- Regulation of banks — Investment business
Systemic Risk, Regulation, and Proportionality
11.01 The topic of executive and employee compensation in financial institutions raises some interesting issues at the frontier between corporate governance and banking regulation that this chapter tries to highlight. Section II shows that excessive pay at financial institutions has often been indicated as one of the possible causes of the recent financial crisis, but the case for regulating compensation structure is rather weak, while regulation of remuneration and risk governance, and of remuneration disclosure are to some extent justified. Section III argues that the international principles interfere with compensation structures in a prescriptive way, particularly with regard to deferred variable pay and pay-out in instruments. For the rest, they mostly track best practices already followed by large institutions before the (p. 254) financial crisis, leaving some room to flexibility. Section IV argues that EU law shifted the setting of compensation from a supervisory approach to a regulatory one, adopting detailed and rigid provisions on the structure, governance, and disclosure of pay. Moreover, CRD IV introduced an unprecedented cap to variable remuneration, which may distort incentives and produce unintended consequences on bank risk-taking. Section V analyses the EU provisions on remuneration which apply to insurance undertakings, asset managers, and investment firms. Section VI examines possible ways to overcome the shortcomings of EU regulation of financial institutions in the remuneration area and suggests that it should be made more flexible and proportionate within the limits allowed by the international principles. The focus will be on systemic risk in order to identify the institutions which should be subject to the most stringent provisions as to the setting and monitoring of pay. Moreover, it is suggested that proportionality be implemented in wider terms than recently proposed by the European Commission in its review of CRD IV.1
A. Role of incentives in the crisis
11.02 Official policy documents issued after the crisis argue that the recourse to flawed remuneration structures, including the excessive use of short-term incentives for managers and other risk-taking employees, contributed to the failure of many banks and other financial institutions.2 Some scholars take issue with this hypothesis, while others offer empirical evidence in support of the same. This topic is intertwined with the more general one concerning the role of corporate governance in the crisis, as governance structures shape managerial incentives and monitor risk-taking by financial institutions. Official policy documents converge on the fact that the malfunctioning of corporate governance at banks and other financial institutions contributed to their crisis in the financial turmoil.3 Once again, scholars (p. 255) are divided: some argue that failed institutions often complied with best corporate governance standards (or at least appointed a majority of independent directors to their boards), while others criticize pre-crisis governance practices for lack of adequate monitoring on internal control and risk-management systems.
11.03 Some empirical studies analyse the structure of bank CEOs’ pay before the crisis asking whether short-term incentives may have distorted risk-taking by their institutions. Rüdiger Fahlenbrach and René Stulz4 identify some evidence that banks led by CEOs whose interests were better aligned with those of their shareholders had worse stock returns and a worse return on equity in the crisis.5 Lucian Bebchuk, Alma Cohen, and Holger Spamann6 offer a different view arguing that the large losses on shares that the top financiers suffered when their firms melted down do not offer a full picture of their payoffs. In the observed timeframe (2000–2008), the relevant executives received large amounts of cash bonus compensation and ‘regularly took large amounts of money off the table by unloading shares and options’.7
11.04 Other scholars offer different explanations, especially with regard to the circumstance that CEOs were heavily invested in their firms and lost tremendous amounts of money when the latter were brought down by the crisis. What could have led top managers to take huge risks threatening their firms’ survival if they were heavily invested in the same? One possible explanation is sheer incompetence of these managers, who arguably knew little of what was really going on at their firms.8 However, it is hard to apply a similar explanation to the majority of top managers (p. 256) at financial institutions in the early 2000s, given that ‘the corporate hierarchy is inherently a tough climb and weeds out a lot of incompetents, especially in the unforgiving and fiercely competitive financial sector’.9 A seemingly better explanation focuses on CEOs competing ‘for prestige by making more profits in the short-term or by heading league tables for underwriting or lending, regardless of the longer-term risk involved’.10 Another explanation may be found in a study showing that some banks had a culture of risk-taking and of compensating very heavily over the short term which influenced their performance.11 When these banks did well during boom times, their CEOs were acclaimed as heroes; however, in the recent crisis the same banks either did poorly or failed, and their CEOs became villains. Indeed, aggressive risk-taking in some banks paid off handsomely for a considerable period of time, but this was largely based on luck. Similar bets made by bank managers more recently led to disastrous outcomes in the financial crisis, once the tail risks which had been taken materialized.12
11.05 The studies cited so far focus on the remuneration of top executives at large banks. However, the remuneration of other employees should also be taken into account, particularly that of high-earners who contribute to risk-taking by the firm. Even though precise empirical data are lacking, it is well known that many of these employees were paid short-term incentives in amounts much greater than that of their fixed salaries. As explained by Diamond and Rajan, in the case of traders ‘many of the compensation schemes paid for short-term risk-adjusted performance. This gave traders an incentive to take risks that were not recognized by the system, so they could generate income that appeared to stem from their superior abilities, even though it was in fact only a market-risk premium’.13
11.06 No doubt, assuming that CEOs and other top managers had the right incentives—that is, not only short-term, but also long-term incentives—the fact that other employees had mainly short-term incentives should not have been a big problem, provided that sound risk-management systems were in place and an effective oversight was exercised on risk-takers by their superiors. However, as widely acknowledged in the aftermath of the crisis, this was not always the case at large banks, where risk-management systems were often deficient and top managers did not (p. 257) always understand, either as a result of flawed risk-management systems or just out of sheer incompetence, what their subordinates were doing. The problem was exacerbated by the huge amounts at play both for employers and employees, who were often incentivized to place financial bets in the crazy way aptly described by Professor Alan Blinder: ‘Heads, you become richer than Croesus; tails, you get no bonus, receive instead about four times the national average salary, and may (or may not) have to look for another job … Faced with such skewed incentives, they place lots of big bets. If heads come up, they acquire dynastic wealth. If tails come up, OPM [other people money] absorbs almost all losses’.14
B. Policy issues
11.07 As the author argued in another paper, the case for regulating the structure of compensation appears to be rather weak.15 Firstly, it is not sure that pay structures generally contributed to excessive risk-taking before the recent crisis. According to some of the studies cited above, corporate governance and compensation structures of CEOs at institutions that failed were not necessarily flawed. Secondly, even assuming that compensation structures were flawed—particularly those of traders and other middle-managers taking excessive risks—the need for their regulation would not be automatically established. In fact, excessive risk-taking could be curbed directly through prudential regulation of financial institutions, rather than by modelling the incentives of their employees, given that regulators may not be professionally qualified for designing pay structures.16 Thirdly, mandating pay structures hampers the flexibility of compensation arrangements, which need tailoring to individual firms and managers, in light of the latter’s portfolios of their institutions’ securities. Moreover, boards of directors lose one of their key governance functions, finding it more difficult to align executives’ incentives to corporate strategy and risk profile. This may also create problems in keeping and attracting managerial talent, particularly from countries that adopt a more liberal stance or from firms that are not subject to regulatory constraints (such as hedge funds or private equities).
