Journal of Banking Regulation

, Volume 18, Issue 2, pp 103–131 | Cite as

A jurisdictional comparison of the twin peaks model of financial regulation

Original Article
  • 182 Downloads

Abstract

The last two decades have seen an increasing number of jurisdictions adopting the ‘twin peaks’ model of financial regulation. Since it was pioneered in Australia, the model has been adopted by the Netherlands, Belgium, New Zealand and the United Kingdom. South Africa is currently in the process of changing its model, and it has also been considered by the US. This paper evaluates the twin peaks model as it has been adopted in these jurisdictions and investigates the extent to which there is uniformity in its design and implementation from a regulatory perspective. The comparative analysis indicates that there are many variations in the design of a twin peaks model and that there is no archetypal twin peaks model. In particular, choices need to be made around key elements in areas such as structural design, operational independence and regulatory coordination.

Keywords

financial regulation twin peaks jurisdictional comparison 

Introduction

Since 1998, almost 80 per cent of OECD countries have changed their regulatory architecture.1 The growing complexity of financial products, the increasing challenge of regulating large financial conglomerates, and the Global Financial Crisis (GFC) have made effective regulation a key priority for developed economies. One of the identified trends in recent years has been the departure from the sectoral or institutional model of regulation,2 which sees entities regulated according to the sector in which they operate or their legal form. In moving away from this and other models, jurisdictions confront the challenge of choosing a new regulatory model.

Over the last two decades, one model has gained significant traction since its inception in 1998. The ‘twin peaks’ model of financial regulation was pioneered in Australia and sees regulation split into two broad functions: market conduct regulation and prudential regulation. The model has since been adopted by the Netherlands, Belgium, New Zealand (NZ) and the United Kingdom (UK). South Africa is currently in the process of changing to this model,3 and it has also been considered by the US.4

According to the International Monetary Fund (IMF), part of Australia’s relative success throughout the GFC was its ‘well-developed regulatory and supervisory structure.’5 Despite the model being severely tested in Australia with the HIH Insurance collapse in 2001 and various other financial crises and scandals,6 and also in the Netherlands with the onset of the GFC, the model appears to have maintained its resilience and neither jurisdiction has seriously considered changing to another model.

This paper seeks to evaluate the twin peaks model as it has been adopted in jurisdictions around the world. It will focus on the experience in Australia, NZ, Belgium, the United Kingdom and the Netherlands. In doing so, it will investigate the extent to which there is uniformity in the design and implementation of the twin peaks model and the various design choices that must be made.

Part 2 of this paper provides an overview of the dominant models of financial regulation and the perceived advantages and disadvantages of the twin peaks model. Part 3 summarises each jurisdiction’s journey to the twin peaks model and considers some of the key factors that led to its adoption. Part 4 provides a detailed jurisdictional comparison, examining the key elements of the twin peaks model and how it has been designed in each jurisdiction. These key elements include structural design, operational independence, and the regulatory coordination that is necessary in order to make the model viable.

Part 5 draws inferences from the jurisdictional comparison, and considers how the international twin peaks experience explains trends in financial regulation more broadly. In addition, it considers the ongoing importance of regulatory ‘culture’ and the desirability of achieving ‘synergies’. Part 6 concludes.

The Twin Peaks Model of Financial Regulation

The dominant models of financial regulation 7

The dominant models of financial regulation can be summarised as falling into three broad categories; the ‘institutional model’, the ‘unified’ or ‘super-regulator’ model, and the functional model as reflected in ‘twin peaks’.7 The experience of the jurisdictions surveyed reveals that some countries have frequently changed models, particularly in response to financial collapse.8

The ‘institutional approach’ focuses on the form of the regulated institution (e.g. a bank, insurer or a securities firm) and establishes a separate specialist regulator for that institution. Under this approach, the relevant regulator supervises all activities undertaken by the institution that fall within the scope of financial regulation, irrespective of the market or sector in which the activities take place, and the institution is normally regulated by one regulator alone. The institutional approach is often referred to as an offshoot of the broader sectoral or ‘operational’ approach, under which institutions are regulated by reference to the sector in which they operate or the products or business in which they engage. For example, where a financial institution offers banking products and life insurance, it might be regulated by both the banking regulator and the insurance regulator.9

As early as 1996, Taylor argued that ‘a regulatory system which presupposes a clear separation between banking, securities and insurance is no longer the best way to regulate a financial system in which these distinctions are increasingly irrelevant.’10 The sectoral or operational model, like the institutional model, becomes increasingly difficult to operate as the complexity of financial products increases as well as the complexity of financial institutions, as reflected in the emergence of financial conglomerates. This potentially causes coordination problems and regulatory overlap between the relevant regulators.

The ‘unified’ or ‘super-regulator’ model attempts to address the problems experienced by the institutional and sectoral approaches by creating a single regulator to monitor both the conduct of market participants and also the prudential soundness of financial institutions. This model was championed by the UK prior to its move to twin peaks. One of the perceived problems with this model, however, is that ‘[p]rudential and conduct of business… regulation require[s] fundamentally different approaches and cultures and there may be doubt about whether a single regulator would, in practice, be able to effectively encompass these to the necessary degree.’7 Another issue with the unified approach is that a single regulator ‘might not have a clear focus on the different objectives and rationales of regulation and supervision, and might not make the necessary differentiations between different types of institution and business.’7

The final approach, and the focus of this paper, is the ‘twin peaks’ model. Twin peaks model regulates the market in accordance with two broad regulatory functions: first, market conduct integrity and consumer protection; secondly, prudential regulation and financial system stability. Each objective is pursued by a separate regulator, thus lending the name ‘twin peaks’ to the model.

The perceived advantages and disadvantages of the twin peaks model

The twin peaks model is considered to have a number of advantages. First, the two peak regulators are more likely to have ‘dedicated objectives and clear mandates to which they are exclusively committed.’7 Secondly, there is minimal danger that one aspect of regulation – such as prudential regulation – will come to dominate the regulatory landscape.7 Regulatory culture, which encompasses the attitudes, policies and practices adopted by regulators in fulfilling their objectives, can be fostered depending on the function of the regulator and the culture that it needs to perform its role effectively.7 This avoids the issue of having multiple ‘cultures’ under the one roof, as might be the case with a super-regulator where different cultures arise as a result of the different regulatory objectives.1 Thirdly, the model may be better adapted towards keeping pace with the growing complexity of financial markets and the continuing rise of financial conglomerates.11 Finally, the twin peaks model may avoid an inherent conflict of interest in having a super-regulator. As was noted by the UK Joint Committee on the draft Financial Services Bill (JCFSB) in 2010

[T]he evidence of the recent financial crisis suggests that mixing functions can contribute to a lack of focus on rising macro-prudential risk and difficulties in moving to a ‘war footing’ when that risk becomes substantial. In addition, the incentives are different. For example consumer protection can be well served by keeping a bank open, while stability is well served by closing it.12

There are also a number of perceived disadvantages of the twin peaks model. First, twin peaks may create regulatory overlap with dual-regulated entities. The twin peaks model means that it is ‘inevitable that two separate regulators would have two separate rule books and two separate systems.’12 In the UK, it was noted in 2013 that ‘approximately 2,000 firms [would] be subject to dual regulation.’13 If not carefully managed, this could place a ‘considerable burden’12 on regulated entities and lead to poor information-sharing and coordination.14

Secondly, there is a general risk that cooperation and coordination between the regulators will not be sufficient 15 with potentially serious consequences.16 While these risks can be managed through robust coordination and liaison channels, it nevertheless remains a key concern for jurisdictions that have adopted the model.17

Finally, and depending on how the twin peaks model is implemented, there may be a conflict of interest within the central bank. This can arise if the central is bank is responsible for both prudential regulation and monetary policy, as is the case in NZ. Llewellyn has suggested that ‘a central bank with responsibility for preventing systemic risk is more likely to loosen monetary policy on occasions of difficulty.’7 At the same time, this aspect may also be viewed as creating a synergy ‘arising from [the central bank’s] knowledge of monetary policy and financial market developments.’1

Overview of Twin Peaks Jurisdictions

Australia

In Australia, the twin peaks model arose out of the Wallis Inquiry, which was established in 1996 to review Australia’s financial system. Amongst a number of important recommendations to emerge from the report, one recommendation was that a single agency should be established for the regulation of companies, market conduct and consumer protection.18 The regulation of conduct and disclosure was previously ‘provided through a variety of agencies, with arrangements governed by the institutional form of the service provider.’ This was said to be ‘inconsistent with the emerging structure of markets’ and ‘resulted in inefficiencies, inconsistencies and regulatory gaps and [was] not conducive to effective competition in financial markets.’19 Innovation in product design had ‘blurred the boundaries between financial instruments and institutions’,19 which meant that regulating entities in accordance with their institutional form was becoming increasingly haphazard.

The APRA Bill 1998 was introduced in a suite of financial sector legislative reforms that brought the twin peaks model into existence. The position in Australia can be summarised as follows:

[T]he Australian Securities and Investment Commission (ASIC)… is responsible for the regulation of companies, market conduct and consumer protection; and the Australian Prudential Regulation Authority (APRA)… is responsible for prudential regulation. Under this model, the central bank, the Reserve Bank of Australia (RBA), remains responsible for monetary policy and financial stability, including ensuring a safe and reliable payments system.6

The model was reformed following the collapse of the general insurer, HIH Insurance Limited, in 2001. Following the recommendations made by the HIH Royal Commission, several major changes were made to the legislation governing ASIC and APRA.6

One of the issues identified with the Australian twin peaks model was the ‘apparent confusion in the context of the HIH collapse about the allocation of the functions between APRA and ASIC and how they should coordinate their roles.’6 Other reasons were the less than desirable arrangements in place to facilitate information-sharing between the regulators, and how APRA responded to the information it had been provided.16 Despite the significant difficulties experienced, the HIH Commissioner provided a qualified endorsement of the twin peaks model, and did not recommend the creation of a ‘super-regulator.’16

The Netherlands

In 2002, at ‘around the same time that regulatory reform was being contemplated in France and Germany’,20 the Netherlands started to explore reform to its financial sector. The impetus for reform, according to Pan, was

recognition that the Dutch financial system was ill-prepared to handle the growth of the cross-border markets, the rise of financial firms involved in banking, insurance, and securities activities, the growth of a single market for financial services in the European Union, and the ability of current agencies to meet the new demands proposed by the EU Financial Services Action Plan.21

A number of justifications were put forward in favour of the twin peaks model. As noted by Pan, these included the ability to exercise more effective supervision of financial conglomerates, the reduction in regulatory arbitrage and the need to clarify the role of the Dutch National Bank (DNB).20 One reason why a ‘super-regulator’ model was not favoured was that ‘[t]he objective of conduct of business supervision is considered to be inherently different from prudential supervision and requires a separate supervisor with its own mandate and focus.’1

Since 1 March 2002, the market conduct regulator has been the ‘Netherlands Authority for the Financial Markets’ (AFM). The AFM replaced the ‘Securities Board of the Netherlands’22 with the result that ‘savings, investment insurance and loans’ were all brought under the supervision of one regulator.22 The Dutch Central Bank, De Nederlandsche Bank (DNB), has been responsible for prudential supervision since 2004.20 It replaced the Pensioen en Verzekeringskamer. The Act on Financial Supervision of 28 September 2006 underpins the operation of the Dutch twin peaks model.23

The IMF has commented that the ‘full transition to “twin peaks” was completed only in 2007.’23 This saw the Netherlands ‘[join] Australia as the other prominent example of a country that has adopted the twin peaks regulatory model.’20 Interestingly, the Dutch do not appear to have drawn extensively on the Australian experience in designing their twin peak model.

New Zealand

Like many developed nations, NZ’s financial regulatory reform was largely reactive in nature. Its financial system ‘faced three extraordinary and largely unrelated challenges: an unprecedented GFC; numerous finance company failures; and the Canterbury earthquakes.’24 While NZ had commenced some prudential reform prior to these three events – it designated the Reserve Bank of New Zealand (RBNZ) as prudential regulator in 2005 25 – the bulk of prudential reform occurred after NZ encountered difficulties.26

The GFC and major collapses exposed significant problems in NZ’s system of financial regulation. In April 2010, the NZ Cabinet identified the ‘fragmentation of the market regulators [and] the adequacy of regulators’ powers’ as key concerns. 27 Summarising the key issues with the previous model, the House of Representatives stated there were ‘too many regulators, [a] narrow mandate and limited enforcement powers, under-resourcing and limited… avenues of redress.’27

Under the new framework, NZ’s recently created market conduct regulator is the Financial Markets Authority (FMA). The FMA was established on 1 May 2011 and ‘consolidated functions previously fragmented among the Securities Commission, parts of the Ministry of Economic Development and NZX Limited.’27 The FMA is an Independent Crown Entity.28 The RBNZ is the peak prudential regulator, with supervision carried out by the RBNZ’s Prudential Supervision Department (PSD).

