Home Country Control with Consent: A New Paradigm for Ensuring Trust and Cooperation in the Internal Market?

Abstract Home country control has been a long-standing principle of supervisory governance in the internal market. However, in the wake of the financial crisis, the principle has come under stress. This chapter looks at ways to deal with home country control by putting forward for discussion a new paradigm which I will coin ‘home country control with consent’ (HCC-C). My aim is to examine the building blocks of HCC-C but also to reflect more generally on the merit of a (mostly horizontal) supervisory arrangement which allows other (host) actors to get involved in the decision making of a home state authority. To describe such involvement, I will use the term ‘interference’. The basic problematic that I seek to address is that of ensuring cooperation and trust between national competent authorities. To identify the building blocks of HCC-C, I will turn to the recently enacted European Market Infrastructure Regulation (EMIR) which provides a possible, even if embryonic, template for HCC-C.


I. INTRODUCTION
H OME COUNTRY CONTROL has been a long-standing principle of supervisory governance in the internal market. However, in the wake of the financial crisis, the principle has been severely * A version of this chapter was presented at a seminar organised by the Centre for European Legal Studies, Cambridge in October 2012. I wish to thank all the participants for their questions and helpful comments. Thanks are also due to Jean-Louis Beckers and Oleg Shmeljov. questioned. Specifically, doubts were expressed about whether it can be an effective arrangement absent burden-sharing arrangements between Member States. 1 The European System of Financial Supervision (ESFS) was partly adopted as a response to issues with home country control. New European Supervisory Authorities (ESAs) were established. They were vested, inter alia, with intervention powers that allow them to intervene in the decision making of competent authorities 2 under predefined conditions and requirements. Efforts to regulate the post-trading industry have also led to the adoption of new supervisory arrangements. The relevant rules and requirements are found in the European Market Infrastructure Regulation (EMIR). 3 They concern, inter alia, the approval of central counterparties (CCPs). Under these rules, various actors get involved in the decision making of one authority, that is, the CCP's competent authority, which is, in essence, the CCP's home authority. 4 This chapter examines these new supervisory arrangements in an effort to consider ways of improving the home country control principle. For the present purposes, they provide a template-albeit a somewhat embryonic one-for a new procedure which I will coin hereinafter 'home country control with consent' or HCC-C for short. 5 The aim of the chapter is to examine the building blocks of HCC-C, but also to reflect more generally on the merit of a (mostly horizontal) supervisory arrangement, which allows other (host) actors to get involved in the decision making of a home state authority. To describe the involvement of these other actors, I will use the term 'interference'. The basic problematic that I will seek to address is that of ensuring cooperation and trust between national competent authorities. Specifically, the question is whether interference has anything to offer in this context. Improving cooperation and trust has long been a concern at the EU level. Both are considered crucial for delivering a well-functioning single market. Most recently, the issue of trust has fed into discussions on the establishment of a Eurozone banking union and on vesting supervisory competence in the European Central Bank (ECB). 6 In this chapter, I do not seek to discuss the merit of a possible banking union or, for that matter, to make proposals on the supervision of the banking sector. Instead, I hope to discuss what might be required to ensure cooperation and trust in a decentralised setting.
The chapter proceeds as follows. Section II begins by discussing home country control and the problems related to it. Section III turns to HCC-C. It first introduces EMIR and its subject matter, after which it turns to the EMIR arrangements as a potential template for HCC-C. Section IV seeks to reflect more generally on the rationale underpinning HCC-C: greater host state interference in the decision making of a home competent authority. Section V concludes.

II. HOME COUNTRY CONTROL UNDER STRESS
The starting point for assessing HCC-C is the much more familiar principle of home country control. Home country control has a long history in the internal market or, for that matter, the banking and financial market sectors. The thrust of this principle is that the main locus of control over the internal market activities of an economic actor resides with the competent authority/authorities of a single Member State: the home Member State. In the banking sector, home country control was introduced in the 1989 Second Banking Coordination Directive together with the principle of mutual recognition (known as the single banking licence). 7 It replaced a system based on host country control. Under the latter, a market actor was meant to comply with the rules and regulations that were also applicable to domestic actors in the host Member State. 8 Home country control has become a firm fixture of the internal market and, together with mutual recognition, has contributed to generating greater cross-border mobility and ultimately market integration. It operates across fields for the 'solo' 6 Thus, the Commission noted in relation to the supervision of the banking sector that it was 'too fragmented to face current challenges' and as such was 'not conducive to the necessary trust between Member States'. It concluded that there was a need for 'political agreement on more and independent EU supervision'; see European Commission, 'Update-The Banking supervision of individual actors. 9 But also in more complex settings, such as in the context of cross-border group supervision, a similar preference for vesting certain powers in the competent authority of a single Member State can be witnessed. 10 In the case of cross-border banking groups, for instance, this preference translates into vesting decision-making powers over the group in the so-called consolidating supervisor, ie, the home authority of the group. The home country principle is not, however, an absolute principle. Even for the solo supervision of single entities where home country control is most developed, limitations will subsist: some powers or areas of action will generally remain in the hands of host state supervisors. 11 But these (host state) concessions are not such as to undermine the home state bias.
Given the home state bias, it is not surprising that especially in a time of crisis, the home state principle is likely to cause friction between states. Thus, despite its strengths as an internal market principle, home country control has not necessarily mapped well onto the realities of increasingly integrated markets for host Member States. 12 In the banking sector, the failure of the Icelandic banking industry provides a telling example of the problems faced by host states. Icelandic banks operated in Member States through branches that were under the control of the domestic (ie, Icelandic) supervisor. 13 As a result, the UK found itself exposed to the pre-and postcrisis actions of Icelandic authorities. The prospect of UK depositors failing to be reimbursed in the wake of the collapse of the Icelandic banking industry ultimately led the UK to take extraordinary and controversial action by securing relevant assets on the basis of powers under the Anti-terrorism, Crime and Security Act 2001. 14 Problems have also emerged because of the geographical distribution of large banking groups and the competence imbalances between Member States that have ensued. Thus, the fact that mainly West European banking groups have dominated the banking markets of Central and East European 9 Note that while home state control applies when activities are exercised through branches in host Member States, subsidiaries will be subject to the supervisory competence of the authority of the Member State in which the subsidiary is established. See below in relation to cross-border groups. 10 In the case of group supervision, it is worth noting that the 'home-host' terminology is often used more loosely. Thus, when cross-border groups are involved, it is common to use the term 'host' even in relation to subsidiaries (eg, FSC Report (n 7) 16). 11  OJ L145/1, arts 13(9), 32(7) and 62. 12 See also P Schammo, 'EU Day-to-Day Supervision or Intervention-based Supervision' (2012) 32 OJLS 771. 13 Icelandic banks could operate in Member States under EU mutual recognition arrangements, which extend to members of the European Economic Area (EEA). 14 See, eg, HL Report on EU Financial Regulation and Supervision (n 1) 51. countries 15 has left these 'host' states feeling particularly exposed to the decisions and courses of action taken by competent authorities located outside their territories. Given the issues associated with the home country principle for 'host' Member States, it should not come as a surprise that Member States have been less willing to see the principle being extended. This was apparent in the context of two earlier Commission proposals, which sought, inter alia, to strengthen group supervision: 16 that is, the amended Capital Requirements Directive (CRD) and the Solvency II Directive. 17 The former is about to be repealed and our interest in it is merely historical. The amended CRD in fact comprised two directives. It dealt with capital requirements for banks and investment firms and transposed the Basel requirements on capital standards in the EU. 18 New legislation was adopted in June 2013. 19 The Solvency II Directive deals with the regulation of the insurance industry and especially with capital requirements for insurance undertakings. In both cases, Member States resisted proposals to extend the powers of the 'home' supervisor, that is: (i) the consolidating supervisor under the CRD; 20 and (ii) the group supervisor under the Solvency II Directive. Thus, in the case of the CRD, the EU legislature made provision for the consolidating supervisor to decide certain matters over and above other competent authorities 15 'Report of the High-level Group on Financial Supervision in the EU' (Brussels, 25 February 2009) 71. Available at: http://ec.europa.eu/internal_market/finances/docs/de_ larosiere_report_en.pdf (hereinafter 'de Larosière Report'); HL Report on EU Financial Regulation and Supervision (n 1) 51. 16  (for example, in respect of model validation). 21 But, crucially, it decided against extending this power to other proposed matters. Specifically, Member States opposed the Commission proposal to reinforce the power of the consolidating supervisor in relation to so-called 'Pillar 2' capital add-ons-ie, additional capital requirements-for subsidiaries in other Member States. 22 Instead, the CRD foresaw, as elsewhere, 23 a joint decision-making process, but a process which, if unsuccessful, allowed national competent authorities to decide their own courses of action at their respective competence levels. 24 In the case of the Solvency II Directive, the Commission's proposal also faced obstacles. The proposal in question foresaw that the parent undertaking of a group be allowed, under the supervision of the group supervisor, to satisfy part of the solvency requirements of its subsidiaries by way of promises of support. 25 However, the Commission's attempt to introduce such a 'group support' regime was defeated. Opposition emerged from a group of Member States that were fearful of being made dependent on the actions of supervisors from some larger Member States where insurance groups had their headquarters. 26 The CRD and the Solvency II Directive are representative of the problems associated with the logic that underpins the home country principle in a more integrated market; that is, of seeking to consolidate decision-making powers in one Member State at the expense of others in order to further improve the functioning of the internal market. As the de Larosière report put it when looking at EU financial supervision in the wake of the financial crisis: 'In both cases a strong number of countries-including all new Member States for Solvency II and Member States unanimously in the case of the CRD-have decisively rejected changing the current balance of home and host state regulation'. 27 21 Article 129(2), para 5. Note that the power of the consolidating supervisor could nevertheless be curtailed by the European Banking Authority exercising its power to settle disagreements under art 19 of its founding regulation (Regulation (