11.08 There is no doubt that competent authorities should supervise compensation from the perspective of the supervised institutions’ safety and soundness.17 Rather than designing compensation structures ex ante, which is a matter for boards, they (p. 258) should analyse the impact of remuneration structures on risk-taking and conduct their surveillance activities accordingly, for instance by imposing higher capital requirements to institutions adopting ‘aggressive’ remuneration mechanisms which may lead to excessive risk-taking. Moreover, supervisors should check bank compliance with compensation governance requirements and with the disclosure requirements concerning remuneration policies. Rather than interfering with pay structures, this type of regulation aims to ensure that organizational structures and procedures are in place for the setting of pay in compliance with safety and soundness requirements.
11.09 The case for regulating remuneration governance is stronger than that for mandating structure, at least to the extent that corporate governance was found deficient and in need of improvement after the financial crisis.18 In fact, some of the studies already cited show that ailing institutions were often the ones with the best corporate governance in place under international standards.19 However, the alignment of managerial interests with those of shareholders which ‘good’ governance determines is not enough from a financial stability perspective, for shareholders’ interests are not aligned with those of other stakeholders, like depositors and taxpayers, who will in the end pay for the costs of a crisis.20 Indeed, shareholders tend to push managers and boards to take risks in an amount higher than socially desirable, while the interest alignment deriving from good governance and remuneration practices works in the same direction, rather than that of financial stability.21
11.10 To the extent that risk-management practices are flawed and board oversight on them is also deficient, an improvement of board organization and functioning is in the interest of shareholders, who would otherwise be negatively affected by (p. 259) excessive risk-taking.22 Therefore, some regulation of corporate governance is justified from a prudential perspective; moreover, supervision should ensure that there is proper oversight of risk management within institutions, not only from a shareholders’ perspective, but also from a societal and systemic viewpoint. Also remuneration practices of risk-takers are subject to oversight by the board, which should therefore check that they are sound from a risk-management perspective, while prudential supervision could exert further pressure on boards in the same way.23
11.11 In addition, mandatory disclosure of compensation practices is justified on at least two counts. Firstly, detailed disclosure of pay structure and amounts makes boards and managers more accountable to shareholders and the capital markets. Secondly, disclosure allows shareholders to better exercise their say-on-pay rights, while enabling supervisors to perform their function more effectively.24
A. The FSB principles and standards
11.12 The FSB principles are addressed to ‘significant financial institutions’, which more than others deserve an internationally uniform regime. They cover four main compensation areas: governance, structure, disclosure, and supervision. As to compensation governance, they incorporate well-known best practices concerning the strategic and supervisory role of the board. In addition, they reflect the post-crisis emphasis on bank risk management and monitoring of the same by the board of directors, who should determine the risk appetite of the firm. They reiterate the role of the remuneration committee, requiring its liaison with the risk committee to ensure compliance with the relevant requirements.
11.13 Compensation structures are considered by the principles along lines that reflect, to a large extent, best practices generally followed before the crisis. Indeed, the role and limits of equity-based compensation, as well as the potentially perverse effects of short-term incentives, have attracted much attention over the last twenty years.25 However, pre-crisis practices mainly emphasized the alignment of managers’ incentives with shareholder wealth maximization. The principles break new ground by requiring financial institutions to align compensation with prudent risk-taking. (p. 260) Accordingly, compensation should be adjusted for all types of risk, including those considered difficult-to-measure, such as liquidity risk, reputation risk, and capital cost. Compensation outcomes should be symmetric with risk outcomes.
11.14 Deferment of compensation, traditionally used as a retention mechanism (on the basis that a ‘bad leaver’ would generally lose unpaid deferrals), should make compensation pay-out schedules sensitive to the time horizon of risks. In particular, a substantial portion of variable compensation (i.e. 40–60 per cent) should be payable under deferral arrangements over a period of not less than three years, provided that this period is correctly aligned with the nature of the business, its risks, and the activities of the employee in question. Furthermore, a substantial portion (i.e. more than 50 per cent) of variable compensation should be awarded in shares or share-linked instruments, as long as the same create incentives aligned with long-term value creation and the time horizons of risk. In any event, awards in shares or share-linked instruments should be subject to an appropriate retention policy.
11.15 The principles also tackle concerns relative to bonuses, which famously emerged during the recent crisis. They require ‘malus’ and ‘clawback’ mechanisms, which enable boards to reduce or reclaim bonuses paid on the basis of results that are unrepresentative of the company’s performance over the long term or later prove to have been misstated. They consider ‘guaranteed’ bonuses (i.e. contracts guaranteeing variable pay for several years) as conflicting with sound risk management and the pay-for-performance principle. Severance packages need to be related to performance achieved over time and designed in a way that does not reward failure.
11.16 Compensation disclosure, despite being widely practiced pre-crisis, did not always meet the relevant standards. After the crisis, there has been consensus that disclosure should benefit not only shareholders, but also other stakeholders (e.g. creditors and employees). Moreover, disclosure should identify the relevant risk management and control systems and facilitate the work of supervisors in this area. The principles add new items of disclosure, such as deferral, share-based incentives, and criteria for risk adjustment. They also require effective supervision. In the case of a failure by a firm to implement ‘sound’ compensation policies, prompt remedial action should be taken by supervisors and appropriate corrective measures should be adopted to offset any additional risk that may result from non-compliance or partial compliance with the relevant provisions.