The Australian twin peaks experience was raised in the NZ debates concerning financial sector reform. The former Chief Executive of the FMA made the following remarks to an Australian Parliamentary Committee:

[I]t was personally encouraging that New Zealand decided to implement a twin peaks model similar to the Australian ASIC/APRA design… This spelt the end of the previous fragmented regime with a multiplicity of regulatory bodies.29

United Kingdom

The GFC had ‘an enormous economic, social and political impact on the United Kingdom.’12 In 2008, ‘Britain’s banking system came perilously close to collapse.’12 This was narrowly avoided and required ‘the Government… to step in with multi-billion pound bailouts.’12 The UK JCFSB noted that the crisis ‘highlighted weaknesses with the tri-partite regulatory structure.’12 Use of the term ‘tri-partite’ refers to the three institutions previously involved in financial regulation – the Bank of England (BoE), Treasury and the FSA. The FSA was a ‘super-regulator’, having responsible for both market conduct and prudential supervision.

The FSA was ‘severely criticised for inadequate regulation and supervision of UK banks and wholesale capital markets, and for failing to contain systemic risk.’12 Two key deficiencies with UK financial regulation during the crisis were a lack of coordination between the regulators and confusion surrounding the regulators’ responsibilities:

The lack of proper prudential supervision was only part of the problem. The run on Northern Rock in 2007 revealed major weaknesses in the process for crisis management and in particular in co-ordination and division of responsibility between the three players in the tri-partite system: the Bank, the FSA and the Treasury. As the crisis unfolded, there was disagreement about how to respond.12

The BoE, in its evidence to the JCFSB, noted that there was an inherent cultural conflict between market conduct and prudential regulation, and adopting the super-regulator model exposed these shortcomings:

The conduct supervisor can measure success partly by enforcement, convictions and fines: the prudential supervisor seeks to ensure the continued prudent management of firms in the interests of the system as a whole. Putting these two cultures together was a mistake, and in our view directly contributed to the FSA’s taking its eye off the build-up of prudential risks in a number of major institutions.30

The Deputy CEO designate of the new prudential regulator explained that the introduction of twin peaks would entail ‘a much clearer focus from splitting prudential and conduct regulation for banks and insurers.’31

The FSA had been established under the Financial Services and Markets Act 2000 (FSM Act). The FSM Act was significantly amended by the Financial Services Act 2012 (FS Act), which created a new regulatory framework. The changes became effective in April 2013.32 The JCFSB noted that the ‘fundamental purpose of the draft Bill [was] to change the structure of regulation from the tripartite system to a twin peaks model.’12 The FSA was replaced by two regulators – the Financial Conduct Authority (FCA) and the Prudential Regulation Authority (PRA). The PRA was established as a subsidiary of the BoE. As outlined in Part 4 below, legislative reform in the UK has now ended the PRA’s status as a subsidiary and has brought it within the BoE acting through the Prudential Regulation Committee.

Belgium

Belgium was heavily affected by European bank collapses that struck around the time of the GFC. Two of the country’s largest banks faced significant financial difficulties – Fortis and Dexia, the ‘Franco-Belgian mega-bank that collapsed twice and was bailed out twice within 3 years.’33 The Belgian Government was ultimately forced to nationalise, guarantee and bailout the banking sector.34 The Belgian Government was singled out for these failures, and ‘became the first national administration to fall as a direct result of events linked to the global financial crisis.’35

It is interesting to note that the Belgian model of financial regulation transitioned from the super-regulator model to a twin peaks model. The ‘Banking, Financial and Insurance Commission’ was replaced on 1 April 2011 by ‘Twin Peaks I’ – a suite of measures contained in the ‘Law of 2 July 2010 (henceforth the “Twin Peaks Decree”).

The current market conduct regulator is the Financial Services and Markets Authority (FSMA). Prudential regulation is undertaken by the National Bank of Belgium (NBB), which was established under the Law of 22 February 1998 Establishing the Organic Statute of the NBB.36

The initial twin peaks legislation was significantly amended in 2013 by the ‘Twin Peaks II’ package of legislation – the Law of 30 July 2013 and Law of 31 July 2013. According to one analysis

Twin Peaks II substantially strengthens the supervisory powers of the FSMA and aims to improve market transparency and protection of investors. It builds upon the “Twin Peaks” supervisory architecture created in April 2011, under which the supervision of the financial sector and financial markets is shared between the FSMA and the National Bank of Belgium.37

Some of the main changes included broadening the FSMA’s investigative powers and reinforcing the sanctioning regime.37 The amendments were consolidated in the Law of 2 August 2002 on the Supervision of the Financial Sector and on Financial Services (Law of 2 August 2002).38

Interestingly, prior to having a super-regulator, Belgium operated a sectoral model of regulation. As a result, Belgium has operated three types of financial regulation model: the sectoral approach, the super-regulator approach and, most recently, the twin peaks approach.

Unlike the UK and NZ, which considered the Australian experience in designing their own twin peaks model, Belgium does not appear to have relied on the Australian experience. The most obvious explanation for this is that it looked to the Dutch model. Both Belgium and the Netherlands sit within the broader framework of the EU, which provides an easier reference point than Australia in the design of a twin peaks model.39

Designing Twin Peaks – A Jurisdictional Comparison

The comparative analysis set out below examines five questions that are fundamental to the design of the twin peaks model: (a) Where should the prudential regulator be housed? (b) How should the functions and objectives of the regulators be defined? (c) How should the relationship between each regulator and the government be structured? 40 (d) How should coordination between the regulators be achieved? (e) Should a coordinating body be established between the regulators?

Where should the prudential regulator be housed?

One of the key considerations in implementing a twin peaks model is deciding where the prudential regulator should be housed. The jurisdictions considered in this paper reveal three options:
  1. 1.

    The prudential regulator is an entirely distinct entity that sits outside of the central bank.

     
  2. 2.

    The prudential regulator is a subsidiary of the central bank.

     
  3. 3.

    The prudential regulator exists within the central bank.

     

Prudential regulation by an entity independent of the central bank

Turning to the first option – where the prudential regulator is a distinct entity from the central bank – it is relevant to note that Australia stands alone in adopting this approach. The Australian Prudential Regulation Authority is established under the APRA Act 41 as a body corporate,42 and operates outside of the RBA.

The Wallis Inquiry recommended that the prudential regulator ‘should be separate from, but cooperate closely with, the Reserve Bank of Australia’ and that ‘strong mechanisms should be established to ensure appropriate coordination and cooperation between the two agencies.’43 The following reasons were given for keeping the prudential regulator separate from the central bank:
  • The combination of deposit taking, insurance and superannuation regulation is unlikely to be carried out efficiently and flexibly by a central bank whose primary operational relationships are with banks alone and whose operational skills and culture have long been focused on banking.

  • Separation will clarify that, while the central bank may still provide support to maintain financial stability, there is no implied or automatic guarantee of any financial institution or its promises in the event of insolvency.

  • Separation will enable both the RBA and the [prudential regulator] to focus clearly on their primary objectives and will clarify the lines of accountability for the regulatory task.19

Dr Edey, the Assistant Governor (Economic) of the Reserve Bank of Australia, appeared before the English JCFSB to give evidence on the FSB. When asked whether Australia had considered combining prudential regulation with the RBA, he noted as follows:

When we had bank supervision inside the central bank we found that for most of the time outside a period of crisis nothing was happening in the regulatory sphere, so central banks tend to focus most of their energies at the very senior level on the monetary policy function. I think there can be a case for separating out and specialising the institution that does regulation so it is more focused just on that task, but if you are to do that you need good co-ordination mechanisms between the two.44

Prudential regulation by a subsidiary of the central bank

Prior to recent legislative reforms, the UK operated a unique model where the PRA was a subsidiary of the Bank of England (BoE).45 In considering whether to structure the PRA as a subsidiary of the BoE, the Treasury noted:

[A] case can be made for a close relationship between monetary and prudential policy-makers because of the impact of a central bank’s policy interest rate on financial conditions and the impact of regulatory tools on the monetary transmission mechanism. That case is stronger in a setting where regulatory tools are adjusted more frequently as a means of mitigating macro-prudential risk, as the reform proposals currently suggest they will…46

Two potential justifications emerged in support of having a prudential regulator that operated as a subsidiary of the central bank, as distinct from operating as an entity independent of the central bank. First, because the functions of prudential policy and monetary policy are inherently linked, it was considered that a close relationship between the BoE and PRA would be desirable to achieve the best regulatory outcome. Secondly, there would be a lower risk of ‘regulatory gaps’ because the BoE, through the Financial Policy Committee statutory coordinating committee, could compel the PRA to take regulatory action.46

The Bank of England and Financial Services Act 2016, which was enacted on 4 May 2016, will end the PRA’s status as a subsidiary and constitute the BoE as the PRA, the functions of which will be exercised by the BoE acting through a new committee of the BoE to be known as the Prudential Regulation Committee. The purpose of this reform is to ‘maximise the synergies of having monetary policy, macro-prudential policy and micro-prudential policy under the aegis of one institution.’47

Prudential regulation by the central bank

In addition to the UK following the recent reforms, the Netherlands, Belgium and New Zealand operate a third option.

Turning first to the Netherlands, the DNB identified in its ‘Supervisory Strategy’ the key benefits of housing the prudential regulator within the central bank:

[T]he credit crisis has shown the added value of a combined central bank–prudential supervisor: information and expertise on supervision, payments, financial markets, macroeconomics and financial stability were all present under one roof. But even in normal circumstances it is useful to harness the full power of this combination.1

As is the case in the Netherlands, the justification of achieving ‘synergies’ by having the central bank conduct prudential supervision was recognised in Belgium. According to the NBB

One of the reasons why the lawmaker wanted the central bank to combine all the activities concerning financial stability and prudential supervision is that the Bank possesses skills and data which are particularly useful for the performance of those tasks.48

In New Zealand, prudential supervision of ‘specified institutions’ has been the responsibility of the Reserve Bank since 1986.49 The role of the RBNZ with respect to prudential regulation was further clarified in 2005, when it was ‘agreed… that the Reserve Bank would be the prudential regulator of financial market participants.’50 Thus, while the RBNZ had prudential supervision powers as early as 1986, its status as the prudential regulator was more formally recognised from 2005.

A specialist ‘Prudential Supervision Department’ (PSD) was formed on 1 November 2007 with responsibility ‘for the microprudential regulation of banks, payment and settlement systems (PASS), insurance companies and Non-Bank Deposit Takers (NBDTs).’24

How should the functions and objectives of the regulators be defined?

Adequately demarcating the responsibilities of the market conduct regulator and prudential regulator is critical to the success of a twin peaks model. The necessity of regulators having clear responsibilities and objectives is reflected in the Basel ‘Core Principles for Effective Banking Supervision’51 (“Basel Core Principles”) and the ‘Insurance Core Principles’.52

The next subsection considers the functions and objectives of the regulators in each jurisdiction.

The functions and objectives of the market conduct regulator

While being broadly consistent, each jurisdiction varies in how it determines the responsibilities of the market conduct regulator. The analysis below proceeds by identifying some of the nuances in how the regulators’ functions and objectives are expressed:
  1. 1.

    Is there a ‘main objective’ or ‘strategic objective’?

     
  2. 2.

    Is the market conduct regulator required to have regard to competition?

     

i) Is there a ‘main objective’ or ‘strategic objective’?