A. EMIR and OTC Derivatives Trading: Overview
To make sense of EMIR's arrangements, it is helpful to start first with its subject matter: over-the-counter (OTC) derivatives and market infrastructures. A derivative is a financial instrument-essentially a contract-whose value is based on something else: for instance, an underlying financial asset, a commodity, an index or an event. 29 Derivatives have been popular for different reasons. They have been used to speculate, but they are also often a very useful and legitimate tool for hedging risk-for example, credit risk or movements in, say, interest rates. 30 Credit risk is the risk of a person defaulting on his or her obligations. A credit derivative contract seeks to address this type of risk. 31 In other words, by entering into a credit derivative contract, an entity having granted, say, a loan will seek to transfer the risk of the borrower failing to repay the loan to its derivative counterparty. As a result, the party that granted the loan (the 'protection buyer') will be able to get protection from the risk of default of the borrower; while the party to which the risk is transferred (the 'protection seller') will agree to assume it in the expectation of making a gain. Credit default swaps (CDSs) 28  are a type of credit derivatives. 32 They were traded widely after they first appeared in the 1990s. 33 By 2007, they had seen 'dramatic' growth rates. 34 They were often used in intricate securitisations where the risks associated with them were complex to assess. They became widely known for the part that they played in the financial crisis, after firms such as American International Group (AIG), which had sold billions in protection, suffered immense exposure as a protection seller. Ultimately, these exposures were too significant for AIG to be able to honour its obligations. 35 One aspect of derivatives trading that explains much of the thinking behind the EMIR arrangements is that derivative contracts have mostly been traded off-exchange or OTC. 36 OTC trading offers flexibility, which is useful in order to deal with 'bespoke' types of derivatives, as opposed to standardised derivative contracts. 37 But OTC markets can also lack transparency and they have proved vulnerable to risks building up unnoticed, ultimately posing a threat to financial stability. 38 The financial crisis brought all of the problems with OTC derivatives trading and derivative contracts into sharp relief. With hindsight, the problem drivers and interlinks seem obvious enough. The European Commission put it as follows when discussing the role played by the failed former investment banks, Bear Stearns and Lehman Brothers, as well as the bailed-out insurance firm AIG in the OTC derivatives market: Bear Ste[a]rns, Lehman Brothers and AIG were important players in the OTC derivatives market, either as dealers or users of OTC derivatives, or both. The trouble they experienced originated outside the OTC derivatives markets, it entered the derivatives market via the CDS written by these three institutions and, because of these institutions' central role in all OTC derivatives markets, it spread beyond CDSs and affected the world economy. The opaqueness of the market prevented, on the one hand, other market participants from knowing exactly 32 Commission Staff Working Paper on Derivatives Markets (n 29) para 3.1.1, which defines a CDS as a 'a contract between two counterparties under which the protection buyer will pay an annual fee (on a quarterly basis) to the protection seller until the maturity date of the contract or until a credit event occurs on the reference entity. In the latter case, the protection buyer must deliver bonds or loans of that reference entity for the amount of the protection (notional value of the contract) to the protection seller and receives the par value in return'. 33  what the exposures of their counterparties were to these three entities, which resulted in mistrust and in the sudden drying up of liquidity. On the other hand, it also prevented regulators from being able to identify early the risks building up in the system, the extent to which risks were being concentrated and consequently the effects that their default would have for financial stability. The light regulatory coverage of the market exacerbated this problem as supervisors did not have sufficient information. 39 In response, the EU legislature adopted EMIR in July 2012. By doing so, it sought to implement commitments agreed upon at the G20 summits of Pittsburgh and Toronto. 40 EMIR has essentially two limbs. The first concerns the mandatory clearing of eligible derivative contracts through a CCP. 41 Clearing through a CCP has become the weapon of choice for dealing with OTC derivatives. 42 It improves transparency and helps to manage risk. In the wider scheme of things, it also contributes to reducing the risk that a successive failure of interconnected market actors might ultimately jeopardise the whole financial system. A CCP is a legal person whose purpose is to stand in-between two parties to a contract, 'becoming the buyer to every seller and the seller to every buyer'. 43 If one of the original parties defaults, the CCP will ensure that the contractual obligation to the non-defaulting party can be honoured. 44 To deal with defaults, the CCP uses various strategies. These include netting, 45 collecting margin 46 and establishing a default fund. 47 The CCP's intervention will shift the counterparty risk, ie, the risk that one party  46 Margin can be defined as: 'An asset (or third-party commitment) that is accepted by a counterparty to ensure performance on potential obligations to it or cover market movements on unsettled transactions' (ibid 89). 47 A default fund is a 'fund composed of assets contributed by a CCP's participants that may be used by the CCP in certain circumstances to cover losses and liquidity pressures resulting from defaults by the CCP's participants' (ibid 89). fails to honour its obligations. But note that the risk will only shift; it will not vanish: 'CCPs do not make risk disappear: they reallocate it'. 48 In order to ensure that CCPs can play their role effectively and are 'safe and sound', 49 EMIR put in place harmonised rules for CCPs. They include organisational and conduct of business requirements, prudential obligations as well as provisions on the establishment of a mutualised default fund to deal with defaults. 50 However, not all derivative contracts can be cleared through a CCP. Some derivatives might, for example, be too bespoke and not sufficiently standardised to be suitable for CCP clearing. 51 Thus, EMIR establishes criteria and arrangements for determining whether a class of OTC derivatives is eligible for clearing through a CCP. 52 Moreover, for derivative contracts that do not meet the predetermined eligibility criteria and are therefore cleared on a bilateral basis, EMIR requires various procedures and arrangements to be put in place in order to deal with risk. 53 The obligation to report trades to trade repositories (TRs) is the second limb of EMIR. I will be brief with respect to TRs, given that they are not the subject matter of this chapter. The purpose of a TR is to collect information. EMIR defines it as a 'legal person that centrally collects and maintains the records of derivatives'. 54 As with CCP clearing, TR reporting has found international support. 55 The thinking here is straightforward: in an OTC derivative market which lacks transparency, mandating TR reporting promises to improve market transparency and decrease systemic risk. 56 To ensure that EMIR's new arrangements are properly applied, the EU legislature made provision for CCPs and TRs to be subject to authorisation and registration (respectively), as well as proper oversight. 57 The rest of the chapter will look at the arrangements that were adopted for the authorisation of CCPs, but mainly from the perspective of internal market governance. As far as TRs are concerned, it is worth noting that unlike CCPs, they are registered and supervised by ESMA rather than by competent 48 Pirrong (n 41) 10. 49 EMIR rec (49). 50 Note that there is a variety of other means and strategies that a CCP uses in order to deal with defaults. For details, see Pirrong (n 41) 6-10. 51 European Commission, 'Ensuring Efficient, Safe and Sound Derivatives Markets: Future Policy Actions' (Communication) COM(2009) 563 final, 5. There might be other reasons: for example, they might lack sufficient liquidity to be suitable for clearing (ibid). Note, however, that standardisation is not necessarily a prerequisite for central clearing (HL Report on Derivative Markets (n 29) 31). For a critical discussion, see Braithwaite (n 41) 106-09. 52 EMIR rec (16); arts 4 and 5. 53 Article 11. 54 Article 2(2). 55 See, eg, the G20 Pittsburgh Summit communiqué, stating that 'OTC derivative contracts should be reported to trade repositories'. 56 HL Report on Derivative Markets (n 29) 23. 57 Articles 14 and 55.
(national) authorities. 58 Yet, in both cases-that is, for CCP authorisation or TR registration-the decision to authorise or to register will be effective throughout the EU. 59