B. The choice between standards and rules
11.17 The FSB principles represent a political compromise between the various interests at stake in the area of compensation, incorporating traditional criteria and adapting them to new circumstances. Firstly, they focus on long-term incentives, in order to counter the role allegedly played by short-term incentives in the crisis. Since executive compensation packages at most large banks before the crisis were (p. 261) already balanced between short-term and long-term incentives at least for CEOs (as shown by the Fahlenbrach and Stulz paper cited above),26 the international principles track already existing practices, but extend the same to a greater number of bank employees. Secondly, the principles widen the powers of supervisors by explicitly making pay at financial institutions subject to prudential supervision. Thirdly, similar to other international financial standards, the principles remain at a sufficient level of generality and allow for flexibility in implementation; in several instances, financial institutions are permitted to depart from a given principle or standard, if application of the same would lead to unsound consequences.
11.18 However, the principles also interfere with compensation structures by asking institutions, for instance, to defer 40–60 per cent of variable compensation and to award at least half of variable compensation in shares. This type of ‘one-size-fits-all’ approach is open to criticism for all the reasons indicated in Section I. There is no doubt that states are free to implement the principles through either regulation or supervision and, if they adopt a supervisory approach to implementation, the interference with remuneration structures might be softer. Nonetheless, the existence of detailed principles and standards such as the ones just indicated—which are indeed ‘rules’ rather than ‘standards’ for their level of specificity—will inevitably shape supervisory actions producing results not entirely dissimilar from those of ad hoc regulation.
11.19 The FSB principles have been implemented along different models.27 Some jurisdictions follow a primarily supervisory approach to implementation, involving principles and guidance and the associated supervisory reviews. In other jurisdictions the model includes a mix of regulation and supervisory oversight, with new regulations often supported by supervisory guidance that illustrates how the rules can be met. In jurisdictions like the European Union, a regulatory approach to implementation prevails grounded on two layers of directives and leaving only a narrow scope to supervisory discretion (see Sections III and IV below).
11.20 In all jurisdictions, however, the law in action for remuneration at financial institutions consists of rules rather than standards. In fact, either the law foresees detailed requirements—such as those concerning the deferment of compensation over a stated period of time—or similar requirements are enforced by supervisors in practice on the basis of the standards foreseen by the law. Similar comments hold for the FSB principles and standards, which on one side offer a significant level of specificity; on the other, are enforced internationally by the FSB checking the level of compliance with the principles by individual states and making the results of (p. 262) similar reviews public so as to stimulate convergence by all the jurisdictions concerned.28 Clearly, financial supervisors are comfortable with this approach, to the extent that their actions towards the supervised institutions are based on pre-fixed standards. But even supervisees often prefer to know in advance the criteria under which their remuneration practices will be assessed, despite the fact that this inevitably reduces the space for autonomy and flexibility in the setting of bankers’ pay.
11.21 The European Union initially adopted a supervisory approach to remuneration through the Commission Recommendation on remuneration in the financial sector (2009) touching upon the governance and structure of pay along lines similar to those followed by the FSB principles.29 At the same time, the Committee of European Banking Supervisors (CEBS) issued high-level principles for remuneration policies at banks.30 However, subsequent reviews on the national implementation of these documents revealed shortcomings in several areas,31 such as the measurement of risk-adjusted performance, the scope of the new standards, proportionality, and home/host relationships. Similar differences, together with increased pressure from the media, politicians, and the public, led to a change in regulatory approach. The Capital Requirements Directive (implementing the Basel capital requirements) was amended twice also to include provisions on bankers’ remuneration: in 2010 when CRD III was enacted32 and in 2013 when the CRD IV package was adopted, including a new directive concerning, inter alia, bankers’ remuneration.33 In addition, CEBS issued supervisory guidance in order to facilitate (p. 263) compliance with the remuneration principles included in CRD III,34 while the EBA issued new Guidelines under the new directive.35
A. CRD IV
11.22 Notwithstanding the fact that executive pay at large banks was lower and more balanced after the crisis,36 the European regulation in this area was deeply overhauled by Directive 2013/36/EU (Capital Requirements Directive, CRD IV).37 The new regime applies on a consolidated basis, that is, to ‘institutions at group, parent company and subsidiary levels, including those established in offshore financial centres’ (Article 92(1)). The ratio for an EU-wide scope of application is ‘to protect and foster financial stability within the Union and to address any possible avoidance of the requirements laid down in this Directive’ (67th considerandum). Moreover, the new regime applies to different categories of staff including senior management, risk-takers, staff engaged in control functions, and any employee receiving total remuneration that takes them into the same remuneration bracket as senior management and risk-takers, whose professional activities have a material impact on their risk profile (Article 92(2)). In this regard, the Commission has recently adopted a delegated Regulation including regulatory technical standards on the identification of risk-takers.38
11.23 Article 74 (1) of CRD IV requires institutions to have in place a remuneration policy for all staff, which should comply with the principles set out in Articles 92 and 93 of the Directive and with EBA Guidelines on sound remuneration policies. The remuneration policy should specify all components of remuneration and include the pension policy too. It should be consistent with the objectives of the institution’s business and risk strategy, corporate culture and values, long-term (p. 264) interests of the institution, and the measures used to avoid conflicts of interest, and should not encourage excessive risk-taking. The remuneration policy should contain the performance objectives for the institution, the methods for the measurement of performance, the structure of variable remuneration, and the ex ante and ex post risk-adjustment measures of the variable remuneration.39
11.24 Under Article 92(2) institutions comply with the principles just stated and other principles ‘in a manner and to the extent that is appropriate to their size, internal organization and the nature, scope and complexity of their activities’. As explained in Recital 66:
The provisions of this Directive on remuneration should reflect differences between different types of institutions in a proportionate manner, taking into account their size, internal organisation and the nature, scope and complexity of their activities. In particular, it would not be proportionate to require certain types of investment firms to comply with all those principles.