There are variations in whether the regulators’ objectives are expressed as being a ‘main objective’ or a ‘strategic objective’. Use of the word ‘main’ suggests that in the event there is a conflict between objectives, the main objective will prevail. The use of the word ‘strategic’ suggests that the regulator must generally act in a way that is compatible with the strategic objective and that the operational objectives must be informed by the strategic objective.53

The NZ FMA Act provides for the ‘main objective’ of ‘promoting and facilitat[ing] the development of fair, efficient, and transparent financial markets.’54 The detailed functions of the FMA are then listed.55 The ASIC Act, by contrast, simply states that the ‘objects of this act are to provide for ASIC’s functions, powers and business.’56 The Act then goes on to state a number of outcomes that ASIC must strive towards in carrying out its functions.

In line with its objectives-based approach to regulation, the UK’s FCA provides for two types of high level objective: a ‘strategic’ objective of ensuring that ‘the relevant markets function well’,57 and the ‘operational’ objective of consumer protection, integrity and competition.58 In addition to these objectives, the UK recognises a diverse range of ‘regulatory principles to be applied by both regulators.’59 These relate to notions of efficiency, placing the onus of responsibility on consumers, and transparency.60

The Dutch AFM similarly lists three strategic objectives in its ‘mission statement’; ‘promot[ing] the fair and conscientious provision of financial services; promot[ing] the fair and efficient operation of the capital markets [and] contribut[ing] to the stability of the financial system.’61 The objectives and functions of the AFM are then listed in the legislation.62

The Belgian FSMA has objectives and responsibilities that are broadly aligned with the other jurisdictions.63 Article 45 of the Law of 2 August 2002 spells out the powers of the FSMA in considerable detail.

ii) Is the market conduct regulator required to have regard to competition?

Another difference vis-à-vis the market conduct regulators’ powers is whether or not the regulator must have regard to competition in carrying out its functions.

In the UK, the FCA has a ‘competition objective’ as one of its operational objectives.64 Interestingly, this objective is capable of being amended by Treasury.65 This can be contrasted with ASIC which, unlike APRA, is not currently required to have regard to competition. It has been noted that the effect of including a formal competition objective would be that ASIC would ‘select the most “competition-friendly” option from a range of potential regulatory responses, provided that this option was also capable of achieving ASIC’s other regulatory objectives.’66 This was accepted by the recent Financial System Inquiry (FSI) in its Final Report, which recommended that ASIC ‘be given an explicit competition mandate’ and that ‘periodic reviews of the state of competition should be conducted.’67

There is no explicit competition objective that applies to the regulators in NZ, Belgium or the Netherlands.

The functions and objectives of the prudential regulator

The variations in the functions and objectives of the prudential regulators are greater than the market conduct regulators. The difference may be analysed by asking the following questions:
  1. 1.

    How are the objectives of prudential regulation expressed?

     
  2. 2.

    Is the prudential regulator required to have regard to competition?

     
  3. 3.

    To what extent can the prudential regulator influence the market conduct regulator?

     

i) How are the objectives of prudential regulation expressed?

In NZ, the objectives of the RBNZ relating specifically to prudential regulation are found across a patchwork of legislation. The general provisions relating to prudential regulation are found under the RBNZ Act, which states that ‘the Bank shall… undertake prudential supervision of registered banks 68 for the purposes of ‘promoting the maintenance of a sound and efficient financial system’ and ‘avoiding significant damage to the financial system that could result from the failure of a registered bank.’69

Following the RBNZ’s expanded mandate to regulate the insurance sector in 2010, the Insurance Prudential Supervision Act 2010 states that one of the purposes of the Act is to ‘promote the maintenance of a sound and efficient insurance sector.’70 The reach of the RBNZ’s prudential supervision purpose was extended yet again in 2013 by the Non-bank Deposit Takers Act 2013.71 The RBNZ now has the function of ‘act[ing] as the prudential regulator and licensing authority of NBDTs’,72 and has wide powers to give directions to NBDTs and associated persons.73

In Australia, the APRA Act provides that APRA exists for the purpose of ‘regulating bodies in the financial sector in accordance with other laws of the Commonwealth that provide for prudential regulation or for retirement income standards’.74

Under the Dutch legislation, the Dutch regulator, the DNB, is ‘required to exercise the prudential supervision of financial enterprises and to decide on the admission of financial enterprises to the financial markets.’75 This is supplemented by detailed provisions concerning the DNB’s powers.

In the UK, the BoE, as the PRA, has the general objective of ‘promoting the safety and soundness of PRA-authorised persons’,76 which is achieved primarily by ensuring that the business of PRA-authorised persons is ‘carried on in a way which avoids any adverse effect on the stability of the UK financial system, and seeking to minimise the effect of a failure of a PRA-authorised person.’77 The PRA has more specific objectives with respect to insurance.78

It is interesting to note that the UK Treasurer has the power to provide for new objectives with respect to the PRA.79 Where the Treasurer decides to exercise this power, it is subject to a Parliamentary approval procedure.80

With respect to Belgium, a key objective of the NBB relates to assisting the European System of Central Banks in maintaining price stability 81 and various other European-oriented objectives.82 The objectives of the NBB include ‘[contributing] to the stability of the financial system’ and ‘carrying out its supervisory tasks exclusively in the general interest…’83

ii) Is the prudential regulator required to have regard to competition?

Differences arise between jurisdictions as to whether the regulator must have regard to competition.

APRA, unlike ASIC, is required to have regard to ‘competition’ and ‘competitive neutrality.’ Subsection 8(2) of the APRA Act provides that in ‘performing and exercising its functions and powers, APRA is to balance the objectives of financial safety and efficiency, competition, contestability and competitive neutrality and, in balancing these objectives, is to promote financial system stability in Australia.’

As noted above, the BoE, acting as the PRA, has the general objective of ‘promoting the safety and soundness of PRA-authorised persons’,84 which is to be ‘carried on in a way which avoids any adverse effect on the stability of the UK financial system, and [seeks] to minimise the effect of a failure of a PRA-authorised person.’85 It must also have regard to ‘the need to minimise any adverse effect on competition in the relevant markets that may result from the manner in which the PRA discharges those functions.’86 The rationale behind this secondary competition objective was addressed by the JCFSB:

Competition within the financial sector is an important part of developing a stronger, more diverse system. The actions of the PRA have the potential to affect the costs of individual firms or of particular types of institution, and affect the barriers to entry and expansion in the market. While the need to protect and promote competition in the sector should not dictate the actions of the PRA… we believe it is a factor that ought to be considered in the course of PRA decision making.12

There is no express requirement for the prudential regulator to have regard to competition in Belgium, the Netherlands or NZ.

iii) To what extent can the prudential regulator influence the market conduct regulator?

The UK is unique because the BoE, acting as the PRA, has the power to prevent the FCA from acting in limited circumstances. This is effectively a veto power, and will be used where ‘an action to be taken by the FCA… is likely to threaten the stability of the UK financial system…’87 In 2011, the IMF raised concerns about the veto power, and stated that ‘without appropriate safeguards, this arrangement [had] the potential to limit FCA independence and also to cause uncertainty in decision making.’88

Other jurisdictions do not recognise an equivalent regulator veto power.

How should the relationship between each regulator and the government be structured?

The government has a legitimate interest in the performance of financial regulators. This is reflected in the various ways the governments in twin peak jurisdictions can influence, direct or commandeer the functions of the regulators. It is an area that sees the largest degree of divergence between twin peaks countries, and demonstrates why there is no paradigmatic twin peaks model. This is a critical issue as it goes to the question of operational independence.

While some jurisdictions are readily comparable, many are not. The following questions are relevant:
  1. 1.

    Can the government direct the regulator to undertake an investigation?

     
  2. 2.

    Are the regulators under specific duties to advise, or report to, the government?

     
  3. 3.

    Can the government direct the regulators to take action?

     
  4. 4.

    Can the government determine which regulator is the lead regulator in certain circumstances?

     

Can the government direct the regulator to undertake an investigation?

A common power given to the governments of twin peak countries is the ability to direct the regulators to undertake an investigation. The existence of this power varies between the market conduct and prudential regulators.

In NZ, the Minister may request that the FMA inquire into and report on ‘any matter relating to the financial markets, financial markets participants, or other persons engaged in conduct relating to those markets.’89 Similarly, the Australian Minister may direct ASIC to investigate particular matters where it is in the ‘public interest.’90 The UK Treasury may direct both the FCA and the BoE acting as the PRA to undertake an investigation.91

The NZ Minister may request the RBNZ to provide advice ‘on any matter… that is connected with the functions of the Bank.’92 Where such advice is given, the RBNZ Governor is under a further obligation to advise the Minister of the likely effect on monetary policy of that advice.93 In Australia, the Minister may require the prudential regulator to give advice.94 In the Netherlands, the DNB is expected to provide advice to the government, when requested, on matters of economic importance.95

Are the regulators under specific duties to advise, or report to, the government?

In Australia, pursuant to reforms introduced after the collapse of the HIH Group, APRA has an obligation to advise the Minister where ‘it considers that a body regulated by APRA is in financial difficulty.’96 The rationale for imposing this duty was articulated as follows:

The Government [has] a legitimate interest in APRA’s operations, its policies and the way they are applied. The possibility of a prudentially regulated institution experiencing financial difficulty is properly a matter of concern to the Government, particularly when the institution is a deposit-taking institution involved in the payments system. It is appropriate, therefore, that the Treasurer, as the Minister with overall responsibility for the financial system, should be informed when an institution is in financial difficulty.97

ASIC is not subject to an equivalent duty. Interestingly, nor is the RBNZ or FMA in NZ. This can be compared with the UK’s PRA and FCA, both of which are under a duty to investigate and report on possible regulatory failures.98

Can the government direct the regulator to take action?

There are substantial differences between jurisdictions in terms of the extent to which the government can direct the regulator to undertake a certain course of action. This can vary from having no powers of direction, to the government stepping in and carrying out a function on behalf of the regulator.

i) Market conduct regulator

Some jurisdictions do not allow the executive government to direct the market conduct regulator in any way. NZ does not provide for any direction over the FMA yet, interestingly, does for the RBNZ with respect to ‘government policy objectives’.99 The lack of direction power may stem from the fact that the FMA, as an ‘independent Crown entity’,27 is afforded a greater level of independence than other crown entities recognised in NZ. The Australian government can require ASIC and APRA to have regard to government policy, but is not permitted to give a direction about a particular case.100

Previously, the Dutch MoF had the power to ‘carry out directly’ the duties of the supervisors if either the AFM or DNB ‘seriously failed to perform its duties.’101 This was identified as a significant area of concern by the IMF, despite this power never having been exercised.101 The relevant provision was repealed on 1 January 2013, effectively abolishing the Minister’s intervention powers in relation to the regulators.102 This addressed one of the key concerns in the IMF assessments, and also brought the Netherlands in line with Australia.

ii) Prudential regulator

The NZ government can influence the RBNZ by requiring it to have regard to ‘government policy objectives.’ The ‘Minister may direct the Bank to have regard to a government policy’ where a direction is set out in writing after consultation with the bank.103 Importantly, this is qualified by the fact that ‘the Minister may not give a direction that requires the performance or non-performance of a particular act by the Bank, or any employee or office holder of the Bank, or the bringing about of a particular result, in respect of a particular person.’104

The NZ Cabinet noted that the power was designed to ‘enable the Minister to engage with the Reserve Bank on the outcomes it is seeking to achieve, but… maintain the Reserve Bank’s independence in its regulatory functions.’105

Concerns were raised in Parliamentary debate about the introduction of a power to direct the RBNZ. Two MPs voiced their concerns that ‘the last thing we need is what is meant to be an independent regulatory environment becoming one that the politicians can start to meddle with’106 and that this could result in the ‘possible politicisation of monetary policy.’107 The justification put forward by the NZ Cabinet was that

Government should have the opportunity to be involved in issues that raise policy considerations, to ensure that the value judgements underpinning policy reflect those of society at large, that policy in specific areas supports the government’s wider strategic objectives and that priorities and policy are coherent and consistent across government.105

Cabinet also noted that the direction power could potentially result in ‘poorer quality regulation, if political concerns lead to inconsistency in regulatory approach’ and ‘more opportunity for lobbying by industry.’105

The position in Australia is very similar to NZ – the Australian Government may give APRA written directions regarding the ‘policies it should pursue’, but ‘must not give a direction… about a particular case.’108

Can the government determine which regulator is the lead regulator?

A key challenge for twin peak jurisdictions is how they manage ‘dual-regulated entities.’ These entities are regulated by both the market conduct regulator and prudential regulator. While effective coordination between the regulators is central to ensuring that they are managed appropriately, the Government may have the power to determine who the ‘lead’ regulator is.