B. The Building Blocks of HCC-C
This sub-section will examine the building blocks of HCC-C by turning for inspiration to EMIR and its requirements regarding CCP approval. I will begin by highlighting the core aspect of this procedure-that is, decision making with consent-after which I will discuss the role of supervisory colleges and the role of ESMA under HCC-C.

i. Decision Making with Consent
HCC-C, as I understand it here, can be described as a procedure under which a home authority must share some of its power to make a decision with other national authorities which have, albeit to a lesser extent, a say over the decision of the former. I will describe such a say as given expression by some form of consent procedure. But I will use the term 'consent' rather loosely here for reasons that will become apparent in a moment. For now, the point is that because of the consent aspect, HCC-C is different from ordinary home-host arrangements and especially the home country control principle as we find it in many fields. Under the latter, the home authority and, albeit to a much lesser extent, the host authorities make decisions unilaterally (ie, independently) in their reserved, or possibly shared, areas of competences. 60 Likewise, home country control with consent is different from joint decision making. Recall, for instance, the provisions on consolidated supervision, mentioned earlier in relation to the CRD. Although joint decision making under the CRD involved other competent authorities, the decision-making process, if unsuccessful, culminated-depending on the competence area-either in the consolidating supervisor taking the decision, irrespective of the extent of opposition by host authorities, or national competent authorities deciding a given matter separately at their competence level which effectively reflected a 'host' state-based approach. 61 58 Article 55(1). The only other field in which ESMA has such powers concerns credit rating agencies. See P Schammo, 'The European Securities and Markets Authority: Lifting the Veil on the Allocation of Powers' (2011) 48 CML Rev 1911-46. 59 Articles 14(2) and 55(3). 60 For the avoidance of doubt, the home authority's unilateral decision will produce effects beyond its territorial boundaries as a result of the principle of mutual recognition. This is not the case for host state authorities. 61 Article 129(2); art 129(3), para 5 of Directive 2006/48/EC. Note that I use the term 'host' in a broader sense here (see n 10 for details). HCC-C is also different from the type of collective decision making that is characteristic of the decision making of the ESAs. The latter are EU bodies and are meant to act in the interests of the EU. In our case, the forum of decision making is national and involves colleges of supervisors, to which I will return below. 62 One important point about HCC-C, as I define it here, is that by referring to the requirement of consent, I do not mean to refer to, or even imply, the exercise of equal powers between relevant authorities. Even though there is room for variation (just as with ordinary home-host arrangements), the requirement on involving other actors in the decision making of the home authority does not go as far as changing the decision-making arrangements into a more finely balanced system of 'shared country control'. Under such a procedure, one would expect all relevant authorities to become intrinsically involved in the assessment and control of a market actor irrespective of territorial links. Any reference to home country control would as a result be mistaken. Under the arrangement that is presented here, the decision-making arrangements continue to be biased towards a specific Member State authority. Recall that such a bias is the defining feature of home country control: the distribution of powers between competent authorities will by definition be biased towards the authority of the home Member State of a market actor.
To be sure, generally one can imagine various ways in which a host state authority can hope to have a say on matters that are under the competence of a home competent authority. For the present purposes, they can be termed soft say, soft legal say, hard operational say and hard decisional say. Soft say describes ways to influence home state decision making by relying on argument and persuasion. As a soft strategy, it has obvious limitations. It does not create legal obligations and is simply an attempt to influence the home supervisor. Soft legal say, as the second type of say, translates into the right for a host authority to have its views considered. To put it otherwise, this type of say places a legal obligation on its addressee (the home state authority) to consider the views expressed by other actors. But this say is 'soft' in the sense that it only gives rise to an obligation to give due consideration to the views expressed by others as part of the process of making a decision. 63 The addressee is not bound to uphold these views. As such, the obligation is more constraining than, for example, a mere generic obligation to cooperate. But it does not go as far as placing an obligation on its addressee to reach a specific outcome. 64 Hard say can be of an operational type or of a decisional type. Hard operational say includes a rich category of requirements and obligations which are imposed on competent authorities to provide their counterparts with assistance or specific information as part of the operation of a regulatory regime. Finally, hard decisional say is the type of arrangement that is contemplated here. It allows host competent authorities to get involved in the decision making of a home competent authority. The thrust of decisional hard say is that, depending on matters such as voting modalities, it might ultimately allow host authorities to prevent the home competent authority from approving a market actor or market activities.
It is now worth turning to EMIR, for it provides for our purposes a template, albeit still a rather embryonic one, for the arrangement described above. Specifically, it is worth considering the provisions governing the ex ante authorisation of CCPs. 65 The latter is a sine qua non under EMIR. This is because OTC derivative contracts that require clearing must be cleared by a CCP which has been granted authorisation. 66 Once granted, such approval will be effective throughout the internal market. 67 For the present purposes, the requirements on CCP approval are interesting for two reasons. First, they testify to a 'home' state bias-that is, towards the CCP's competent authority which is the authority of the Member State in which the CCP is established 68 or, for our purposes, the CCP's home competent authority. 69 It is the latter which is meant to authorise a CCP, but second-and this is the crucial point-it must share some of its say over this decision with other actors. 70 These other actors are part of a college of supervisors (together with the CCP's competent authority) and are meant to issue an opinion on whether the CCP satisfies the relevant regulatory 64 As a consequence, verifying that the obligation has been discharged will in practice raise concerns. To address these concerns, this type of say might be coupled with a type of explanatory accountability that requires its addressee to explain its courses of action in light of the views expressed by other actors. EMIR includes such an obligation (art 17(4)). At the outset, it requires the CCP's competent authority to 'duly consider' the opinion reached by the college of supervisors (art 17(4)). It adds that where the CCP's competent authority disagrees with the college which has reached a positive opinion on the approval of a CCP, the former should give 'full reasons' in its decision and explain 'any significant deviation' from the college opinion (ibid). 65 Articles 14-20. Note that the same procedure applies where a CCP wishes to extend its activities or services. See art 15. 66 Article 4(3). Note that for third-country CCPs, EMIR states that a third-country CCP can provide clearing services in the EU if it is recognised by ESMA (art 25). 67 Article 14(2). 68 Article 14(1). 69 As mentioned earlier, EMIR does not as such use the home-host terminology, but for the sake of convenience, I will use it at times. 70 Article 17.
requirements. 71 This opinion may take the form of a joint opinion or, in its absence, a majority opinion. 72 Crucially, if all college members 'excluding the authorities of the Member State where the CCP is established' come to a 'joint opinion by mutual agreement' that the CCP should not be authorised, the CCP's competent authority cannot approve it. 73 Hence, if college members, minus the CCP's ('home') competent authority, refuse consent, 74 the CCP's competent authority is prevented from authorising it. 75 As a result, the CCP will effectively be unable to exercise the clearing activities that EMIR regulates, be it domestically or, via mutual recognition, cross-border.
To be sure, the threshold of 'unanimity minus one' 76 is a high threshold to meet. But it is nevertheless sufficient to alter a fundamental aspect of home country control as we have seen it in many sectors of activity: the fact that under home country control, the home competent authority makes decisions unilaterally-that is, at the exclusion of other Member State authorities-in its designated areas of competence. 77 The point is all the more remarkable given that in the securities markets, ex ante approval of a market actor or of market activities has traditionally been a reserved competence area of home state authorities. 78 It is worth noting, however, that EMIR does not go as far as empowering college members to overrule the CCP's competent authority if the latter decides not to authorise a CCP. In other words, a positive opinion of college 71 Article 19(1). 72 Ibid. 73 Article 17(4). 74 Or, perhaps more accurately, if they decide to dissent. 75 Note that according to art 17(4), such a decision must be motivated in writing and disclose why members of the college consider that obligations under EMIR or generally under EU law have not been met. This requirement will obviously matter in order to determine whether the reasons for disagreeing with the CCP's competent authority are legitimate or not. 76 Note that, strictly speaking, the terminology of 'unanimity minus one', although often used in order to describe the voting requirements of EMIR, is somewhat misleading for the reason that more than one authority from the 'home' Member State might be involved. The reference to 'one' must therefore be understood as encompassing any number of authorities from the CCP's 'home' Member State which might have voting rights. This fact is also reflected in art 17(4), which speaks of 'authorities from the Member State where the CCP is established'. 77 To be sure, the fact that the ESAs were vested with intervention powers (see below for details) already somewhat erodes home country control. The exercise of such intervention powers is also foreseen in EMIR. Additional conditions and requirements will, however, apply. In any event, the present scenario is different, given that it concerns national authorities acting at the national level. Moreover, the use of intervention powers is meant to be exceptional. Under EMIR, the normal course of action is for college members to be involved in the decision making of the CCP's competent authorities. 78  members cannot have the effect of requiring the CCP's competent authority to approve the CCP if this goes against the judgment of the latter authority. In such a case, the final decision is left to the CCP's competent authority. 79 The reason for restricting the say of college members in these circumstances appears obvious. It has to do with the potential financial implications which a Member State may have to bear in a case where a CCP were to fail: given that the Member State in which the CCP is established would be the first state to suffer the consequences of a failure, it seems natural to reserve the decision to refuse approval to the competent authority of that Member State (ie, the CCP's competent authority).
Besides the above voting arrangements, one can think of another reason why the procedure, which governs the say of college members, might be seen as somewhat embryonic. Indeed, it is apparent that under EMIR, CCP approval is subject to the absence of a negative opinion of the college. This opinion, it is recalled, must reflect a certain degree of opposition (or dissent), as expressed by the requirements of 'unanimity minus one'. But the regulation does not as such require a positive opinion from the college-consent, stricto sensu-for the CCP's competent authority to go ahead. One could therefore say that the procedure is not yet an (active) consent procedure; instead, it produces binding effects if college members express dissent. In any event, for our purposes, this is not problematic as it is not suggested that EMIR enacts HCC-C. All that is suggested is that EMIR is a starting point to think of the building blocks of HCC-C and its possible arrangements.