The EBA Guidelines further specify that the proportionality principle ‘aims to match remuneration policies and practices consistently with the individual risk profile, risk appetite and strategy of an institution, so that the objectives of the obligations are effectively achieved’.40 However, the EBA’s recent opinion on proportionality expressed the view—shared by the European Commission—that ‘the wording of Article 92 (2) does not permit exemptions or waivers to the application of the remuneration principles’.41
Where the CRD sets some specific requirements with numerical criteria (i.e. the minimum deferral period of three to five years; the minimum proportion of 40% to 60% of variable remuneration that should be deferred …), institutions should apply the criteria based on proportionality, considering that in particular for significant institutions and their senior management and members of the management body more strict criteria should be set, and in any case apply at least the minima criteria set in the CRD.42
According to this interpretation, proportionality can only determine an enhancement of the applicable criteria, not a waiver of the same. As a result, ‘neutralization’ practices, which were allowed under the CEBS Guidelines with respect to the (p. 265) requirements previously in force,43 are no longer admitted by the European authorities with a significant restraint of the proportionality principle.
11.26 Nonetheless, EBA opinion acknowledged that information provided by competent authorities, together with evidence gathered from stakeholders during the consultation period, show that ‘there are different legal interpretations of the proportionality clause as established in Article 92(2) of Directive 2013/36/EU, which have led to different applications of the remuneration principles at national level. These approaches would be in line with the CEBS Guidelines on remuneration policies and practices’.44 As a result, the EBA concluded that action is necessary at the level of the EU institutions in order to ensure that remuneration requirements are applied consistently across the Union. In the EBA’s view, CRD IV should be amended ‘to exclude certain small, non-complex institutions from the requirements to apply the remuneration principles regarding deferral and payment in instruments for variable remuneration, and to limit the scope of those remuneration principles as regards staff who receive low amounts of variable remuneration, including large institutions’.45 The European Commission has recently presented a proposal to this effect, as explained in Section IV.C below.
11.27 Article 94(1) provides several requirements for the variable elements of remuneration. Some of them are rather generic, such as the one requiring performance pay to be based on a combination of the assessment of the performance of the individual and of the business unit concerned and of the overall results of the institution. In addition, performance should be assessed in a multi-year framework in order to ensure that the assessment process is based on longer-term performance and that the actual payment of performance-based components of remuneration is spread over a period which takes account of the underlying business cycle of the credit institution and its business risks. The EBA Guidelines specify that when the award of variable remuneration, including long-term incentive plans (LTIP), is based on past performance of at least one year, but also depends on future performance conditions, institutions should clearly set out to staff the additional performance conditions that have to be met after the award for the variable remuneration to vest.46 The additional performance conditions should be set for a predefined performance period of at least one year and, when they are not met, up to 100 per cent of the variable remuneration awarded under those conditions should be subject to malus arrangements.47
(p. 266) 11.28 In addition under Article 94(1), the total variable remuneration should not limit the ability of the institution to strengthen its capital base. Furthermore, the fixed and variable components of total remuneration should be appropriately balanced and the fixed component should represent a sufficiently high proportion of the total remuneration to allow the operation of a fully flexible policy on variable remuneration components, including the possibility to pay no variable remuneration component.
11.29 Other requirements are more specific, particularly the cap on variable remuneration that the European Parliament asked to be included in CRD IV and that will be analysed in the following paragraph. Severance payments are also covered by the provision: they will have to reflect performance achieved over time and should not reward failure or misconduct. In addition, remuneration packages relating to compensation or buy out from contracts in previous employment must align with the long-term interests of the institution concerned, including retention, deferral, performance, and clawback arrangements.
11.30 In general, the measurement of performance used to calculate variable remuneration should include an adjustment for all types of current and future risks and take into account the cost of the capital and the liquidity required. A substantial portion, and in any event at least 50 per cent of variable remuneration shall consist of a balance of (i) shares or equivalent ownership interests, subject to the legal structure of the institution concerned or share-linked instruments or equivalent non-cash instruments, in the case of a non-listed institution; (ii) where possible, other instruments within the meaning of Articles 52 or 63 of Regulation (EU) No 575/2013 or other instruments which can be fully converted to Common Equity Tier 1 instruments or written down, that in each case adequately reflect the credit quality of the institution as a going concern and are appropriate to be used for the purposes of variable remuneration.
B. The cap on variable remuneration
11.31 While Directive 2010/76/EU (CRD III) simply required an ‘appropriate balance’ between fixed and variable remuneration (Article 23), CRD IV also establishes a maximum ratio between the two remuneration components. The definition of these components is found in the Directive’s 64th considerandum, stating that fixed remuneration includes ‘payments, proportionate regular pension contributions, or benefits (where such benefits are without consideration of any performance criteria)’, while variable remuneration includes ‘additional payments, or benefits depending on performance or, in exceptional circumstances, other contractual elements but not those which form part of routine employment packages (such as healthcare, child care facilities or proportionate regular pension contributions)’. Both monetary and non-monetary benefits are comprised in the relevant (p. 267) definitions. The criteria for setting fixed and variable remuneration are found in Article 92(2)(g) stating that basic fixed remuneration should primarily reflect relevant professional experience and organizational responsibility, while variable remuneration should reflect ‘a sustainable and risk adjusted performance as well as performance in excess of that required to fulfil the employee’s job description as part of the terms of employment’.
11.32 Under Article 94(1)(f), the fixed and variable components of total remuneration should be appropriately balanced and the fixed component should represent a sufficiently high proportion of the total remuneration ‘to allow the operation of a fully flexible policy on variable remuneration components, including the possibility to pay no variable remuneration component’. However, Article 94(1)(g) further constrains this proportion by stating that the variable component should not exceed 100 per cent of the fixed component of the total remuneration for each individual. Moreover, Member States may set a lower maximum percentage (as Belgium and the Netherlands did, by setting 50 per cent and 20 per cent respectively). Alternatively, Member States may allow shareholders of the institution concerned to approve a higher maximum level of the ratio between fixed and variable remuneration provided the overall level of the variable component shall not exceed 200 per cent of the fixed component of the total remuneration for each individual. Member States may also set a lower percentage. In any case, approval of a higher percentage should occur through a special procedure that is described in detail by Article 94(1)(g)(ii).