The UK Treasury has the power to establish the ‘boundaries’ between the two regulators. This effectively allows the Treasury to determine who should be the lead regulator in a regulatory investigation:

3G Power to establish boundary between FCA and PRA responsibilities

(1) The Treasury may by order specify matters that, in relation to the exercise by either regulator of its functions relating to PRA-authorised persons, are to be, or are to be primarily, the responsibility of one regulator rather than the other.

(2) The order may

(a) provide that one regulator is or is not to have regard to specified matters when exercising specified functions; and

(b) require one regulator to consult the other.109

Where this power is exercised, a draft order must be laid before Parliament unless it is a matter of urgency.110

The ability to determine who is the ‘lead supervisor’ or ‘primary regulator’ has been said to ‘[create] problems in terms of coordinating supervisory action, information flows and, ultimately, crisis intervention’7 and was not recommended in the Australian FSI Final Report. When read alongside the PRA’s veto power in relation to the FCA, it is clear that the UK has deliberately built features into its twin peaks model to ensure that only one regulator will act where necessary. This provides an interesting point of difference with Australia, where decisions are largely consensual between the regulators, and soft power is relied on extensively.6 While both Australian regulators are subject to a general Ministerial direction’s power, this falls short of the Minister being able to determine which regulator is to take primary responsibility for a matter.111

It remains to be seen to what extent the ‘power to establish boundaries’ in the UK undermines the independence of the regulators, and whether it evinces a lack of faith in the ability of the regulators to coordinate matters effectively between themselves.

How should coordination between the regulators be achieved?

Coordination between the regulators in a twin peaks jurisdiction is critical to the model’s success.112 As has been noted with respect to the UK, the model is ‘unlikely to function effectively if the different parts… operate in a silo manner.’13 The chief difficulty is recognising what is necessary for effective coordination, and how this is best achieved. Some jurisdictions establish a formalised regime for coordination, whereas other such as Australia see ‘coordination among the agencies [being]…a largely informal arrangement.’113

The following discussion highlights the wide divergences in approach in terms of addressing the issue of coordination.

Establishing a duty to cooperate under legislation

In Australia, while the APRA Act makes express reference to coordination between the regulators, the ‘intention’ that the regulators cooperate is not expressed as a legally binding obligation. To this end, its function is more aspirational in nature.114 Interestingly, there is no equivalent duty in the ASIC Act. While both of the NZ regulators have the power to share information, neither is under an express duty to cooperate with the other regulator.

The UK, on the other hand, imposes a statutory duty on the regulators to cooperate in discharging their functions. The previous regulator, the FSA, was subject to a duty to ‘cooperate with others… who have similar functions to those of the Authority.’115 The current duty is more specific, and is expressed as a duty of the FCA and PRA to ensure the coordinated exercise of their functions.116

The FCA and PRA must coordinate by obtaining information from each other in matters of ‘common regulatory interest’.117 This only applies to the extent that it is compatible with the advancement of each regulator’s objectives and the burden of compliance is not unduly disproportionate to the benefit.’118 For the purposes of the FCA, ‘objectives’ means ‘operational objectives.’119

The Memorandum of Understanding between the FCA, BoE and PRA echoes this duty by stating that ‘the FCA, the Bank and the PRA, will share information related to markets and markets infrastructure where materially relevant to another of them both at their own initiative and upon each other’s request, where legally permissible.’120 Interestingly, a separate duty to cooperate exists between the Treasury, the BoE and the PRA to coordinate the discharge of their functions ‘so far as they… relate to the stability of the UK financial system… and affect the public interest’.121

The UK is not the only jurisdiction that recognises a statutory duty to cooperate. In the Netherlands, both the AFM and DNB are under general statutory duties to cooperate. They are also required to cooperate in specific areas.101 One translation of the general duty reads ‘the supervisor shall collaborate closely with a view to laying down generally binding regulations and policy rules, in order to ensure that these are equivalent wherever possible insofar as they relate to matters that are both subject to prudential supervision and supervision of conduct.’122

The Belgian Law of 2 August 2002 states that ‘the FSMA and the Bank may agree terms of cooperation in areas which they shall determine.’123 The NBB has noted that the regulators ‘conclude[d] a protocol in order to lay down the practical arrangements’124 necessary for coordination. The Belgian ‘protocol’ was concluded on 14 March 2013 and has the same legal status as a Memorandum of Understanding.

Interestingly, in Belgium it is only the FSMA that is under a duty to cooperate with other Belgian authorities.125 The IMF picked up on this idiosyncrasy, and suggested ‘consideration should be given to introducing [an equivalent] provision to supplement other cooperation.’14 This would ensure that the FSMA was ‘aware of failures or weaknesses in an institution’s capital, liquidity or governance structure.’14 It further noted that the terms of the current legislation were unclear and should be amended to ‘clarify the FSMA’s obligations to share information with the NBB for the purposes of prudential supervision.’14

The NBB has noted that ‘cooperation’ has three facets: ‘the exchange of information, consultation and coordination.’124 Despite a very detailed Memorandum of Understanding addressing cooperation, the Twin Peaks Decree ‘does not address in great detail how the NBB and the FSMA will cooperate in the future.’126 Cooperation between the NBB and FSMA was seen as a particularly important issue following the twin peaks reform, and has been identified as a High Priority action for Belgium by the IMF.126

Legislative provisions dealing with information-sharing and confidentiality

A central part of coordination is the ability of the regulators to share information that they have obtained about financial market participants.

In Australia, each of APRA and ASIC is subject to confidentiality or secrecy requirements. However, provision is made for each body to share information with the other.127 In NZ, the FMA has express power to share information with other regulators where it considers that this will assist the other regulator in carrying out its function. Appropriate safeguards must be in place to protect confidential information shared with the other entity. Similarly, the FMA may use any information that it receives from other entities in carrying out its own functions.128 The confidentiality of information exchanged between the regulators must be preserved by both the FMA 129 and RBNZ.130

In the UK, there are a number of provisions in the FSM Act dealing with the disclosure of confidential information.131 Broadly speaking, the FCA cannot disclose confidential information unless it does so under one of the limited exceptions.132 The ‘Memorandum of Understanding between the Financial Conduct Authority and the Bank of England, including the Prudential Regulation Authority’ provides that the ‘the FCA and the Bank (including the PRA) will protect the confidentiality and sensitivity of all unpublished regulatory and other confidential information received from the other.’133

The Dutch DNB has a ‘duty of confidentiality in respect of all information obtained from supervised institutions.’134 The AFM is permitted to provide confidential information to the DNB.135 The Dutch Financial Stability Committee has also been a positive development for information exchange between the regulators.

With respect to Belgium, in 2013 the IMF recommended as a High Priority action that ‘the authorities should remove legal impediments to information,’ and stated that ‘there is scope for the FSMA and the NBB to review how best to strengthen supervisory cooperation as they gain experience in implementing the twin-peaks structure.136

There are various duties of confidentiality imposed on both the FSMA and NBB. The NBB staff ‘shall be subject to professional secrecy and may not divulge to any person or authority whatsoever confidential information of which they have had knowledge on account of their duties.’137 This applies in largely the same terms to the FSMA.138 If the NBB receives any confidential information from another regulator, it ‘must… respect any restrictions that may be set out by the foreign authority.’136 The IMF has identified specific circumstances where information is exchanged; for example, in conducting ‘fit and proper assessments’ of directors, and prior to undertaking enforcement action against NBB supervised entities.136

Memoranda of understanding

A widely used mechanism to help facilitate coordination between the regulators is the preparation of Memoranda of Understanding (MoU). MoUs typically detail the arrangements between the regulators regarding areas such as the exchange of information, coordination, sharing of fees and dispute resolution. Despite the fact that a MoU is not legally enforceable, it nevertheless takes on different levels of importance in different jurisdictions.

In NZ, an MoU between the FMA and RBNZ was entered into on 9 September 2011.139 It outlines various mechanisms for cooperation and coordination, and also makes provision for information sharing between the regulators. The MoU is designed to ‘ensure cooperation, promote transparency, help prevent unnecessary duplication of effort and identify risks developing in the financial system.’139 The provisions concerning information sharing are dealt with at a very general level,140 which is in line with NZ’s ‘soft law’ approach to regulatory cooperation.

Striking similarities emerge when the equivalent Australian MoU 141 is read alongside the NZ MoU. The NZ MoU contains the same sections and substance as the Australian MoU, indicating that the Australian MoU was used as a model.

One difference in the NZ MoU is the inclusion of the statement that ‘[t]he FMA and the RBNZ will hold meetings of senior officials at least every 6 months to discuss the operations of this MoU.’142 This finds no direct counterpart in the Australian MoU.

In the UK, there is an extensive MoU framework between various UK regulators and international organisations. Many of the MoUs have statutory backing and are therefore expected to be taken seriously and implemented in accordance with their terms. For example, the BoE considers that ‘these are documents through which the Bank expects to be held to account by Parliament, through the Treasury Select Committee.’143 The fact the BoE considers that it is held to account by Parliament through the MoU is a significant point, particularly when compared with Australia. There is no expectation that either ASIC or APRA will be held to what is written in the MoU – to the contrary, the Australian MoU appears to be used primarily for signalling purposes to indicate how the regulators intend to achieve effective coordination.6

Another interesting feature of the arrangements in the UK is that the PRA and FCA are under a duty to ‘prepare and maintain a memorandum which describes in general terms… the role of each regulator in relation to the exercise of functions conferred by or under this Act.’144 This MoU must be reviewed at least once a year, given to the Treasury and tabled before Parliament.145 This provision is prescriptive about the types of information that must be contained in the MoU vis-à-vis coordination between the regulators. This potentially provides greater certainty about what is meant by ‘coordination.’

Another MoU is required to be prepared and maintained between the Treasury, BoE and PRA with respect to crisis management.146 This is designed to set out ‘who is in charge of what and when between HM Treasury and the Bank (including the PRA) in a financial crisis.’147 Additionally, the Treasury, BoE, FCA and the PRA must prepare and maintain an MoU with the European Supervisory Authorities, EU institutions and other international organisations.148

In the Netherlands, there is a ‘Covenant’149 in place between the AFM and DNB, which appears to take on the same significance as an MoU in other twin peaks jurisdictions.150 Commentators have suggested that the Covenant was a factor that made navigating the GFC an easier task.151 The Covenant is broadly aligned with the Australian and NZ MoUs. It addresses information exchange 152 and coordination 152 in similar terms to other twin peak jurisdictions. It also makes provision for ‘netting arrangements’, which allow for costs to be shared between the regulators where both are involved in regulating an entity.152

The position in Belgium finds no ready parallels with the other countries. The FSMA and NBB concluded a ‘Protocol’ – equivalent to an MoU – called the ‘Cooperation Protocol Between the NBB and the Financial Services and Markets Authority in the Field of Supervision and Oversight of Market Infrastructures.’153 The Protocol was made in accordance with the Twin Peaks Decree, which states that ‘the FSMA and the Bank may agree terms of cooperation in areas which they shall determine.’154 It is important to note that this is permissive rather than prescriptive in determining how the regulators are to cooperate. It also does not equate to a duty to conclude an MoU as is seen in the UK.

The preamble to the Protocol states that ‘there is a need for co-operation and the sharing of information between the two Authorities in order to ensure the highest level of efficiency in the exercise of their supervision.’155 Article 2 of the Protocol reads ‘the Authorities commit to co-operating to the best of their ability to ensure the exercise of their respective supervisory powers.’ The Protocol contains detailed provisions on information-sharing.156

Notwithstanding the fact that the Protocol uses words such as ‘shall’ in describing the various arrangements, Article 28 makes it clear that the Protocol ‘may not form the basis of any claim for liability or any other dispute before the courts.’157 Another feature of the Protocol is that it establishes a ‘Liaison Committee’ and a ‘Joint Supervisory Committee’ between the regulators to further enhance cooperation.124

Neither the MoU in Australia nor the Protocol in Belgium is legally binding. The key difference between the two jurisdictions is the extent to which the MoU forms the basis for regulatory cooperation. In Belgium, the Protocol is detailed. The implementation of the Protocol, as discussed below, is specifically supervised by the Joint Supervisory Committee. Conversely, the Australian MoU is very brief and appears to serve primarily as an indication of how the regulators intend to achieve effective coordination.6 Thus, the Belgian Protocol is a focal point for coordination between the regulators – even if it is not legally binding. The same cannot be said in the Australian context.