ii. Colleges of Supervisors
It is plain that under HCC-C, proper cooperation and information sharing between actors is especially important. It requires policy makers to have regard to deeply practical considerations: for example, how to ensure that all authorities, which have some stake in the authorisation process of a market actor, are properly involved; and how to make sure that all necessary information is available to all relevant actors. Hence, the second building block of the procedure is a college of supervisors. Here too, the EMIR provisions provide a useful illustration of the possible requirements and challenges.
To begin with, it is worth noting that the concept of supervisory colleges was not invented by EMIR. Colleges have been in place elsewhere for many years-for the supervision of cross-border banking groups, for example. However, EMIR turns to colleges for the supervision of a single 79 The CCP's home competent authority will, however, be subject to a form of explanatory accountability (art 17(4); see also above, n 64). Note that the decision to withdraw an authorisation is also a matter for the CCP's competent authority which must only consult college members (save in the case of urgency). See art 20. actor, that is, a CCP. Admittedly, describing what colleges actually are is not without its problems. Colleges are not legal bodies. Nor are they, for that matter, European bodies such as the ESAs which are subject to EU constitutional law. Moreover, colleges have no legal personality or indeed physical existence. They have no representative office. Nor do they have a separate secretariat. The college itself has no decision-making powers. 80 The decision-making powers are those of the members of the college. The latter are national actors. What is more, the decisions that college members adopt must be translated into national decisions for them to be legally binding on market actors. Also worth noting is that, unlike the ESAs, colleges are not subject to prescribed normative orientations such as to act in the interests of the EU. 81 Given these various limitations, it is perhaps not surprising that colleges have been described in various ways, that is, simply as a meeting of individual supervisors, a 'supervisory process', 82 or perhaps more generously as an 'instrument for stronger coordination and cooperation' 83 or a 'forum of cooperation'. 84 Although colleges of supervisors lack 'backbone', they can nevertheless be a pragmatic means of organising multilateral cooperation. 85 They are meant to be adaptable and are also meant to offer flexibility, 86 as college members can decide upon their structure and organisation. Hence, the challenge with colleges is to find the appropriate balance between flexibility and prescription. In the case of EMIR, the regulation seeks to achieve this by mandating certain rules, but leaving others to be decided by college members.
EMIR provides a legislative basis for the establishment of colleges and also specifies their tasks and regulates their membership. 87 It is plain that under these provisions, college membership is restricted and that, as a result, not all authorities which might have an interest in the activities of 80  the CCP will be able to get involved. 88 However, EMIR adds that these competent authorities can ask the college for information if it is necessary for them to perform their supervisory tasks. 89 EMIR also deals with the voting modalities in colleges. Some of the relevant rules have been highlighted above. They concern the requirements that determine the say of college members over the decision to authorise a CCP. 90 Similarly, the regulation specifies the voting requirements that must be met before a college member, disgruntled with the decision to authorise a CCP, can refer the matter to ESMA for binding mediation. 91 Besides specifying voting requirements, EMIR also sets out rules that make it possible to determine the number of votes of college members. The rules are especially important because possibly more than one authority from a single Member State might be involved in a college. Thus, EMIR restricts the total number of votes per Member State. It provides that for colleges 'up to and including 12 members', up to two college members that are of the same Member State can vote, and each voting member can have one vote. 92 Beyond 12 members, EMIR provides for up to three college members that are part of the same Member State to vote, and each voting member can have one vote. 93 It also specifies that ESMA cannot vote on college opinions. 94 This rule makes good sense, given that ESMA has a role to play as 'mediator' under the rules which are examined below. As noted above, other matters regarding the setting-up and working of colleges are to be determined by the college itself. EMIR makes this explicit too, as it specifies that the 'establishment and functioning' of a college are to be 'based on a written agreement between all its members'. 95   for the working of the college which also encompasses the specifics of the voting procedures. 96 Hence, EMIR sets up a two-tier system in relation to colleges: some matters are mandated, while others can be agreed upon by college members.

C. ESMA as Arbiter of Disagreements
Like the previous building block, the final building block of HCC-C is not exclusive to this procedure. Rather, it reflects the fact that the EU is increasingly intent on intervening in matters concerning financial supervision, an area that was traditionally left to Member States. Thus, the third building block concerns the power of ESMA to intervene in disagreements between members of the college. This power to address and possibly settle such disagreements with binding effect is among a range of intervention powers which have their legal bases in the regulations establishing the ESAs. 97 Together, these powers are the basis of an intervention-based system, that is, a type of hybrid system which vests the day-to-day supervisory competence in national authorities, but which allows the ESAs to intervene in: (i) the relationship between competent authorities; or (ii) the relationship between a competent authority and a market actor. 98 The aim is to combine the strength of supervision at the grassroots level with an intervention system that is effectively meant to ensure that the interests of the internal market, or the EU as a whole, are taken into account in a cross-border context. 99 Crucially, because of its hybrid nature, this intervention-based system can be combined with home country control or, as in the present case, with HCC-C. As far as home country control is concerned, examples of the new intervention powers can be found throughout securities markets legislation-for example, in the Prospectus Directive (PD) or the Markets in Financial Instruments Directive (MiFID). 100 As far as HCC-C is concerned, OJ L331/84, arts 17-20. Note that according to art 19 of ESMA's founding regulation, for ESMA to exercise its power to mediate and settle disagreements, provision must be made for the exercise of this power in sectoral acts (such as EMIR). The use of the power in a college setting is envisaged in art 21(4) of the founding regulation. It is worth noting that the usual safeguards found in ESMA's founding regulation on the use of dispute settlement will also apply. Thus, the fiscal responsibility clause, which is provided for in ESMA's founding regulation, will apply (see art 38 of ESMA's founding regulation). 98 For details, see Schammo (n 12). 99 Ibid. 100  EMIR makes (as noted) provision for ESMA to get involved in the case of a disagreement between college members. 101 Specifically, it might be involved because the CCP's competent authority wants to proceed with authorising a CCP in the face of opposition from college members. This opposition might be widespread, but not sufficient to allow college members to simply block a decision. Alternatively, ESMA might also get involved because college members have succeeded in blocking a decision to authorise a CCP and the CCP's competent authority responds by asking ESMA to intervene in the matter. 102 For the dispute settlement powers to be triggered by college members unhappy with the decision to authorise a CCP, however, EMIR requires a certain voting threshold to be crossed. Specifically, it states that 'a majority of two-thirds of the college' must have 'expressed a negative opinion'. In this case, EMIR states further that 'any of the competent authorities concerned' can refer the matter to ESMA 'based on that majority of two-thirds of the college'. 103 These voting requirements are demanding and appear to go beyond the minimum required for mediation under ESMA's founding regulation. 104 The EU legislature seems to recognise the point implicitly. Recital (56) states that EMIR's provisions, especially those dealing with the voting requirements for referrals to ESMA, are not meant to set a precedent.