11.33 The official justification for the cap on variable remuneration is ‘to avoid excessive risk taking’ (65th considerandum). The Directive implicitly assumes that an excessive level of variable remuneration is likely to induce excessive risk taking by the managers of financial institutions, as (arguably) shown by the financial crisis. The need for capping variable pay, based on the assumption that excessive bonuses contributed to recent bank failures, was brought forward quite vigorously by France and Germany in the aftermath of the crisis, but was not recognized at international level, where the FSB rejected any suggestion of introducing a cap on bonuses given the firm opposition of the US government.48 However, the initial Commission proposal of CRD IV did not include a similar cap, which was later suggested by the European Parliament among a number of amendments to the Commission’s proposal.49
(p. 268) 11.34 The adoption of a cap was nonetheless controversial. The United Kingdom, in particular, vehemently contested the same, reflecting the City of London’s concerns that capping variable pay would disrupt remuneration practices of investment banks, which rely heavily on bonuses and other types of performance-related pay.50 The UK government brought proceedings against the European Parliament and the Council seeking the annulment of the CRD IV provisions regarding the limits on variable remuneration, the EBA’s rule-making powers in this area and the public disclosure of certain details of the material risk-takers’ salaries required by the Capital Requirements Regulation (CRR). The United Kingdom mainly maintained that the contested provisions have an inadequate Treaty legal base in addition to being disproportionate and failing to comply with the principle of subsidiarity.51
11.35 However, the Opinion of Advocate General Jääskinen found the UK’s pleas ungrounded, firstly by arguing that the cap on variable remuneration ‘does not impact directly on the level of pay’, rather it ‘merely establishes a ratio between the fixed and variable element without affecting the level of remuneration as such’.52 As contended below, this is disputable from an economic perspective, as it is likely that the cap on variable pay will push fixed pay upwards, while variable remuneration will stay below the amount that the labour market would otherwise require. Therefore, the cap may have an impact both on the structure and on the level of remuneration. Secondly the Advocate General argued that ‘all the procedural requirements relating to the assessment of the compliance of the proposal with the principles of proportionality and subsidiarity were duly respected by the EU legislature’, which ‘possesses a wide margin of discretion’.53 However, the Opinion largely ignores (or is agnostic about) the possible arguments—summarized below—against regulating the level of bankers’ pay, implicitly relying on the post-crisis populist debate about excessive pay at financial institutions and on the assumption that all arguments in favour and against regulation have been duly considered by the legislature.54 Nevertheless, the United Kingdom—immediately after publication of the (p. 269) Advocate General’s Opinion—withdrew its application and the case was therefore discontinued.55
C. Criticism of the cap
one is the often excessive level of remuneration in the financial sector; the other one is the structure of this remuneration … Social-political dissatisfaction has tended recently to focus, for understandable reasons, on the former. However, it is primarily the latter issue which has had an adverse impact on risk management and has thereby contributed to the crisis. It is therefore on the structure of remuneration that policy-makers should concentrate reforms going forward.56
Similar arguments are found, across the Atlantic, in a report by a committee of highly distinguished economists (the ‘Squam Lake Group’) who argued that governments should generally not regulate the level of executive compensation in financial institutions. Indeed, no convincing evidence has been seen to suggest that high levels of compensation in financial companies are inherently risky for the companies themselves or the overall economy. Moreover, limits on pay are likely to cause unintended consequences, so that society is better off if compensation levels are set by market forces.57 In a similar vein, Richard Posner argues: ‘… efforts to place legal limits on compensation are bound to fail, or to be defeated by loopholes, or to cause distortions in the executive labour market and in corporate behaviour’.58
11.37 Indeed, experience relative to the use of ‘role-based allowances’ by around thirty-nine European banks shows that efforts to circumvent the EU cap were soon made after its introduction. Role-based allowances are linked to the position and organizational responsibility of staff. As explained by the EBA in an opinion given to the European Commission, ‘allowances’ are payments or benefits paid in addition (p. 270) to basic salary and variable remuneration (bonus).59 Banks tended to consider all allowances, including role-based allowances, as fixed remuneration, arguing that they were not based on performance. However, role-based allowances are generally not part of the basic salary and are not pensionable; are initially granted for a limited period of time; can be reduced, suspended, or cancelled by banks on a fully discretionary basis; and include other contractual conditions which do not form part of routine employment packages. In the EBA’s opinion, in order to qualify as fixed remuneration ‘the conditions for their granting and the amount of the role-based allowance should be predetermined, transparent to staff, permanent, i.e. maintained over time and tied to specific role and organisational responsibilities, not provide incentives to take risks and, without prejudice to national law, be non-revocable.’60 As a result, the EBA believes that role-based allowance not complying with these conditions (e.g. it is not predetermined, is not permanent, or provides incentives to take risks) should be classified as variable remuneration ‘in line with the letter and purpose of the CRD’.61 Similar criteria are now included in the the EBA Guidelines, which in section 7 identify the categories of remuneration (fixed and variable) and in section 8.1 deal specifically with allowances.
11.38 Going back to the widespread criticism of the maximum ratio, several arguments show that neither the objective to reduce excessive risk-taking nor the one to reduce perceived excesses in the level of banking remuneration will be achieved by capping variable remuneration.62 Firstly, the cap will likely increase the level of fixed remuneration, making banks more vulnerable to business cycles and therefore increasing the risk of bank failure. Anecdotal evidence already shows that fixed pay at large European banks is on the rise,63 while the EBA reached similar findings for EU (p. 271) banks in general (as reported in the following section), even though a similar trend predates the introduction of a maximum ratio for fixed-variable pay.