Should a coordinating body be established between the regulators?

A common element of twin peak jurisdictions is the existence of a ‘coordinating body’. This body may be a creature of statute, or it may be an informal forum for representatives of the regulators to meet. The membership of the body tends to vary by jurisdiction. This section examines the following three questions:
  1. 1.

    Is there a coordinating body?

     
  2. 2.

    Is the coordinating body enshrined in statute?

     
  3. 3.

    Is the government a member of the coordinating body?

     

Is there a coordinating body?

NZ has a Council of Financial Regulators (CFR) to assist with regulatory cooperation. The CFR typically meets ‘on a quarterly basis and the chair will rotate between the RBNZ and the FMA.’158 The issues discussed will be confidential ‘unless disclosure is required by law or agreed by the permanent members.’158 The aim of the CFR is to ‘contribute to the efficiency and effectiveness of New Zealand’s prudential and financial markets regulatory model and to promote the stability of the New Zealand financial system by providing a forum to review industry trends and issues.’158 Its functions include sharing information, identifying important issues and trends, and ensuring responses are coordinated.158

Australia has a Council of Financial Regulators (CFR) that operates in a similar way to NZ’s CFR. The Assistant Governor of Australia’s RBA has noted that ‘[it has] never found any conflicting objectives or perspectives on policy’44 at CFR meetings. The CFR ‘has no legal functions or powers separate from those of its individual member agencies.’159 As outlined by the RBA in its submission to the FSI, its membership comprises APRA, ASIC, the RBA and the Treasury; meetings are chaired by the Reserve Bank Governor and are ‘typically held four times per year but can occur more frequently if required’. Further, ‘[m]uch of the input into CFR meetings is undertaken by interagency working groups, which has the additional benefit of promoting productive working relationships and an appreciation of cross-agency issues at the staff level’. Noting that the CFR has worked well since its establishment, the RBA submitted that the ‘experience since its establishment, and especially during the crisis, [had] highlighted the benefits of the existing non-statutory basis of the CFR.’160

In the UK, the nearest thing to a coordinating body is the ‘Financial Policy Committee’ (FPC) which has ‘responsibility for regulation of stability of the financial system as a whole.’161 The FPC is a statutory committee of the Bank of England’s Court of Directors. The FPC has a suite of powers including the power to give directions to the FCA and PRA in relation to macro-prudential measures,162 and the ability to make recommendations to the BoE,163 Treasury,164 PRA and FCA.165 The FPC’s policy decisions ‘[are] announced via the Financial Stability Report or in an official statement to the market shortly after a meeting.’32

The Netherlands has a coordinating body called the Financial Stability Committee (FSC). The Financial Stability Board (FSB) has viewed the creation of the FSC favourably, noting ‘cooperation and information exchange between the relevant institutions responsible for safeguarding financial system stability have been strengthened via the creation of the FSC.’166 The main reason behind establishing the FSC was ‘to create a forum in which the authorities with a responsibility for financial stability could meet to identify and discuss potential risks and ways to mitigate them, including by making recommendations with respect to those risks.’166 The FSC is able to issue warnings, but its recommendations are cautionary and not binding on the regulators.166 Summaries of the FSC meetings are published on its website,167 and it also produces annual reports for the Ministry of Finance.168

Belgium operates a different structure: it has both a coordinating committee, the ‘Liaison Committee’ (LC), and a ‘Joint Supervisory Committee’ (JSC) established by the Protocol.169 The LC is in place to:

[M]onitor the implementation of the MoU, resolve any questions of interpretation that may arise, evaluate whether the MoU needs adapting - especially in order to take into account any potential legislative changes at a European Union level - and endorse the templates to use for letters for the implementation of the MoU.170

The JSC takes on a more conventional role as a coordinating body, and is established ‘with a view to coordinating [the regulators’] respective supervision policies.’170

Interestingly, the emphasis on implementing the Protocol is a distinguishing factor between Belgium and Australia. In Australia, the MoU between ASIC and APRA is brief, and is not fundamental to cooperation between the regulators. In Belgium, it is clear that the Protocol is intended to be followed closely and forms the basis of coordination.

Is the coordinating body enshrined in statute?

The NZ CFR is not established by legislation. Australia’s CFR also exists outside of legislation, but there have been calls to formalise its existence and give it statutory recognition.171

The UK’s FPC is established under statute. Strictly speaking, it is unlike other coordinating bodies because it exists within the BoE.

The Dutch FSC was created ‘by means of Ministerial Decree rather than in primary legislation.’166 The Financial Stability Board was concerned by the relatively informal nature of the FSC and ultimately recommended introducing legislation to recognise its role.

‘[T]he authorities should consider enhancing the FSC’s legal status… [T]o further improve its effectiveness and to enhance its credibility as a key part of the macroprudential framework, the authorities should consider embedding the FSC in primary legislation.’166

The suggestions of the FSB that the FSC should be reformed and given a legislative basis are particularly interesting in the Australian context. Australia’s Council of Financial Regulators (CFR) has no statutory basis and is an informal body.172 Belgium is also interesting because, while its JSC is created under the Protocol, its existence is not enshrined in primary legislation.

Is the government a member of the coordinating body?

In NZ, the permanent members of the NZ CFR are the RBNZ, the FMA, the Ministry of Business, Innovation and Employment and the Treasury.158 Thus, there are two government representatives sitting on the CFR.

Membership of the Australian CFR includes APRA, ASIC, the RBA and Treasury.44 The key difference between the NZ and Australian CFRs is the presence of the Ministry of Business, Innovation and Employment in the NZ CFR.

In the UK, the FPC’s members are the Governor, the four Deputy Governors, the Chief Executive of the FCA, the Bank’s Executive Director for Financial Stability Strategy and Risk, five external members appointed by the Chancellor, and a non-voting representative of the Treasury.173

The Dutch FSC is:

made up of seven representatives from the authorities: three from DNB (including the President as FSC chair) and two each from the MoF and the AFM. Each representative attends in a personal capacity. There are no formal independent external members, though there is scope to invite external experts as required.166

Thus, the Dutch government is represented through the MoF. Additionally, a ‘secretariat at DNB works with staff from DNB, AFM and MoF to prepare the meetings.’166 Interestingly, the MoF representative has no voting rights in FSC meetings ‘to ensure a degree of independence’.166

The Belgian Liaison Committee ‘consists of four members of at least management level, with each Authority designating two members.’174 The JSC consists of ‘the heads of the ‘Prudential Policy and Financial Stability’ and the ‘Policy’ departments of the Bank and the FSMA respectively. Other people may join on a case-by-case basis.’174 It is important to note that the Government does not have any involvement in the coordinating body, as is the case in other twin peak jurisdictions. To some extent, this is addressed by the fact that the Belgian Minister may send a personal representative to attend board meetings of the FSMA.

What Does the Jurisdictional Comparison Reveal About Twin Peaks and Financial Regulation More Broadly?

The preceding analysis has highlighted in detail that there are many variations in designing a twin peaks model. There is no ‘one size fits all’ model and there is no rulebook about how a jurisdiction should approach the task. In looking at the design choices, and considering why they were made, a number of insights can be gleaned about the twin peaks model and financial regulation more broadly.

The relationship between regulatory culture and regulatory structure

At the heart of most of the decisions in implementing the twin peaks model is the desire to create a more effective regulatory culture. When the previous UK and NZ models failed, regulatory culture was cited as a leading reason. The UK suggested that ‘successful regulation depends more on the regulatory culture, focus and philosophy than on structure’.12 This is in line with broad views about regulation:

[F]undamentally, better regulation comes from stronger laws, better-trained staff and better enforcement. Any country that thinks that tinkering with the structure of agencies will, by itself, fix past shortcomings is doomed to relive its past crises.175

It is worth considering the relationship between culture and structure. One view may be that culture can only be sustained by the appropriate structures, and is therefore reliant on having a sound model. Another is that culture exists independently from the structure within which it is embedded.

The most likely answer, it is suggested, lies somewhere in the middle. Fault lines in regulation are created when there is a misalignment between culture and structure. The idea that completely overhauling a jurisdiction’s regulators would somehow enhance the culture of the new regulator is potentially myopic. Countries that initiate regulatory reform following financial disasters run the risk of overlooking the fact that changing structure alone is unlikely to achieve much. The structure should, however, be one that fosters the change needed within the regulators, and actively encourages them to work more effectively.

As noted by Llewellyn:

[I]nstitutional structure may have an impact on the overall effectiveness of regulation and supervision because of the expertise, experience and culture that develops within particular regulatory agencies and the approaches they adopt.7

It is perhaps too early to tell whether the twin peaks model is the optimal model for encouraging the appropriate regulatory culture — most jurisdictions have operated the model for less than 10 years, and many have not yet been tested by a financial crisis. What it may do better than other models is encourage the two regulators to coordinate and communicate with each other as neither regulator is able to ignore the other.

It is also worth noting the view of the DNB, who have said ‘the quality of supervision depends largely on the quality of the regulatory framework.’47 The significance of structure, while not to be overstated, is nevertheless important for regulatory culture.

Achieving synergies and central bank power

Australia stands alone in housing its prudential regulator completely outside of its central bank. Prior to recent legislative amendments, the UK used a ‘subsidiary’ model and has now adopted a model under which the BoE acts as the prudential regulator through the Prudential Regulation Committee. The Netherlands and Belgium rely on their central bank which has diminished responsibility for monetary policy, while NZ sees the central bank having complete responsibility for monetary policy and prudential regulation. There are two key observations that emerge from these decisions. The first is that achieving synergies is considered by some jurisdictions to be important. The second is that there appears to be a trend towards giving the central bank more power, not less.

The notion of achieving synergies is captured by the view that the central bank – with its highly specialised set of skills, processes and experience – is best situated to undertake the task of prudential supervision. There is great sway in the argument that the central bank would be the most obvious choice, particularly given the interconnected nature of monetary policy and prudential regulation. The counter-argument may be that this in itself creates a conflict, because the differences between prudential supervision and monetary policy are important and at times contradictory. It is clear from the experience in the Netherlands, Belgium, NZ and the UK that synergies are desirable. On the other hand, APRA has so far demonstrated that it is possible to conduct effective prudential supervision outside of the central bank.

The trend towards giving the central bank more power is important because it contradicts global patterns up until 2006 – a central bank conducting prudential supervision was the touchstone of a developing country, as compared to an ‘industrial’ one. Llewellyn notes

It is… the case that a higher proportion of central banks in developing countries are responsible for bank supervision than is the case with industrial countries where, until last year, the proportion has been falling (e.g. bank supervision has recently been taken away from the central bank in the UK, Austria and Australia) as more countries have adopted the integrated agency but have chosen for this not to be the central bank. This general trend has recently been reversed as some countries have extended the remit of the central bank to include the supervision of more than banks. The Netherlands is a notable example.7

Whereas up until 2006, having a central bank as the prudential supervisor was likely to be found in developing countries, the UK, Belgium, NZ and the Netherlands have bucked this trend. In fact, the legislative frameworks in these countries have been continuously reinforced to expand the powers of the central bank. For example, Belgium introduced ‘Twin Peaks II’ and NZ introduced two acts to expand the prudential supervision remit of the RBNZ. This suggests there is a trend, but it is in fact a trend of enhancing the power of the central bank – not diminishing it.

In 2010 the European Central Bank, after conducting a survey of various EU jurisdictions, made reference to this apparent trend. It noted

[T]he survey confirms the departure from the sectoral model and highlights a clear tendency towards further enhancing the role of central banks in supervisory activities. The latter development is underpinned by the experience during the financial crisis, which highlighted the information-related synergies between the central banking and the prudential supervisory function. This rationale may explain another important finding of the survey, namely that the involvement of the central bank in financial supervision seems to be increasingly strengthened through the adoption of the “twin peaks” model.2

It follows that there may well be a trend towards giving the central bank more power, and equally, a trend towards adoption of the twin peaks model.