IV. HCC-C: A NEW PARADIGM FOR THE INTERNAL MARKET?
Section III described the building blocks of HCC-C. It was argued that a central feature of this procedure is that the home authority must share its say over whether to authorise a market actor with other national authorities, without however that such 'shared say' means that all authorities have an 'equal say'. This section will single out this basic feature in order to investigate further whether it has anything to offer as a new paradigm for the internal market.
I will begin by looking at the problematic which underpins home-host relationships (section IV.A). This will take me back to issues that I discussed in section II. But hereinafter I will focus more on the theoretical and doctrinal underpinnings of the home-host problematic, which I have left aside 101 Article 17(4). It is worth noting that EMIR also makes provision for ESMA to exercise its powers to deal with breaches of EU law (see art 17(5) of EMIR and also art 17 of the ESMA founding regulation (n 97)). 102 Article 17(4). 103 Ibid. Note that the decision to refer must disclose in writing why college members think that relevant obligations under EMIR, or generally under EU law, have not been met. 104 Article 19(1). until now. Next I will reflect on the basic rationale of HCC-C: that of greater interference in home state decisions (section IV.B).

A. Pathology
At issue is the cooperative relationship between competent authorities and especially how to ensure cooperation and trust in the face of power asymmetries between them. These power asymmetries are profound under home country control, for the home supervisor concentrates decision-making powers over market actors or activities taking place domestically and abroad as a result of the application of the principle of mutual recognition. Host supervisors, on the other hand, are left with little room to resist the decisions taken by the former in its role as home authority and, if at all, can only make decisions whose reach is territorial . As such, the cooperative relationship between supervisors is not one of equals.
To be sure, the EU legislature has enacted numerous provisions which require competent authorities to cooperate with each other. 105 These provisions should prima facie prevent the home competent authority from withdrawing from the cooperative relationship. But a moment's thought will show why this is not so under the home country control principle. First of all, although the home authority and host authorities are formally under the same obligation to cooperate, generally worded obligations to cooperate often remain vague and unspecified, and their application remains difficult to monitor. Moreover, in the financial markets field, there appears to be little evidence that they have been effectively enforced. 106 Furthermore, even if more specific obligations were enacted (for example, an obligation to assist or provide information), it remains arguably difficult to make provision ex ante for all possible types of contingencies. Hence, the unilateral decision-making powers of the home supervisor will ensure that it will continue being able to effectively 'exit' the cooperative relationship. That is not so for host supervisors: for matters that fall under the competence of the home authority, host authorities will continue to be dependent on the cooperation of the former for the supervision and the proper application of rules and regulations in their territory, including for the proper provision of information.
The asymmetry between home and host authorities can be illustrated by one of the well-known episodes of the financial crisis: the rescue of Fortis Bank in 2008. Having highlighted that one of the problems of home-host relations in supervisory colleges is that host supervisors, unlike their home country counterparts, were not necessarily meant to have the 'full picture', Lannoo highlights the consequences of the asymmetry. He notes that in the case of Fortis Bank, the Belgian authorities only got in touch with the two main host countries 'after about 48 hours of discussions', 107 a delay that was widely judged to be inadequate. This delay does not necessarily illustrate an absence of cooperation, but, arguably, it shows a lack of quality in the cooperative behaviour of authorities.
The Court of Justice of the European Union has contributed to exacerbating the power asymmetries described above. It has repeatedly held that Member States must place trust in each other. 108 In Hedley Lomas, a case in which the UK refused to grant a licence which Hedley Lomas requested for the export of livestock to Spain, the Court held that 'Member States must rely on trust in each other to carry out inspections on their respective territories'. 109 The UK had refused to grant export licences on the grounds that it doubted that animals would be slaughtered in Spain in accordance with the requirements of a Council directive. 110 The Court's insistence on trust between Member States contributed to its conclusion that the UK was not entitled to withhold the export licence. This decision appears to have found its way into the banking, insurance and financial fields. Thus, the ESA's founding regulations declare that 'the parties to the ESFS shall cooperate with trust and full mutual respect, in particular in ensuring the flow of appropriate and reliable information between them'. 111 There are reasons to be critical of this case law. One such reason is that for the notion of trust to have any kind of meaningful denotation, it cannot be prescribed: 'I cannot will myself to believe that X is my friend, I can only believe that he is.' 112 What the Court then really appears to say, under the cover of the notion of mutual trust, is the following: Member States should not question each other's actions or, for that matter, their lack of actions outside the boundaries drawn by the Court of Justice and the EU legislature. The point about these boundaries is of course that they are narrow. This is reflected generally in the case law on factors that justify restrictions to the free movement. It is also reflected in the EU legislature's approach to precautionary measures as found in financial markets regulation. Under these provisions, host competent authorities can take precautionary action in relation to matters that are reserved to home authorities, but only after satisfying specified conditions and requirements. 113 Last but not least, it is reflected in the Court's case law on the actions that a Member State can take if it considers that another Member State has breached EU law. For example, it is well established that a Member State cannot take unilateral action of a 'corrective or protective' nature in order to deal with such a breach. 114 Overall, it is fair to say that the Court's approach will generally benefit the integrationist objectives of the EU, but it also implies that, generally, interference by host competent authorities in decisions or actions of home competent authorities will be seen as unsatisfactory. 115 As far as mutual trust is concerned, we are then essentially left with an empty and hollow concept of trust. But that is not the end of the story. Indeed, it is plain that while the Court's case law on prescribed trust is convenient from an integration point of view, it fundamentally eschews the real problematic of trust under home country control. This problematic is shaped by the fact that trust and power are involved concepts. 116 As Farrell notes, 'the degree to which one party trusts another may vary according to the power relations between them'. 117 According to the author, if power asymmetries become too significant, it will undermine trust and may indeed foster distrust. 118 The less powerful actor will have no grounds to believe that the more powerful actor will take its interests into account: If I am so much more powerful than you that I am no longer capable of giving credible commitments, then it follows that our relationship is insufficient to bind me to act in your interest. You will have no reason to trust me and in many 113  Moreover, as Farrell puts it, 'to the extent that power affects the possibility of trust and trustworthiness, it also may affect the kinds of cooperation that take place on the basis of trust and trustworthiness'. 120 Thus, in our context of home country control, Farrell's observations about power and trust suggest that the power asymmetries sketched out above, which are at the heart of home country control, will affect trust between competent authorities. As such, his observations underscore the point that power asymmetries may adversely impact on the quality of cooperation between authorities, notwithstanding the formal obligations to cooperate under EU law.
To be sure, the idea that trust might be at issue in interactions between composite actors such as competent authorities might cause some debate in the first place. To see why, it is necessary to step back for a moment and to reflect on the meaning of trust. 121 For influential authors such as Hardin, Cook and others, trust is best suited for interpersonal relations. 122 These authors rely on a specific description of trust, which is based on 'encapsulated interests'. According to this description, my trust will depend on whether you consider my interests in part as your own (that is, whether my interests will become encapsulated in your interest). 123 Specifically, my interests will be part of your interests 'just because they are my interests'. 124 The reason why a person might encapsulate another person's interests might differ. Most often it is because the trusted wishes the relationship with the other person to carry on. 125 For Hardin, trust is then a 'three-part relation': 'A trusts B to do, or with respect to, X'. 126 Moreover, it is said to be a 'cognitive notion, in the family of such notions as knowledge, belief, and the kind of judgment that might be called assessment ' It is obvious that the above definition is a specific definition of trust. 128 It is based on a certain vision of human behaviour. 129 Moreover, as noted, it generally conceives of trust as something which is best reserved to interpersonal relations. As a result, Hardin and others take the view that trusting governments or large institutions is almost impossible. 130 The reason why one cannot trust large institutions has to do with the extensive knowledge that is required for trust to develop. 131 For the authors, trust cannot develop absent knowledge about an actor's motivations; about him or her encapsulating our interests. 132 Given what we said about the concept of trust-that for Hardin and others, trust is something that should be reserved for relationships at the interpersonal level-one might conclude that the language of trust is simply out of place when discussing cooperation between competent authorities. That said, focusing narrowly on trust at the interpersonal level-in our case, between individual officials from competent authorities-will not allow us to gain much ground unless we elaborate on how trust and trustworthy behaviour might come to matter for the likely behaviour of the organisation as a whole, of which such officials are part. 133 In this context, it is worth observing that officials representing each authority are supposed to act according to expectations attached to their role and according to the objectives and interests of the authority as a whole. I will return to the point below, but for now, the more important point is that even if we step outside the realm of interpersonal relations and choose to no longer talk of trust, our conclusions appear to remain similar: the power asymmetries inherent in the relationship described above are likely to affect the assessment of the cooperative relationship and the judgment, albeit grounded in the abstract knowledge of the existence of the power asymmetries, about whether the home authority has incentives to take the interests of the host authorities into account. Whether we call the outcome a lack of trust or something else does not change the essential problematic: it might come to 128 See Cook, Hardin and Levi (n 116) 4, who note that: 'Ours is a relatively specific definition that imposes clear requirements on those we claim are trusting.' 129 There are other conceptions of trust. See also J Lewis and A Weigert, 'Trust as a Social Reality' (1985) 63 Social Forces 967, 972, who note that 'Trust in everyday life is a mix of feeling and rational thinking' (reference omitted). 130 Cook, Hardin and Levi (n 116) 4-5. 131 Ibid 105. 132 Ibid 8: 'Trust involves a genuine involvement between you and the trusted other and a specific, not abstract, assessment of that other's motivations toward you.' 133 See also Hardin (n 119) 200, who notes that: 'It may still be true that trust and trustworthiness are fundamentally important in making large-scale activities, and especially, large social institutions function. To show how they do this, however, requires substantial unpacking of the relationships within those institutions to understand how trust plays a role at the micro level. Trust is inherently a micro-level phenomenon.' affect cooperation or, perhaps better, the quality of cooperation between competent authorities.