11.39 Secondly, the traditional bonus system at investment banks, which is characterized by below-market salaries and high-bonus opportunities, provides strong incentives to avoid ‘bad’ risks and take ‘good’ ones. On the contrary, the new system—which will be characterized by above-market salaries and ‘capped’ bonuses—provides incentives to take ‘bad’ risks and avoid ‘good’ ones. In fact, if bad risks materialize, the bank manager will not suffer, for his or her remuneration is to a large extent fixed. But, if the bank shuns good risks and the relevant profits, the responsible manager will not be worse off given that his or her bonus is capped. Indeed, the bonus cap reduces incentives to create value, which is the main purpose of variable pay.64
11.40 Thirdly, executive remuneration is largely set by the markets, so that a bonus cap could have unintended consequences on the firms’ ability to hire people of adequate standing in the international market for managers. In the end, remuneration ‘will reflect a less-talented workforce as the top producers leave for better-paying opportunities in financial firms not subject to the pay restrictions’. In other words, the cap ‘will not lead to lower levels of overall remuneration after adjusting for ability and the risk of the remuneration package’.65 Furthermore, the cap on variable pay will reduce the competitiveness of the EU banking sector relative to non-EU banks and other non-bank financial intermediaries which are not subject to similar restrictions.
11.41 Fourthly, the mandatory cap reflects a ‘one-size-fits-all’ approach which is clearly too rigid, for different types of credit institutions and investment firms present different levels of risk exposure, so that an incentive structure which is appropriate for one firm is not necessarily suited to another. Moreover, the EU bonus cap applies to all credit institutions, without regard to their size and to systemic risk considerations.
11.42 Additional problems may derive from the combination of different tools to deal with the same problem (excessive pay). Indeed, a good part of the CRD IV provisions are based on the international principles’ approach to bankers’ remuneration, which is flexible and relies on pay governance, transparency, and the requirement of an adequate proportion between fixed and variable pay. Other provisions incorporate the international requirements for the deferment of variable pay and the (p. 272) payment of a portion of the same in equity and other financial instruments issued by the bank. The juxtaposition of a cap on variable pay to similar requirements not only may appear redundant and counterproductive, for the reasons explained by Murphy, but could determine further unintended consequences. In particular, the pressure to increase fixed pay deriving from the cap could be enhanced by the requirement that variable pay should be deferred and partly paid in equity or other financial instruments. In fact, a similar requirement pushes variable pay to a higher level than what would be agreed if remuneration were paid in cash and without deferment; but higher variable pay determines an increase in fixed pay given the fixed ratio between the two components of remuneration. The final result of the cumulus of different criteria will be an increase of overall remuneration, including fixed and variable components.
11.43 The EBA conducted a benchmarking exercise on remuneration practices with regard to 2014, examining the impact of the bonus cap.66 One of the EBA’s main findings is that the fixed remuneration of identified staff increased, while the variable remuneration was reduced; on average, following the introduction of the cap, the ratio between fixed and variable plunged to 65.48 per cent from 104.27 per cent in 2013. The highest variable remuneration and total remuneration were paid in investment banking where the ratio of variable to fixed remuneration for identified staff dropped on average from 191.17 per cent in 2013 to 88.89 per cent in 2014. In retail banking the same ratio dropped from 35.05 per cent in 2012 and 24.97 per cent in 2013 to 30.29 per cent in 2014; in asset management from 128.86 per cent in 2012 and 107.88 per cent in 2013 to 100.19 per cent in 2014. Therefore, the mandatory cap introduced by CRD IV easily accommodates the average ratio between fixed and variable remuneration in retail banking, while it is close to the average ratio in asset management and investment banking. In other words, the ratio could be higher in the last two sectors in the absence of the cap.
11.44 These data were commented on by the Commission’s report to the European Parliament including an assessment of the remuneration rules under CRD IV.67 The Commission rejected the idea of a causal link between the Directive’s maximum ratio and the decrease in the variable part of remuneration with respect to the fixed one arguing that a similar trend had already begun several years before (p. 273) the introduction of the maximum ratio.68 The Commission added that there are other elements impacting on the levels and proportions of the remuneration components, such as financial performance, profitability, and general prudential requirements, and that an increase in the fixed portion of remuneration has also been observed in several non-EU jurisdictions, including the United States and some Asian countries. Moreover, the EBA’s findings have been reached on the basis of averages mainly collected from large banking groups and need to be interpreted with care. However, the Commission conceded that ‘while the overall shift towards fixed remuneration cannot be clearly attributed to the maximum ratio, it is likely that in some individual cases the maximum ratio has led to a shift from variable to fixed remuneration’.69 This is consistent with the criticism of the maximum ratio advanced in the previous section on a theoretical level.
11.45 The Commission however dismissed other criticism of the maximum ratio. Firstly, it argued that it is too early to establish whether reducing the variable part of remuneration has substantially affected the risk-taking incentives at financial institutions.70 In addition, a study made by the Institut für Finanzdienstleistungen for the Commission reports the answers to a questionnaire on remuneration structure and incentives sent to banks, showing that 41 per cent of identified staff stated that an increase in fixed pay would not affect their risk-taking behaviour and 27 per cent said that it would have little effect.71 An overwhelming 94 per cent of respondents disagreed that more fixed pay would reduce motivation to take risks. The study comments on these data by arguing that the reaction of bankers to fixed incentives may be cultural: ‘Certain cultures respond to higher fixed pay as an incentive, while others are more motivated by the opportunity to earn more through more variable pay.’72
11.46 Secondly, the Commission downplayed the impact of the maximum ratio on fixed costs and profitability, arguing that fixed remuneration of identified staff in the institutions examined represented, in 2014, below 5 per cent of the total administrative costs, while variable remuneration accounted for only 1 to 2 per cent of the total administrative costs of most of the institutions examined. Moreover, the total fixed remuneration cost of identified staff on an aggregate basis was relatively small compared with the net profits of institutions: ‘This suggests there is a non-negligible margin for fixed remuneration to increase before reaching a level that would threaten the overall profitability of institutions.’73
(p. 274) 11.47 Thirdly, the Commission rejected the claim that the maximum ratio would reduce institutions’ competitiveness by negatively affecting their ability to attract and retain talent. In its view, ‘many elements play a part in a staff’s member decision to move, such as job security, promotion prospects, the reputation enjoyed by the sector, taxation, family, language and living conditions’.74 Also the choice of the proportion between fixed and variable remuneration may depend on personal or cultural preferences. However, the Commission conceded that this issue may merit further assessment when more experience with the rule is gained in practice.