The choice between hard law and soft law

The discussion of the ways in which twin peaks jurisdictions achieve regulatory coordination has revealed conflicting approaches. This is often expressed as a divide between the ‘soft law’ and ‘hard law’ approach to regulation. Hard law is seen as giving rise to legally enforceable obligations – such as duties to cooperate arising under statute or obligations to prepare to an MoU. Conversely, soft law relies on persuasive instruments that are themselves not legally binding, such as MoUs, mission statements, and informal coordinating bodies.176

The IMF, speaking with respect to Belgian-structured product regulation, has discussed the benefits of each approach:

A ‘soft law’ approach … provides a way for regulators to respond in a significantly more timely way to market change and innovation than may be possible under ‘hard law’ regulation. A move to ‘hard law’ on the other hand, provides industry with certainty, reduces opportunities for inconsistent application and interpretation and allows for other regulatory approaches (including rules about product governance) to be applied.14

The jurisdictions surveyed in this paper have revealed there are often many different ways of approaching the critical task of regulatory coordination. While coordination is central to the functioning of a twin peaks model, this does not necessarily mean it must be enshrined in statute. Each jurisdiction strikes its own unique balance of formal and informal means for achieving effective coordination between the market conduct and prudential regulator.

For example, Australia, NZ and the Netherlands fall at the ‘soft law’ of the spectrum and are largely informal. Belgium, as noted by the IMF, contains a mixture of ‘formal and… informal’ procedures for coordination.14 The UK, arguably, reflects a ‘hard law’ approach to cooperation. It establishes its coordinating body under statute, imposes multiple and overlapping duties to cooperate, creates a procedure for the government to define the lead regulator, and sets out a detailed process to enable the prudential regulator to veto the market conduct regulator.

The above table (Table 1) highlights the different approaches.
Table 1

Approaches to regulatory coordination

It is important to note that there is no single ‘best’ approach to ensuring that regulators cooperate. There are disadvantages to both the soft and hard law approach; the soft law approach might be regarded as less transparent and create uncertainty for the regulated community due to a lack of guidance, whereas the hard law approach may create a ‘tick-the-box’ mentality when it comes to complying with the legislation and discourage cooperation through other means.

Whether an approach is successful will likely depend more on the culture of the regulators than on the specific mechanism through which they cooperate. The choice between hard law and soft law is also likely to be influenced by what is expected in a given jurisdiction. For example, if twin peaks is being implemented in a jurisdiction where detailed, precise legislation governs every action of statutory bodies, then it may be most appropriate for the twin peaks governing legislation to follow suite. The key difficulty in assessing approaches is that it is often impossible to do so until a regulatory model fails, or is severely tested by a financial collapse and subsequent investigation.

Conclusion

This paper has shown that there is no archetypal twin peaks model. The variations in the regulatory design of the model reveal that the model is inherently flexible, and can and should be modified to suit local conditions and the local regulatory culture.

A survey of the dominant models of regulation, including twin peaks, reveals that there are unique challenges with each model. Critical to the success of the twin peaks model is ensuring that there is a clear delineation between the functions and objectives of the regulators and achieving effective coordination between the regulators. In fact, many of the choices that relate to implementation revolve around ensuring that the regulators are able effectively to coordinate the discharge of their functions. It remains to be seen whether the more formalised, legislation-heavy jurisdictions such as the UK are more effective at coordinating their functions compared with the soft law approaches of Australia and NZ.

The journey to the twin peaks model in each of the jurisdictions surveyed reveals that the model may be superior at handling the growing complexity of financial products and the rise of large financial conglomerates. As jurisdictions continue their move away from the sectoral or institutional model, they will be faced with the decision about which model is most appropriate. Instilling the appropriate regulatory culture and maximising synergies wherever possible will be influential factors in making this decision.

The DNB remarked in 2010 that the ‘twin peaks supervisory model… that was introduced in the Netherlands in 2004 is increasingly being adopted by other countries.’ Given the relatively embryonic nature of the model, it may be difficult to identify the strength of this long-term trend. However, as the actions of Australia, the UK, Belgium, NZ, the Netherlands and – in the not so distant future – South Africa all reveal, there is growing support in favour of the twin peaks model. It is particularly telling that no jurisdiction has yet abandoned the twin peaks model in favour of another model.

Notes

Acknowledgments

This article is part of a research project funded by the Melbourne Law School and the Centre for International Finance and Regulation entitled ‘Financial System Regulation – is Australia’s “Twin Peaks” Approach a Model for China?’ (see http://www.cifr.edu.au). The authors would like to thank Steve Kourabas and the anonymous reviewers for their helpful and insightful comments on this paper.