B. HCC-C: A New Paradigm for the Internal Market?
If as argued above, power asymmetries might come to affect the cooperative relationship, one might consider that such asymmetries require action. The final part of this section will discuss whether the basic rationale of HCC-C has anything to offer in this context.
Traditionally, choices with respect to supervisory governance have been presented as based on one of three models: home country control, host country control or centralisation of supervisory competence at the EU level. A common feature of these different models is that decision making is essentially unilateral in the assigned areas of competence. In other words, within the relevant competence areas, decisions are taken independently either by home supervisors, host supervisors or a single agency. 134 In the wake of the financial crisis, we have, however, witnessed the adoption of intermediary arrangements. ESMA was vested with intervention powers and of course I discussed at length the powers of host supervisors under the new EMIR arrangements. 135 These two latter arrangements also share a common feature: decision making is no longer unilateral. Under both arrangements, greater involvement in the decision making of a competent authority is seen as justified, albeit under predefined conditions and/or requirements. In the case of ESMA, I used the term intervention to describe such involvement. 136 In the case of HCC-C, I will use the term interference. In the final part of this section, I will reflect on the implications of greater interference by host authorities in the decision making of home supervisors. My aim is not to endorse or reject the precise arrangements provided for under EMIR, but to engage in a richer and deeper reflection on the issue of interference-a concept which is at the heart of HCC-C-in line with the stated aim of assessing whether HCC-C could be a useful new paradigm for the internal market. This enquiry will take us back to the question of how to ensure trust and cooperation between competent authorities. The use of the term interference appears well chosen in this context, for it underlines that hitherto interference by host competent authorities has been resisted by the Court of Justice-at least if it fell outside the mostly narrow boundaries drawn by the Court. Hereinafter, my starting point is that it is possible 134 As mentioned in the introduction, I do not discuss the proposed Eurozone banking union in this chapter. Suffice it to note that under the current proposals, decision making is supposed to be located with the ECB. 135 On ESMA's intervention powers, see Schammo (n 12). 136 Ibid.
to make a case in favour of interference based on arguments about trust and cooperation. I will describe such virtuous interference as constructive interference. But on closer inspection, to make interference work, to make it potentially constructive, various stumbling blocks must be successfully addressed. Moreover, constructive interference as it is understood here must be based on a multi-pronged strategy which seeks to address power asymmetries, transparency and mutual trust. As far as trust is concerned, I will reserve it, in line with authors such as Hardin, to interpersonal relations. However, I will work on the basis that for trust to grow among individuals, the right institutional conditions must be put in place. 137 Moreover, to make trust a useful concept for our purposes, it will be necessary to elaborate on the relationship between interpersonal trust and the behaviour of composite actors such as competent authorities.

i. Addressing the Power Asymmetries Which Characterise Home Country Control
It is useful to begin by briefly reminding ourselves of the line of reasoning of section IV. I began by highlighting the power imbalances that underpin home-host relationships and that are rooted in the institutional rules that structure the interactions between supervisory authorities. I pointed out that they might come to affect the quality of cooperation between authorities. I next asked whether HCC-C could be a response to such problems. Rather than examine each and every aspect of HCC-C, I chose to focus attention holistically on the basic rationale of HCC-C, which is that of interference. I suggested that for interference to be useful, it ought to be based on a three-pronged strategy. The first aspect of this strategy, which is under investigation here, is (obviously) to target the power asymmetries, which are characteristic of home country control. This is achieved by fostering greater interdependence between supervisors.
a. The Need for Interdependence Power asymmetries are of course not something exclusive to the type of relationship that is under investigation here. Such asymmetries characterise many types of social exchanges. In this context, Emerson's work on powerdependence theory provides an influential account on how parties can deal with them. 138 For Emerson, power is a property of a relation rather than a feature of an actor. 139 According to the author, it follows that 'to say that "X has power" is vacant, unless we specify "over whom"'. 140 Thus, for Emerson, the notion of dependence is crucial for determining power. Specifically, power is seen as rooted 'implicitly in the other's dependency '. 141 Given that the dependence of one party ('B') is the basis of the power of another ('A'), Emerson proposes that dependence be defined as 'the amount of resistance on the part of B which can be potentially overcome by A'. 142 Emerson's work has also been used in the literature on trust and cooperation. Cook et al take it as a starting point. 143 They go on to note that the absence of interdependence between parties means that the more powerful actor has no reason to consider the interests of the less powerful. 144 This basic insight can be transferred into our context. Specifically, it might be suggested, based on the above findings, that greater interdependence between competent authorities can be useful for dealing with power asymmetries. In our context, interdependence means giving a say to host supervisors, but not-and this is a crucial point-within the host country control paradigm. It is worth briefly elaborating on this point.
Host country control can be thought of as a conventional line of response to concerns about home country control. Calls in favour of greater host country control were made in the wake of the financial crisis. 145 It is plain that a policy shift towards host country control would have as a consequence to rebalance the power relations between competent authorities dramatically. Host country control thus reflects a vision of an internal market that is very different from the one under home country control. But it is also apparent that host country control is fundamentally an inadequate concept for our purposes, as it does not create the conditions for greater interdependence between parties. This is because under host state control, each competent authority is meant to retain territorial jurisdiction. As a result, each party to the cooperative relationship can effectively withdraw from it at no, or low, cost. It promotes neither dependence nor interdependence. For interdependence between home and host competent authority to be achieved, the host authority must be allowed a say in the decision making of the home authority, 138  as a sort of quid pro quo for mutual recognition. 146 This is the logic that EMIR appears to endorse and which I described earlier when looking at the building blocks of HCC-C. Greater interdependence could plausibly have two benefits: first, home competent authorities would have to take the interests of host state authorities more seriously; and, second, it might help to put in place institutional conditions under which trust is more likely to emerge at the interpersonal level. If those benefits were to materialise, the quality of cooperation could in theory be improved.
That said, interdependence has its complications. At the outset, it is worth bearing in mind that cooperation is not an end in itself. It is a means to an end, which is to ensure the proper application of rules and regulations. Ultimately, the aim is to make certain that the internal market can work effectively. Seen in this light, interference raises at least two noteworthy issues. First of all, one must be mindful of the fact that greater interdependence can complicate the application of secondary rules and more generally make the operation of the internal market more complex. At the supervisory level, there is a premium on the capacity to respond quickly and swiftly to problems with the application of rules and regulations. Thus, in comparison to unilateral decision making, collective decision making will make the decision process much more complex. As such, it might also contribute to undermining the effectiveness of EU law. What is more, from a market actor's perspective, a moment's thought will make it plain that a model based on interference might be a worse paradigm than one which is based on host country control. This is because under the former, disagreement between supervisory authorities on, say, ex ante authorisation of a market actor could easily become an absolute bar to accessing regulated markets or regulated activities. 147 On the other hand, under the host country control paradigm, a prohibition to exercise regulated activities or access a regulated market is only relative: it only affects a market actor in a given host Member State.
A second issue concerns the question of how much power should be shared. It is obvious that to make interdependence meaningful, more than just marginal powers must be shared. 148 This is also the case from a trust point of view. 149 But determining what exactly amounts to marginal powers 146 Mutual recognition is of course the instrument that ensures that home supervisors have a say over host competent authorities. 147 If the decision to approve a market actor were taken by a central authority (eg, a single securities authority), the effects of refusing approval would of course be the same. 148 As Farrell notes, 'there is a point at which asymmetries are such that it is impossible for the more powerful actor to give credible commitments to the weaker. At this point, disparities of power prevent trust from arising and make distrust the likely outcome': H Farrell (n 116) 101. 149 Ibid.
is unlikely to be straightforward. 150 There is at least a bottom line: as long as one does not wish to abandon the home country aspect of the paradigm, there must be some limit on how much power ought to be transferred. On reflection, however, the above issues deserve to be put into perspective. Although home country control is a much neater paradigm, it has shown important limitations in the face of greater integration in times of crisis. To respond adequately to problems of greater integration, it seems inevitable that more intricate solutions will need to be considered-as long as the treaties are, as a matter of political reality, not open for amendment, that is. Thus, within the current treaty constraints, these more intricate solutions will require greater elaboration. Limitations will need to be defined and conditions and requirements will need to be specified.