11.48 Banks are at the core of the remuneration regime fixed at international and EU levels for the simple reason that they experienced the most serious problems in this area throughout the financial crisis and were also the target of the greatest financial stability concerns as a result of repeated crises both in the United States and the European Union. However, other institutions are not immune from the problems concerning employees’ remuneration and its impact on risk-taking, while the international principles on sound compensation practices apply to all financial institutions. The present analysis should therefore briefly be complemented by a review of the EU provisions extending and adapting the remuneration regime to financial institutions other than banks, such as insurance undertakings, asset managers, and investment firms.
11.49 The provisions on insurers’ remuneration policy and its regulation are found in the Commission Delegated Regulation (EU) 2015/35 of 10 October 2014 supplementing Directive 2009/138/EC on the taking-up and pursuit of the business of insurance and reinsurance (Solvency II).75 Article 275 of this Regulation (Solvency II Regulation) provides that, when adopting their remuneration policy, insurance and reinsurance undertakings shall comply with a number of principles, including the following. Firstly, the remuneration policy and remuneration practices shall be established, implemented, and maintained in line with the undertaking’s business and risk-management strategy, its risk profile, objectives, risk-management practices, and the long-term interests and performance of the undertaking as a whole, (p. 275) and shall incorporate measures aimed at avoiding conflicts of interest. Moreover, the remuneration policy shall promote sound and effective risk management and shall not encourage risk-taking that exceeds the risk tolerance limits of the undertaking. In addition, the remuneration policy shall apply to the undertaking as a whole and contain specific arrangements that take into account the tasks and performance of the administrative, management, or supervisory body; persons who effectively run the undertaking or have other key functions; and other categories of staff whose professional activities have a material impact on the undertaking’s risk profile.
11.50 Secondly, the administrative, management, or supervisory body of the undertaking shall establish the remuneration policy for those categories of staff whose professional activities have a material impact on the undertaking’s risk profile and is responsible for the oversight of its implementation. An independent remuneration committee shall be created if appropriate in relation to the significance of the insurance or reinsurance undertakings in terms of size and internal organization.
11.51 Thirdly, some principles shall apply to the remuneration policy of insurance undertakings. To start with, the fixed and variable components of remuneration schemes shall be balanced so that the fixed or guaranteed component represents a sufficiently high proportion of the total remuneration to avoid employees being overly dependent on the variable components and to allow the undertaking to operate a fully flexible bonus policy, including the possibility of paying no variable component. Where variable remuneration is performance related, the total amount of the variable remuneration is based on a combination of the assessment of the performance of the individual and of the business unit concerned and of the overall result of the undertaking or the group to which the undertakings belongs.
11.52 In addition, a substantial portion of the variable remuneration shall contain a flexible, deferred component that takes account of the nature and time horizon of the undertaking’s business. The deferral period shall not be less than three years and the period shall be correctly aligned with the nature of the business, its risks, and the activities of the employees in question. Financial and non-financial criteria shall be taken into account when assessing an individual’s performance. Moreover, the measurement of performance, as a basis for variable remuneration, shall include a downwards adjustment for exposure to current and future risks, taking into account the undertaking’s risk profile and the cost of capital.
B. Asset managers
11.53 Under Article 14a, Directive 2009/65/EC as amended by Directive 2014/91/EU of 23 July 2014 (UCITS V), Member States shall require management companies to establish and apply remuneration policies and practices that are consistent with, and promote, sound and effective risk management and that neither encourage (p. 276) risk-taking which is inconsistent with the risk profiles, rules, or instruments of incorporation of the UCITS that they manage nor impair compliance with the management company’s duty to act in the best interest of the UCITS. The remuneration policies and practices shall include fixed and variable components of salaries and discretionary pension benefits.
11.54 When establishing and applying their remuneration policies, management companies shall comply with the Directive’s principles in a way and to the extent that is appropriate to their size, internal organization and the nature, scope and complexity of their activities (Article 14b). In particular, the remuneration policy must be in line with the business strategy, objectives, values, and interests of the management company and the UCITS that it manages and of the investors in such UCITS, and includes measures to avoid conflicts of interest. In addition, the assessment of performance is set in a multi-year framework appropriate to the holding period recommended to the investors of the UCITS managed by the management company in order to ensure that the assessment process is based on the longer-term performance of the UCITS and its investment risks and that the actual payment of performance-based components of remuneration is spread over the same period. Moreover, subject to the legal structure of the UCITS and its fund rules or instruments of incorporation, a substantial portion, and in any event at least 50 per cent of any variable remuneration component consists of units of the UCITS concerned, equivalent ownership interests, or share-linked instruments or equivalent non-cash instruments with equally effective incentives as any of the instruments referred to above. The instruments shall be subject to an appropriate retention policy designed to align incentives with the interests of the management company and the UCITS that it manages and the investors of such UCITS.
11.55 Furthermore, a substantial portion, and in any event at least 40 per cent, of the variable remuneration component, is deferred over a period which is appropriate in view of the holding period recommended to the investors of the UCITS concerned and is correctly aligned with the nature of the risks of the UCITS in question. This period shall be at least three years. However, the variable remuneration, including the deferred portion, is paid or vests only if it is sustainable according to the financial situation of the management company as a whole, and justified according to the performance of the business unit, the UCITS, and the individual concerned. The total variable remuneration shall generally be considerably contracted where subdued or negative financial performance of the management company or of the UCITS concerned occurs, taking into account both current compensation and reductions in pay-outs of amounts previously earned, including through malus or clawback arrangements.
11.56 Similar principles apply to Alternative Investment Fund Managers (AIFMs) under Directive 2011/61/EU of 8 June 2011.76 Article 13 of this Directive provides that (p. 277) Member States shall require AIFMs to have remuneration policies and practices for those categories of staff, including senior management, risk-takers, control functions, and any employees receiving total remuneration that takes them into the same remuneration bracket as senior management and risk-takers, whose professional activities have a material impact on the risk profiles of the AIFMs or of the AIFs they manage, that are consistent with and promote sound and effective risk management, and do not encourage risk-taking which is inconsistent with the risk profiles, rules, or instruments of incorporation of the AIFs they manage. The AIFMs shall determine the remuneration policies and practices in accordance with Annex II to the Directive.