References and Notes

  1. 1.
    van Hengel, M., Hilbers, P. and Schoenmaker, D. (2013) Experiences with the Dutch twin peaks model: Lessons for Europe. In: Kellermann, JA, de Haan, J and de Vries F (eds.) Financial Supervision in the 21st Century. Springer, Berlin, p. 188.Google Scholar
  2. 2.
    European Central Bank (2010) Recent Developments in Supervisory Structures in the EU Member States (2007–2010), p. 5.Google Scholar
  3. 3.
    See for example, Godwin, A. and Schmulow, A. (2015) The financial sector regulation bill in South Africa: Lessons from Australia. South African Law Journal 132: 756.Google Scholar
  4. 4.
    See Department of Treasury (2008) Blueprint for a Modernized Financial Regulatory Structure. Washington D.C., pp. 13–14, 142–143.Google Scholar
  5. 5.
    International Monetary Fund (2012) Australia: Basel Core Principles for Effective Banking Supervision—Detailed Assessment of Observance. IMF Country Report No. 12/313, p. 4. For an analysis of why Australia weathered the Global Financial Crisis relatively well, see Hill, J. (2012) Why did Australia fare so well in the global financial crisis? In: E. Ferran, N. Moloney, J.G. Hill and J.C. Coffee Jr (eds.) The Regulatory Aftermath of the Global Financial Crisis. Cambridge University Press.Google Scholar
  6. 6.
    For a fuller treatment of these collapses, see Godwin, A. and Ramsay, I. (2015) Twin peaks – the legal and regulatory anatomy of Australia’s system of financial regulation. Journal of Banking and Finance Law and Practice 26: 240.Google Scholar
  7. 7.
    This paper makes no attempt to look at these models in detail, with the exception of twin peaks. For a more thorough analysis of the various models, see Llewellyn, D. (2006) Institutional Structure of Financial Regulation and Supervision: The Basic Rules. Paper presented at a World Bank seminar Aligning Supervisory Structures with Country Needs, Washington DC, 6 and 7 June 2006; Wymeersch, E. (2007) The structure of financial supervision in Europe: About single financial supervisors, twin peaks and multiple financial supervisors. European Business Organization Review 8(2): 237, p. 262.Google Scholar
  8. 8.
    See Part 3.Google Scholar
  9. 9.
    Some commentators refer to the sectoral or operational model as a functional approach, under which each type of business is ‘overseen by a separate, “functional” regulator’: Group of Thirty (6 October 2008) The Structure of Financial Supervision – Approaches and Challenges in a Global Marketplace.Google Scholar
  10. 10.
    Taylor, M. (1996) Peak Practice: How to Reform the UK’s regulator System. London, Centre for Study of Financial Innovation. In: Llewellyn, D. (2006) Institutional Structure of Financial Regulation and Supervision: The Basic Rules. Paper presented at a World Bank seminar Aligning Supervisory Structures with Country Needs. Washington DC, p. 19. Taylor has been identified as a leading proponent of the twin peaks model following his 1995 paper ‘Twin Peaks: A regulatory structure for the new century’: see Jensen, A. and Kingston, M. (2010) The Australian ‘Twin Peaks’ framework of financial system regulation: Australia and UK compared’. Journal of International Banking and Financial Law 548: 549.Google Scholar
  11. 11.
    This is considered in detail with respect to the Netherlands, and its decision to move to the twin peaks model. See Part 3.Google Scholar
  12. 12.
    Joint Committee on the draft Financial Services Bill (JCFSB). Draft Financial Services Bill – Report, together with formal minutes and appendices. HL 2010-12, 236, HC 2010-12, 1447, p. 73 [83].Google Scholar
  13. 13.
    Financial Stability Board (2013) Peer Review of the United Kingdom. Report, pp. 7–8.Google Scholar
  14. 14.
    This was considered with respect to Belgium: International Monetary Fund (IMF) (2013) Financial Sector Assessment Program Update – Technical Note – Securities Markets Regulation and Supervision. Report, p. 20.Google Scholar
  15. 15.
    This was considered in the New Zealand context: The New Zealand Institute of Chartered Accountants, Submission to the Commerce Select Committee, Financial Markets (Regulators and Kiwisaver) Bill, November 2010, para 32.Google Scholar
  16. 16.
    This was a serious concern in the HIH Report, as discussed in Part 3: see generally Commonwealth, The HIH Royal Commission (2003) The failure of HIH insurance: A corporate collapse and its lessons.Google Scholar
  17. 17.
    New Zealand Treasury (2010) Financial Sector Regulatory Agencies – Regulatory Impact Statement. Available at http://www.treasury.govt.nz/publications/informationreleases/ris/pdfs/ris-med-fsra-sep10.pdf/view.
  18. 18.
    For the recommendations of the Wallis Inquiry, see http://fsi.treasury.gov.au/content/downloads/FinalReport/overview.pdf. See also Jensen and Kingston (n 10) p. 549.
  19. 19.
    Financial System Inquiry (‘Wallis Report’) (1997) Overview – The Financial System: Towards 2010. Commonwealth of Australia, p. 17. According to the Wallis Report, ‘[r]egulation for the integrity of market conduct, consumer protection and the regulation of companies have significant synergies. These functions should therefore be combined’: pp 17–18.Google Scholar
  20. 20.
    Pan, E. (2011) Structural reform of financial regulation. Transnational Law and Contemporary Problems 19: 796, p. 854.Google Scholar
  21. 21.
    ibid, p. 854. See also Kremers, J. and Schoenmaker, D. (2010) Twin peaks: Experiences in the Netherlands. LSE Financial Markets Group Paper Series Paper 196, p. 1.Google Scholar
  22. 22.
    Netherlands Authority for the Financial Markets. (2015) About AFM, http://www.afm.nl/en/over-afm, accessed 18 June 2015.
  23. 23.
    International Monetary Fund (2011) Kingdom of The Netherlands: Financial System Stability Assessment. Report, p. 22.Google Scholar
  24. 24.
    Fiennes, T. and O’Connor-Close, C. (2012) The evolution of prudential supervision in New Zealand. Reserve Bank of New Zealand: Bulletin 75:5, p. 5.Google Scholar
  25. 25.
    Ministry of Economic Development (2010) Financial Sector Regulatory Agencies: Agency Disclosure Statement. Regulatory Impact Statement, p. 4. Available at http://www.med.govt.nz/business/business-law/pdf-docs-library/ris-financial-sector-regulatory-agencies.pdf.
  26. 26.
    Prudential supervision of ‘specified institutions’ has officially been the responsibility of the Reserve Bank since 1986, after s 38I was inserted into the Act by an amendment. This was later replaced by the Reserve Bank of New Zealand Act 1989 (RBNZ Act) s 67 that provided ‘The Bank shall… undertake prudential supervision of registered banks.’Google Scholar
  27. 27.
    Cabinet, Office of the Minister of Commerce (2010) Creating a Financial Markets Authority and Enhancing Kiwisaver Governance and Reporting, para 28. See also Commerce Committee, House of Representatives (2011) Inquiry into finance company failures, p. 11: ‘[T]he supervisory framework was fragmented and insufficiently rigorous… The several regulators operated under relatively narrow legislative mandates. Overlapping responsibilities and inadequate funding led to things slipping through the cracks…One of the difficulties with this regulatory framework was that it lacked a clear structure, and the various Acts, amended numerous times over the years, could be confusing.’Google Scholar
  28. 28.
    See for example, State Services Commission. (2015) Board Appointment and Induction Guidelines – Chapter 5: Removal from Office, http://www.ssc.govt.nz/node/9108, accessed 5 June 2015.
  29. 29.
    Parliamentary Joint Committee on Corporations and Financial Services (2014) Oversight of the Australian Securities and Investment Commission and The Takeovers Panel. Opening statement from Sean Hughes, 5 September 2014, p. ii.Google Scholar
  30. 30.
    Bank of England (2011) Joint Committee on the draft Financial Services Bill – Evidence. Evidence presented to Joint Committee, Published 16 December 2011, p. 198. Available at http://www.parliament.uk/business/committees/committees-a-z/joint-select/draft-financial-services-bill/publications/.
  31. 31.
    KPMG (2012) Twin-peaks regulation: key changes and challenges. Report, p. 3.Google Scholar
  32. 32.
    Emma Murphy and Stephen Senior (2013) Changes to the Bank of England. Bank of England Quarterly Bulletin Q1, p. 20.Google Scholar
  33. 33.
    Richter, W. (2012) The Dexia Bailout is Turning into a €162 Billion Nightmare for Belgian Taxpayers. Business Insider Australia, 25 February 2012, http://www.businessinsider.com.au/belgians-get-cold-feet-as-bailout-queen-dexia-drags-them-toward-abyss-2012-2.
  34. 34.
    See generally Kelion, L. (2011) How Dexia was caught out by the Eurozone debt crisis. BBC News, 6 October 2011, http://www.bbc.com/news/business-15180153.
  35. 35.
    Tait, N. and Steen, M. (2008) Financial crisis fells Belgian leadership. Financial Times online, 23 December 2008, http://www.ft.com/intl/cms/s/0/189bfaa2-d03b-11dd-ae00-000077b07658.html#axzz3fqSPmokz. The Belgian banking sector has significant international exposure – ‘around 50 per cent of total bank assets are held by subsidiaries or branches of foreign owned banks.’ In 2008, the banking sector was also very concentrated, with ‘the four largest banking groups account[ing] for approximately 80 per cent of all deposits, assets and loans in Belgium’: High Level Committee on a New Financial Architecture (2009) Final Report, p. 39.
  36. 36.
    The various laws concerning the NBB have been brought together in one document, ‘Statutes of the NBB.’ There is an unofficial translation available on the NBB website: https://www.nbb.be/doc/ts/enterprise/juridisch/e/statutes_e.pdf.
  37. 37.
    Linklaters (2013) ‘Twin Peaks II – Belgium’. Report, p. 1. Available at www.linklaters.com/pdfs/mkt/brussels/The_Twin_Peaks_II_Law.pdf.
  38. 38.
    Unofficial translation, last amended by Law of 31 July 2013 (Belgian Official Gazette 30 August 2013). Available at www.fsma.be/~/media/Files/fsmafiles/…/wet…/law_02-08-2002.ashx.
  39. 39.
    The IMF has noted the similarities between the Belgian model and the Dutch model: International Monetary Fund (2011) Institutional Models for Macroprudential Policy. Staff Discussion Note 11/18, p. 12.Google Scholar
  40. 40.
    This question goes to the issue of operational independence.Google Scholar
  41. 41.
    APRA Act, s 7.Google Scholar
  42. 42.
    APRA Act, s 13.Google Scholar
  43. 43.
    See Recommendation 32. Coordination and cooperation were to be achieved by making provision for ‘full information exchange between the RBA and [APRA]’ and for ‘RBA participation in [APRA] inspection teams’. Further, ‘[a] bilateral operational coordination committee, chaired by an RBA deputy governor, should be established to coordinate information exchange, reporting arrangements on financial system developments, and other ongoing operational cooperation between the RBA and [APRA], including cooperation in establishing clear procedures for the management of regulated entities which experience financial difficulties’: Wallis Report, ‘Overview’ (n 19), p. 11–12.Google Scholar
  44. 44.
    Dr Edey (Assistant Governor of the Reserve Bank of Australia) (2011) Evidence presented to the Joint Committee on the draft Financial Services Bill. Published 16 December 2011, p. 435.Google Scholar
  45. 45.
    IMF (2011) Institutional Models for Macroprudential Policy. Staff Discussion Note 11/18, p. 12.Google Scholar
  46. 46.
    HM Treasury (2011) Financial Regulation: a preliminary consideration of the Government’s proposals. Report, pp. 32–33. Available at http://www.publications.parliament.uk/pa/cm201011/cmselect/cmtreasy/430/43002.htm. It is relevant to note that under s 9H of the Bank of England Act 1998, the Financial Policy Committee (FPC) can give directions to the FCA in respect of macro-prudential measures, even though the FCA is an independent public body.
  47. 47.
    HM Treasury (2015) Bank of England Bill: technical consultation, p. 3.Google Scholar
  48. 48.
    NBB (2011) Report 2011 – Corporate Report. Report, p. 18.Google Scholar
  49. 49.
    Reserve Bank of New Zealand Act 1986, Part Vc ‘Prudential Supervision of Specified Institutions’, s 38I. This was replaced by the Reserve Bank of New Zealand Act 1989.Google Scholar
  50. 50.
    Ministry of Economic Development (2010) Financial Sector Regulatory Agencies: Agency Disclosure Statement. Regulatory Impact Statement, 14 September 2010, p. 4. Available at http://www.med.govt.nz/business/business-law/pdf-docs-library/ris-financial-sector-regulatory-agencies.pdf.
  51. 51.
    Basel Committee on Banking Supervision (2012) Core Principles for Effective Banking Supervision, p. 21.Google Scholar
  52. 52.
    International Association of Insurance Supervisors (2011) Insurance Core Principles, p. 1. See also IOSCO (2010) Objectives and Principles of Securities Regulation, Principle 1: ‘The responsibilities of the Regulator should be clear and objectively stated.’Google Scholar
  53. 53.
    See the UK House of Commons Treasury Committee (2012) Financial Conduct Authority, Twenty-sixth Report of Session 2010–2012, 2012, p. 11.Google Scholar
  54. 54.
    Financial Markets Authority Act 2011 (FMA Act), s 8.Google Scholar
  55. 55.
    FMA Act, s 9.Google Scholar
  56. 56.
    ASIC Act, s 1.Google Scholar
  57. 57.
    Financial Services Act 2012 (FS Act), Part 2 amendments to the FSM Act, s 1B(2).Google Scholar
  58. 58.
    FS Act, Part 2 amendments to the FSM Act, s 1B(3).Google Scholar
  59. 59.
    FS Act, Part 2 amendments to the FSM Act, s 3B.Google Scholar
  60. 60.
    FS Act, Part 2 amendments to the FSM Act.Google Scholar
  61. 61.
    AFM (2015) Strategic Objectives, http://www.afm.nl/en/over-afm/werkzaamheden/strategische-doelstellingen, accessed 15 June 2015.
  62. 62.
    Act on Financial Supervision 2006 (AFS Act) [unofficial translation].Google Scholar
  63. 63.
    Letter from Chairman of the FSMA and Governor of the NBB (2011) Re. new supervisory architecture for the Belgian financial sector, 1 April 2011, p. 1. Available at http://www.iopsweb.org/researchandworkingpapers/47498500.pdf; NBB (2013) Report 2013 – Preamble, Economics and financial developments, Prudential regulation and supervision’. Report, p. 3.
  64. 64.
    FS Act, Part 2 amendments to the FSM Act, s 1E.Google Scholar
  65. 65.
    FS Act, Part 2 amendments to the FSM Act, s 1 J.Google Scholar
  66. 66.
    ASIC Submission to Financial System Inquiry (FSI) (2014) Final Report, p. 18–19. ASIC noted that the Australian Competition and Consumer Commission (ACCC) would remain the competition regulator. The call for a competition mandate was also made in the submission to the FSI by Minter Ellison: Minter Ellison, Submission, Financial System Inquiry, March 2014. The FSI reviewed Australia’s financial system and examined many issues that are relevant to the ways in which the twin peaks model works in Australia.Google Scholar
  67. 67.
    FSI (2014) Final Report, p. 237.Google Scholar
  68. 68.
    RBNZ Act, s 67.Google Scholar
  69. 69.
    RBNZ Act, s 68.Google Scholar
  70. 70.
    Insurance Prudential Supervision Act 2010, s 3.Google Scholar
  71. 71.
    Section 3.Google Scholar
  72. 72.
    NBDT Act, s 7.Google Scholar
  73. 73.
    NBDT Act, s 56.Google Scholar
  74. 74.
    APRA Act, s 8.Google Scholar
  75. 75.
    AFS Act, s 1:24.Google Scholar
  76. 76.
    FS Act, Part 2 amendments to the FSM Act, s 2B(1).Google Scholar
  77. 77.
    FS Act, Part 2 amendments to the FSM Act, s 2B(3)(a), (b).Google Scholar
  78. 78.
    FS Act, Part 2 amendments to the FSM Act, s 2C.Google Scholar
  79. 79.
    FS Act, Part 2 amendments to the FSM Act, s 2D.Google Scholar
  80. 80.
    FS Act, Part 2B amendments to the FSM Act, s 9.Google Scholar