b. Conditions and Requirements
It appears that in assessing whether a greater level of interdependence between competent authorities can be justified, consideration must at least be given to the following questions.
How 'pan-European' are the market activities? It is plain that the type of interdependence that is described here can really only be suitable for a subset of activities. Where the activities in question do not have a significant cross-border dimension, the case for host state intervention will not be apparent. Constructive interference, as it is conceived here, is inextricably linked to the intensity of the cross-border contacts and, as such, to the need for cooperation. HCC-C will become a redundant concept if effective cooperation on supervisory matters is not a prerequisite. It is also apparent that for the reasons mentioned above, (i) interference in approval processes of market actors (or market activities) and (ii) interference in ongoing supervisory matters should be treated separately. However, within a subset of cases, HCC-C may well be an alternative to a further transfer of supervisory powers to the EU.
How does fiscal responsibility affect home and host states? Since the financial crisis, the issue of fiscal responsibility has assumed much importance in discussions on the right distribution of powers. 151 These discussions have mainly been about the vertical distribution of supervisory powers and 150 The allocation of power in exchange networks has been a major topic in the literature. See, eg, K Cook and T Yamagishi, 'Power in Exchange Networks: A Power-Dependence Formulation' (1992) 14 Social Networks 245 (and references therein). 151 For references, see eg, ESMA's founding regulation (n 97) especially recs (5) and (50), art 38; rec (52) of EMIR. It is worth noting that in the banking context and especially in recent discussions on a proposed banking union, the fiscal responsibility problematic has taken a different form from the one that is presented here. In these discussions, the question is about whether the fiscal responsibility of Member States should be pooled together. This would require making provision for some form of financial burden sharing between Member States. The question has proved to be a major bone of contention between Member States. especially about the powers of the ESAs. The fact that Member States such as the UK have, for example, been reluctant to vest ESMA with day-to-day supervisory powers must also be seen in the light of the fiscal responsibility issue. Specifically, the argument goes as follows: as long as Member States bear the fiscal responsibility of a default of a market actor, they should be able to decide on supervisory matters. The point proved prominent in discussions on the supervision of CCPs. Given that the cost of a failing CCP might be substantial, the argument was clearly hard to ignore.
Fiscal responsibility, however, also affects the relationship between competent authorities. It does so in two important ways. First of all, it might come to affect the powers that a home authority is willing to share with host competent authorities. As mentioned previously, the fact that host competent authorities cannot require the CCP's competent authority to approve a CCP if the latter authority does not wish to approve it is best seen in the light of the fiscal responsibility problematic. In the absence of burdensharing agreements, this makes sense. But, importantly, fiscal responsibility can also be an argument for justifying host state interference in the first place. Admittedly, in the case of EMIR, the fiscal responsibility argument is most clearly articulated in order to justify oversight by the CCP's competent authority. Thus, recital (52) states that the CCP should be authorised and supervised by the Member State in which the CCP is established on the grounds that fiscal responsibility 'may lie predominantly' with this Member State. But in the absence of an agreement on cross-border burden sharing between Member States, it is hard to see how, in a well-integrated market which is based on home country control, fiscal responsibility can necessarily be restricted to a single Member State. Indeed, if one follows the reasoning underpinning the fiscal responsibility argument, it is a short step to making it an argument for greater host state interference on the grounds that Member States other than the state in which the market actor is established might ultimately suffer the financial burden of the default of the latter. 152 While EMIR is not entirely explicit on the matter, there is little doubt that it seeks to make the point when noting that: [S]ince a CCP's clearing members may be established in different Member States and they will be the first to be impacted by the CCP's default, it is imperative that all relevant competent authorities and ESMA be involved in the authorisation and supervisory process. 153 152 Think as an example of the financial crisis in this context and how the interests (and ultimately the fate) of major investment banks became entangled. For details, see the Financial Crisis Inquiry Report (n 35).
At any rate, the Commission's Impact Assessment on EMIR drives the point eminently home: [I]n a cross-border context, where a CCP may have members coming from multiple Member States … other Member States may need to use their fiscal resources were a CCP to fail, for example to provide support to their own banks that are members of the failed CCP. 154 Hence, while fiscal responsibility can be a potent argument against ESA intervention in home state decision making, or indeed against the transfer of day-to-day supervisory powers to the ESAs, it is a much more complicated argument to wield in a purely horizontal context. What are potential conflicts of interests in the field under consideration? Admittedly, in their day-to-day activities, competent authorities will often deal with technical matters for which expert understanding is valued and which might not as such be contentious. 155 On the other hand, however, competent authorities are unlikely to be perfectly shielded from conflict of interests that can affect relations between Member States. Moreover, they are unlikely to be fully shielded from considerations linked to the national interest or indeed to the competitiveness of national markets. It is worth bearing in mind in this context that competent authorities have national accountability lines, notwithstanding the fact that they are strictly speaking independent actors. Moreover, there has long been a degree of competition between financial places in Europe and national authorities have at least in the past played a part in regulatory competition between these places. 156 Thus, the fact that host competent authorities are given a say in the decision making of a home competent authority raises the possibility of coalition building when salient issues are involved. 157 The fact that Eurozone Member States increasingly have a distinct agenda from the rest of the EU may become a possible consideration in this context. 158 Indeed, EMIR is a telling example of the tensions which the divide between the EU and the Eurozone can create. During the EMIR negotiations, the UK as a non-Eurozone country with a significant post-trading industry was concerned about the possible impact of the ECB's so-called location policy. Under the latter, CCPs which clear significant amounts of euro-denominated derivatives are meant to be located in the Eurozone area. 159 Specifically, the UK was concerned that the EMIR arrangements might ultimately be used by Eurozone members to implement the ECB's policy. 160 The UK finally secured provisions in EMIR that underline the basic principle of non-discrimination and make it clear that college members cannot discriminate against other Member States as a place for clearing services 'in any currency'. 161 Who should determine the level of interference? Arguably, if one hopes to secure the best possible institutional conditions for interpersonal trust to emerge, simply prescribing greater interference might not be the most effective solution. Indeed, at least from a trust-building point of view, there is plausibly more to be gained if one party has the freedom to voluntarily commit to certain obligations. Such commitment by one party might prove to be a potent signal to others. It might signal that the former is serious about the interests of the latter. Moreover, such behaviour is easy to observe. In the case of HCC-C, this suggests that it would in fact be more effective to let home supervisors define voting modalities unilaterally. Quasi-unanimity would then merely be the default arrangement. Of course, by giving a party the freedom to commit, it might well be that this party decides not to commit in any meaningful way. Hence, such freedom might ultimately only exacerbate distrust. But that is arguably the price to pay for securing trust. At any rate, safeguards could always remain in place. I think, for instance, of mediation by ESMA which could serve as a safety net. Of course, it presupposes that ESMA plays its part in monitoring colleges effectively.
Is there any evidence of a lack of cooperation or trust which justifies action? The financial crisis offers examples of cooperation failures, but one should not be oblivious to the fact that arguments about authorities failing to cooperate or to trust each other are wielded frequently. Such arguments might offer policy actors opportunities to frame the political discourse in ways that favour their preferred course of action. Hence, it is essential to carefully examine the different issues and forces at work in each case. Broad statements, which are not supported by evidence of breakdowns in cooperation, are unlikely to be very useful. Moreover, it is equally important that the impact of greater interdependence be monitored closely, for monitoring is crucial to understand how the behaviour of competent authorities will be affected.