  81. 81.
    Statutes of the NBB (unofficial translation, incorporating amendments by the Council of Regency on 14 January 2015) art 12. This is an unofficial translation available on the NBB website: https://www.nbb.be/doc/ts/enterprise/juridisch/e/statutes_e.pdf.
  82. 82.
    Statutes of the NBB, arts 14-18.Google Scholar
  83. 83.
    Statutes of the NBB, arts 23 and 23bis.Google Scholar
  84. 84.
    FS Act, Part 2 amendments to the FSM Act, s 2B(1).Google Scholar
  85. 85.
    FS Act, Part 2 amendments to the FSM Act, s 2B(3)(a), (b).Google Scholar
  86. 86.
    FS Act, Part 2 amendments to the FSM Act, s 2H.Google Scholar
  87. 87.
    Explanatory Note, FS Act, para 10. The details provision is set out in FS Act, Part 2 amendments to the FSM Act, s 3I.Google Scholar
  88. 88.
    International Monetary Fund (2011) United Kingdom: IOSCO Objectives and Principles of Securities Regulation: Detailed Assessment of Implementation. Report No 11/232, July 2011, p. 23.Google Scholar
  89. 89.
    FMA Act, s 20.Google Scholar
  90. 90.
    ASIC Act, s 14.Google Scholar
  91. 91.
    FS Act, s 77.Google Scholar
  92. 92.
    RBNZ Act, s 33.Google Scholar
  93. 93.
    RBNZ Act, s 14.Google Scholar
  94. 94.
    APRA Act, s 10.Google Scholar
  95. 95.
    De Nederlandsche Bank (2015) Independence and secrecy, http://www.dnb.nl/en/about-dnb/organisation/onafhankelijkheid-en-geheimhouding/index.jsp, accessed 15 June 2015.
  96. 96.
    APRA Act, s 10(1).Google Scholar
  97. 97.
    Explanatory Memorandum to the Australian Prudential Regulation Authority Bill 1998 (Cth), para 4.10.Google Scholar
  98. 98.
    FS Act, ss 73 & 74.Google Scholar
  99. 99.
    RBNZ Act, s 68B.Google Scholar
  100. 100.
    APRA Act, s 12; ASIC Act, s 12.Google Scholar
  101. 101.
    AFS Act, s 1:43 [repealed]; International Monetary Fund (2011) Kingdom of The Netherlands – Netherlands: Publication of Financial Sector Assessment Program Documentation – Detailed Assessment of Observance on IOSCO Principles and Objectives of Securities Regulation. Report, p. 28.Google Scholar
  102. 102.
    AFS Act, s 1:43.Google Scholar
  103. 103.
    RBNZ Act, s 68B. This directions power has been extended by the NBDT Act, s 9 and the Insurance (Prudential Supervision) Act 2010, s 13.Google Scholar
  104. 104.
    RBNZ Act, s 68B(5). Where a direction is given under section 68B of the RBNZ Act, this will be reflected in the Statement of Intent (SOI) that the bank must provide each year under section 162A. Section 162B(1)(da) relevantly provides that the SOI must indicate ‘if a direction has been given under section 68B’ and ‘how the bank had regard to the direction.’ No directions appear to have been issued under section 68B of the RBNZ Act to date.Google Scholar
  105. 105.
    Cabinet Economic Development Committee, Office of the Minister of Finance and Office of the Minister of Commerce (2007) Institutional arrangements for prudential regulation, 13 June 2007, para 18. Available at http://www.med.govt.nz/business/business-law/pdf-docs-library/past-work-and-older-topics-pdfs/review-of-financial-products/arrangements-for-prudential-supervision.pdf.
  106. 106.
    New Zealand, Parliamentary Debates, 11 December 2007, 644-13830 (Dr Lockwood Smith) ‘Reserve Bank of New Zealand Amendment Bill (No 3) – First Reading’.Google Scholar
  107. 107.
    New Zealand, Parliamentary Debates, 2 September 2008, 649-18210 (Craig Foss) ‘Reserve Bank of New Zealand Amendment Bill (No 3) – Second Reading Speech’.Google Scholar
  108. 108.
    APRA Act, s 12.Google Scholar
  109. 109.
    FS Act, Part 2 amendments to the FSM Act.Google Scholar
  110. 110.
    FS Act, Part 2 amendments to the FSM Act, s 3H.Google Scholar
  111. 111.
    See APRA Act, s 12 and ASIC Act, s 12.Google Scholar
  112. 112.
    For a discussion of the importance of coordination in the context of New Zealand, see The New Zealand Institute of Chartered Accountants, Submission to the Commerce Select Committee, Financial Markets (Regulators and Kiwisaver) Bill, November 2010, paras 63-4: ‘coordination between the regulatory agencies will be essential, given the areas of complementary interest between the agencies…’Google Scholar
  113. 113.
    International Monetary Fund (2012) Australia: Financial System Stability Assessment. IMF Country Report No. 12/308, November 2012, p. 28.Google Scholar
  114. 114.
    APRA Act, s 10A.Google Scholar
  115. 115.
    FSM Act, s 354.Google Scholar
  116. 116.
    FS Act, Part 2 amendments to the FSM Act, s 3D.Google Scholar
  117. 117.
    FS Act, Part 2 amendments to the FSM Act, s 3D(1)(b).Google Scholar
  118. 118.
    FS Act, Part 2 amendments to the FSM Act, s 3D(2)(a), (b).Google Scholar
  119. 119.
    ibid s 3D(4). The operation of this provision was explained in the Explanatory Note to the Financial Services Act 2012: Explanatory Note, FS Act, para 168.Google Scholar
  120. 120.
    Memorandum of Understanding between the Financial Conduct Authority and the Bank of England, including the Prudential Regulation Authority (March 2015) para 8.Google Scholar
  121. 121.
    FS Act, s 64.Google Scholar
  122. 122.
    AFS Act, s 1:46(1).Google Scholar
  123. 123.
    Article 45bis.Google Scholar
  124. 124.
    NBB (2013) Report 2013 – Preamble, Economics and financial developments, Prudential regulation and supervision, pp. 227-8.Google Scholar
  125. 125.
    Law of 22 February 1998 Establishing the Organic Statute of the NBB (unofficial translation March 2015) art 77bis § 1(1°). Available at https://www.nbb.be/doc/ts/enterprise/juridisch/e/organic_act.pdf.
  126. 126.
    Linklaters (2011) Twin Peaks – More than a new supervision model. Report, p. 9.Google Scholar
  127. 127.
    APRA Act, s 56(5); ASIC Act, s 127(2A)(a), (b), (c).Google Scholar
  128. 128.
    FMA Act, s 30.Google Scholar
  129. 129.
    For the full circumstances regarding how this is to occur, see FMA Act, s 59.Google Scholar
  130. 130.
    RBNZ Act, s 105.Google Scholar
  131. 131.
    The FS Act amendments to the FSM Act replace various sections of the Act, and include a new provision stating that ‘the body corporate previously known as the Financial Services Authority is renamed the Financial Conduct Authority.’ See FS Act, Part 2 amendments to the FSM Act, s 1A.Google Scholar
  132. 132.
    FSM Act, ss 348 & 349. For a list of the regulations that address the exceptions to confidentiality, see FSM (Disclosure of Confidential Information) Regulations 2001, r 3.Google Scholar
  133. 133.
    Memorandum of Understanding between the Financial Conduct Authority and the Bank of England, including the Prudential Regulation Authority (March 2015) para 35.Google Scholar
  134. 134.
    Dutch National Bank (2010) Supervisory Strategy 2014-2018. Report, pp. 6, 19.Google Scholar
  135. 135.
    AFS Act, art 1:90(4).Google Scholar
  136. 136.
    IMF (2013) Financial Sector Assessment Program: Detailed Assessment of Observance – Insurance Core Principles. IMF Country Report No. 13/134, May 2013, pp. 21-2. On page 49, the IMF noted the following procedures were in place with respect to exchanging information: ‘Where a financial institution commit[s] serious breaches of the rules of conduct, the FSMA shall inform the NBB of the alleged breach before taking the following measures: a) prohibiting or suspending (in whole or in part) the activities of the institution concerned; or b) ordering the replacement of its management. Where such breaches are systematic, the FSMA may request the NBB to revoke the license of the institution. The NBB may only refuse the request if the withdrawal is likely to compromise the stability of the financial system and the FSMA may challenge the NBB’s decision before the MoF, who makes a final decision.’Google Scholar
  137. 137.
    Statutes of the NBB, art 56.Google Scholar
  138. 138.
    Law of 2 August 2002, art 74. The legislation also outlines the limited grounds under which confidential information may be communicated.Google Scholar
  139. 139.
    Memorandum of Understanding between the Financial Markets Authority and the Reserve Bank of New Zealand (9 September 2011). Available at http://www.rbnz.govt.nz/regulation_and_supervision/banks/relationships/4525498.pdf.
  140. 140.
    For example, the MoU makes the following comments with respect to information exchange: (i) For example, the MoU makes the following comments with respect to information exchange: The exchange of relevant information in a timely manner is essential to both FMA and RBNZ… (ii) Both agencies agree to make relevant information available to the other, subject to any legislative requirements, in a timely and efficient manner… (iii) Should either agency, during the course of its work, identify a harm or potential harm falling within the jurisdiction of the other, where legal, all relevant information will be passed to the other agency as a matter of urgency… (iv) Where the cost of providing material under this MoU is considered significant by either agency, the actual and mutually agreed costs can be fairly claimed from the other…Google Scholar
  141. 141.
    Memorandum of Understanding between the Australian Prudential Regulation Authority and the Australian Securities and Investments Commission, 18 May 2010. Available at http://www.apra.gov.au/AboutAPRA/Documents/ASIC-MoU.pdf.
  142. 142.
    Memorandum of Understanding between the Financial Markets Authority and the Reserve Bank of New Zealand (9 September 2011). Available at http://www.rbnz.govt.nz/regulation_and_supervision/banks/relationships/4525498.pdf.
  143. 143.
    Bank of England. (2015) Memoranda of Understanding, http://www.bankofengland.co.uk/about/Pages/mous/default.aspx, accessed 7 June 2015.
  144. 144.
    FS Act, Part 2 amendments to the FSM Act, s 3E(1)(a).Google Scholar
  145. 145.
    FS Act, Part 2 amendments to the FSM Act, s 3E(4),(5) & (6).Google Scholar
  146. 146.
    FS Act, s 65.Google Scholar
  147. 147.
    Bank of England (2015) Memoranda of Understanding, http://www.bankofengland.co.uk/about/Pages/mous/default.aspx, accessed 7 June 2015.
  148. 148.
    FS Act, s 66.Google Scholar
  149. 149.
    Covenant between Stichting Autoriteit Financiële Markten and De Nederlandsche Bank N.V, 2.Google Scholar
  150. 150.
    Netherlands Authority for the Financial Markets. (2015) External Stakeholders, http://www.afm.nl/en/professionals/over-afm/organisatie/ext-stakeholders, accessed 22 February 2015.
  151. 151.
    Hengel and others (n 1) p. 191. They noted with respect to the GFC: ‘An important advantage was that the laws allowed for information sharing in the context of crisis management. Already before the crisis, DNB and the AFM had concluded a covenant that specified cooperation and joint responsibilities between both supervisors. The respective roles of DNB and the Ministry of Finance in crisis management were described in a Memorandum of Understanding. On that basis a crisis management group was set up to include also representatives from the Ministry of Finance and the AFM, which met daily.’Google Scholar
  152. 152.
    Covenant between Stichting Autoriteit Financiële Markten and De Nederlandsche Bank N.V, 2.Google Scholar
  153. 153.
    Cooperation Protocol Between the NBB and the Financial Services and Markets Authority in the Field of Supervision and Oversight of Market Infrastructures (translation of the MoU dated 18 October 2012). Available at https://www.nbb.be/doc/cp/eng/vi/accord_collaboration/accord/pdf/protocole_nbb_fsma.pdf. The NBB has noted that ‘the “twin peaks” model was… reinforced by the signing of cooperation agreements between the Bank and the Financial Services and Markets Authority (FSMA) with the aim of improving the exchange of information between these two institutions’: NBB (n 124), p. 212.
  154. 154.
    Law of 2 August 2002, art 45bis.Google Scholar
  155. 155.
    Cooperation Protocol Between the NBB and the Financial Services and Markets Authority in the Field of Supervision and Oversight of Market Infrastructures (translation of the MoU dated 18 October 2012), p. 1.Google Scholar
  156. 156.
    See, for example, arts 1-7, and arts 9-18 on the more specific aspects of information sharing, such as the deadlines in which requests for information must be addressed.Google Scholar
  157. 157.
    Cooperation Protocol Between the NBB and the Financial Services and Markets Authority in the Field of Supervision and Oversight of Market Infrastructures (translation of the MoU dated 18 October 2012), p. 11.Google Scholar
  158. 158.
    Reserve Bank of New Zealand (2013) Council of Financial Regulators – Terms of Reference, http://www.rbnz.govt.nz/regulation_and_supervision/banks/relationships/4525791.html, accessed 25 May 2015.
  159. 159.
    The Council of Financial Regulators (2015) About the CFR, http://www.cfr.gov.au/about-cfr/index.html, accessed 11 May 2015. See also IMF, ‘Australia: Core Principles’ (n 5): ‘The CFR does not have a legal personality, nor does it have powers separate from its member agencies. Its members share information and views and advise the Government on Australia’s financial system architecture. APRA and the RBA both have mandates for financial stability and have legal gateways to share institution-level data that is needed for them to carry out their respective duties.’
  160. 160.
    Reserve Bank of Australia, Submission, Financial System Inquiry, 2014, p. 66.Google Scholar
  161. 161.
    International Monetary Fund (2011) United Kingdom: Financial System Stability Assessment, p. 37.Google Scholar
  162. 162.
    FS Act, Part 1 amendments to the FSM Act, s 9H.Google Scholar
  163. 163.
    FS Act, Part 1 amendments to the FSM Act, s 9O.Google Scholar
  164. 164.
    FS Act, Part 1 amendments to the FSM Act, s 9P.Google Scholar
  165. 165.
    FS Act, Part 1 amendments to the FSM Act, s 9Q.Google Scholar
  166. 166.
    Financial Stability Board (FSB) (2014) Peer Review of the Netherlands, p. 23.Google Scholar
  167. 167.
  168. 168.
    Financial Stability Committee (2015) About the FSC, http://www.financieelstabiliteitscomite.nl/en/about_fsc, accessed 22 February 2015.
  169. 169.
    Memorandum of Understanding between the NBB and the Financial Services and Markets Authority, art 6.Google Scholar
  170. 170.
    Memorandum of Understanding between the NBB and the Financial Services and Markets Authority, art 24.Google Scholar
  171. 171.
    KPMG, Submission, Financial System Inquiry, 31 March 2014, p. 5.Google Scholar
  172. 172.
    As part of it submission to the FSI, the National Australia Bank recommended that ‘the Council of Financial Regulators (CFR) should be given a more formal structure and be tasked by the Treasurer to coordinate the implementation of regulatory change by APRA and ASIC: NAB, Submission, Financial System Inquiry, March 2014, s 3.2.2.Google Scholar
  173. 173.
    This reflects recent changes under the Bank of England and Financial Services Act 2016. See also Bank of England (2015) Financial Policy Committee, http://www.bankofengland.co.uk/financialstability/pages/fpc/default.aspx, accessed 22 February 2015.
  174. 174.
    Memorandum of Understanding between the NBB and the Financial Services and Markets Authority, art 24.Google Scholar
  175. 175.
    Jeffrey Carmichael (2003) Australia’s Approach to Regulatory Reform. Paper given at World Bank Regulation Conference, December 2003, p. 3. Available at http://info.worldbank.org/etools/docs/library/86146/Carmichael_Australia.doc.
  176. 176.
    For a fuller discussion of hard law and soft law, see generally Godwin and Ramsay (n 6).Google Scholar

Copyright information

© Macmillan Publishers Ltd 2016

Authors and Affiliations

  1. 1.The University of MelbourneMelbourneAustralia

Personalised recommendations