ii. Transparency and Interpersonal Trust
Transparency remains an important consideration in all kinds of cooperative relations, including in relations in which trust is at issue. It can be addressed in various ways-for example, by adopting arrangements that are meant to ensure assistance and exchange of information at the operational level. Moreover, at the decisional level, EMIR includes provisions that give host supervisors a sort of soft legal say which is coupled with a form of explanatory accountability. 162 Generally speaking, it is fair to say that transparency is something that has already received a good deal of attention in EU legislation. But to make interference work, transparency must continue being an important consideration for all the parties involved. This is also the case from a trust point of view.
Perhaps a less obvious point about trust building concerns the organisation of colleges and especially the rules that govern their functioning and composition. Until now, one could plausibly claim that interdependence improves the odds of a home authority considering the interests of host state authorities. But the former might do so for various reasons. It is at the interpersonal level where a home supervisor might ultimately do so for reasons of trust. To be more specific, it is at this level where a home supervisor might consider the host supervisor's interests 'as partly his or her own interests', just because, as Hardin put it, they are the latter's interests. 163 Hence, it is at the interpersonal level that trust might ultimately become the proverbial 'glue' that binds supervisors together.
If, as it was just claimed, trust may develop at the interpersonal level, supervisory colleges, as places of interaction and exchange between officials, might just be the missing link to ensure that trust can emerge and that individuals can gain the knowledge that is required for trust to develop. But there are complications. First of all, it must be borne in mind that interactions which are infrequent or marginal are unlikely to create the conditions 162 Under the relevant provisions, the CCP's competent authority must provide 'full reasons' and explain 'any significant deviation' from the opinion of the college, where the latter is positive about authorising the CCP, but the CCP's competent authority disagrees (art 17(4)). A similar obligation to state reasons also exists for college members if the college blocks the decision of the CCP's competent authority to authorise a CCP or if it refers the matter to ESMA for mediation. Specifically, in these cases, 'full and detailed reasons' must be provided with respect to why college members believe that the obligations of EMIR, or of EU law, have not been met (art 17(4)). 163 Hardin (n 123) 19: 'My trust turns, however, not directly on the Trusted's interests per se, but on whether my own interests are encapsulated in the interests of the Trusted, that is, on whether the Trusted counts my interests as partly his or her own interests just because they are my interests '. under which interpersonal trust can develop. Furthermore, even if trust develops between individual officials, it might not materialise in meaningful action at the level of the competent authority. Thus, as mentioned above, it will be necessary to ensure that relationships of trust between individual officials will come to matter for the behaviour of competent authorities as a whole. To address these concerns, one can think of two lines of action. First of all, to make interactions in colleges meaningful, college meetings must be sufficiently frequent and regular. Moreover, college attendance must be stable: the same officials must attend college meetings over a sufficiently long period of time. If officials do not meet regularly or if the composition of meetings tends to vary, it is hard to see how relationships of trust can emerge. Moreover, in order to ensure that discussions within colleges lead to effective action at the level of competent authorities, officials who attend meetings must be of a sufficient level of seniority. The point is not merely academic. In the past, issues have arisen. The operation of supervisory colleges in the banking sector provides an illustration. The British Banking Association (BBA) thus noted back in 2008 that colleges had 'been found to dissolve, too often, into education sessions for junior supervisors or training seminars for smaller jurisdictions'. 164 The BBA noted further that competent authorities 'should commit to sending personnel with the authority to make binding decisions', adding that a 'key weakness' was 'the lack of certainty as to whether agreements reached during the course of a college will be implemented'. 165 Such criticisms do not seem to have gone unnoticed. In its 2010 guidelines on the operational functioning of supervisory colleges, the Committee of European Banking Supervisors (CEBS), which has now been replaced by the European Banking Authority (EBA), stated that colleges should be places of dialogue with 'senior representatives' from supervisory authorities and noted further that: Besides, depending on the meetings and on the objectives identified under the agenda by the consolidating supervisor, the level of seniority of participants should be such as to being able to propose concrete actions or measures, if need be, and to 'pre-commit' their respective authorities. Supervisors should make sure that their representatives are mandated accordingly and that the work of the college is duly taken into account in day-to-day supervision of the group and its entities. 166 Admittedly, prescribing matters such as the level of seniority, the frequency or the composition of meetings might appear to be overly prescriptive. Indeed, some might see such proposals as somewhat frivolous. And yet, once we see them in the light of the trust problematic, the above points are in fact of 164 British Banking Association, 'Colleges of Supervisors' (10 June 2008). Available at: www.bba.org.uk/policy/article/colleges-of-supervisors/supervision. 165 Ibid. 166  great significance. Hence, there appear to be valid reasons to address them. Certainly, this approach has more to offer than simply prescribing trust without further consideration of how trust is supposed to come about.

V. CONCLUSION
This chapter looked at EMIR and the supervisory arrangements that it enacts with respect to CCPs. It put forward the paradigm of HCC-C, which was inspired by the rules that are found in EMIR. It went on to examine the building blocks of HCC-C. It argued that its basic rationale was that of greater interference in home state decision making. It then went on to reflect critically on the concept of interference and suggested that for interference to be useful (or constructive), various stumbling blocks had to be successfully addressed. Moreover, it submitted that a multi-pronged strategy was required. This strategy should feature greater interdependence, transparency and trust. In the process, the chapter sought to take issue with the approach that the Court of Justice and the EU legislature have taken with regard to trust: the idea that trust can simply be prescribed. It is perhaps not an exaggeration to say that the EU has simply got it wrong when it comes to trust. For one thing, the Court's insistence on 'prescribed trust' shows a lack of serious thinking on how trust can emerge. Together with the power imbalances that are inherent in the principles of home country control and mutual recognition, this approach is on its own unlikely to foster trust. In addition, this chapter suggests that if one wants to be serious about securing trust, the micro-level relations between officials (arguably an important aspect of any concerted effort to secure trust) must receive greater attention. The EBA's efforts in the banking sector are a step in this direction. On the other hand (and perhaps counter-intuitively), the chapter submitted in relation to matters that have received legislative attention-I think in particular of the voting modalities in colleges-that greater freedom to commit might actually be desirable, even though this freedom might ultimately need to be managed carefully.
Admittedly, in the banking sector, the proposed banking union might in the future resolve the horizontal trust issues that can permeate relationships between home and host authorities. And yet trust and cooperation may well continue to be an issue in the future. They might come to affect the vertical relations between the ECB and competent authorities, especially if the ECB were to rely extensively on a delegation of tasks to competent authorities for day-to-day supervisory work. But the bottom line is that any proposal on future supervisory centralisation should be made in light of an informed discussion about supervisory alternatives. By examining arrangements that do not fall squarely within the categories of home country control, host country control or EU supervisory centralisation, this chapter sought to contribute to this debate.