11.57 Investment firms are subject to the regime applicable also to credit institutions (CRD IV), to which MiFID II adds the requirement that the management body has to develop a specific remuneration policy for persons involved in the provision of services to clients so as to ‘encourage responsible business conduct, fair treatment of clients as well as avoiding conflict of interest in the relationship with client’ (Article 9(3)(c)).77 Article 27 of the Delegated Council Regulation (EU) 2017/565 of 25 April 2016 supplementing MiFID II provides more detailed requirements. Firstly, investment firms shall define and implement remuneration policies and practices under appropriate internal procedures taking into account the interests of all the clients of the firm, with a view to ensuring that clients are treated fairly and their interests are not impaired by the remuneration practices adopted by the firm in the short, medium, or long term. In addition, remuneration policies and practices shall be designed in such a way as not to create a conflict of interest or incentive that may lead relevant persons to favour their own interests or the firm’s interests to the potential detriment of any client. Secondly, investment firms shall ensure that their remuneration policies and practices apply to all relevant persons with an impact, directly or indirectly, on investment and ancillary services provided by the investment firm or on its corporate behaviour, regardless of the type of clients, to the extent that the remuneration of such persons and similar incentives may create a conflict of interest that encourages them to act against the interests of any of the firm’s clients. Thirdly, principles reflecting those already analysed in this chapter apply to the governance aspects of remuneration and to the balance between its fixed and variable components.
11.58 In this section, two possible ways to improve EU regulation of remuneration at financial institutions are examined, always assuming that there should be regulation, as is presently agreed at international level. One is focusing on systemic risk, the other is enhancing proportionality. The bonus cap is not discussed here given that it has already been argued in this chapter that there should not be one.
11.59 The FSB principles and standards only apply to ‘significant financial institutions’. The Financial Stability Forum (FSF) (now the Financial Stability Board (FSB)) specified in its Introduction to the Principles that ‘they are intended to apply to significant financial institutions, but they are especially critical for large, systemically important firms’. This establishes a clear link between the Principles and systemic risk. The Principles do not apply only to global systemically important financial institutions (G-SIFIs) or banks (G-SIBs); however, the ‘significance’ of a financial institution, which is required for their application, should be assessed from the perspective of systemic risk too.
11.60 Another way to look at the Principles is by focusing on institutions that are ‘too-big-to-fail’ (TBTF). Indeed, significant financial institutions are likely to benefit from a government bailout in the case of a crisis, despite recent reforms providing for the resolution of insolvent institutions without recourse to taxpayer money. As argued by Thanassoulis and Tanaka, in the presence of deposit insurance and the implicit possibility of government bailouts risk-taking is subsidized to the point that special rules on incentives may be justified.78 Indeed, there are three types of agency problems which might lead to remuneration contracts that incentivize excessive risk-taking. The first runs between the institution’s executives and shareholders, to the extent that the former may not adequately take the long-term interests of the latter into account. This agency problem can be solved through deferred equity-linked pay, as also foreseen under the international principles. The second runs between the institution’s executives and debt holders: the former may have incentives to take excessive risks at the expense of the latter if the debt market cannot monitor the risks and price them accurately. This agency problem can be solved by either linking variable remuneration to the price of debt or to credit default swap (p. 279) (CDS) premia,79 or by using Contingent Convertible bonds (CoCos) as part of the remuneration.
11.61 The third problem runs between the executives and taxpayers and the deposit insurance fund. Given deposit insurance and/or the implicit possibility of a government bailout, higher risk-taking is not reflected by a proportionally higher cost of funding and risk-taking is effectively subsidized. The mechanisms employed to solve the other two agency problems (such as equity-linked pay and pay in debt instruments) cannot help to solve this problem, because the equity prices may be inflated by the explicit or implicit guarantees on debt (deposit insurance and government bailout) and debt prices may be similarly distorted by those guarantees. Thanassoulis and Tanaka see a possible remedy to this agency problem in the pay-adjustment mechanisms, which are also foreseen by the international principles, known as bonus malus and clawback. Both mechanisms ‘have the effect of putting a fixed monetary value of the executive’s pay at risk’, in other words a ‘penalty’.80 Under malus arrangements, an institution may prevent the vesting of all or part of the deferred remuneration already awarded in relation to risk outcomes or performance. Under clawback clauses, staff members agree to repay variable remuneration that has already been paid by the institution under certain circumstances.81 However, these mechanisms work imperfectly ‘if bankers believe that they could be applied in the event their bank suffers large losses, even if they themselves have conducted appropriate risk management’.82 As a result, the authors suggest ‘to link financial losses imposed on bank executives through bonus malus and clawback to ex ante risk management, rather than making them depend on ex post risk outcomes alone’.83
11.62 These arguments shed light on the international principles from various angles. Firstly, they clarify that the deferment of variable pay and the payment of the same in equity instruments solve the first type of agency problem, but not necessarily the others. Secondly, they explicitly connect some of the FSB Principles—such as malus and clawback—to systemic risk (or TBTF) by showing that the relevant mechanisms put an additional pressure on bank managers to prevent excessive risk-taking by the same. Thirdly, they help understanding as to why the FSB Principles (p. 280) are addressed to significant institutions, with a special focus on ‘large, systemically important firms’. Fourthly, they suggest special care in the design of malus and clawback arrangements, particularly with respect to TBTF institutions, so as to avoid risk-taking which may be excessive from a social perspective.
11.63 The FSB Principles and standards have left some room for proportionality in their implementation to the extent that they concern ‘significant’ institutions. Institutions that are not significant can be regulated under a lighter regime, which could further differentiate depending on the size and business of the relevant firms. In fact, medium-sized institutions are less risky from a systemic perspective and less likely to undergo a government bailout, while small institutions do not individually create systemic problems (which could however be generated by them collectively through herding behaviour). Medium and small institutions are therefore less problematic as to the third type of agency costs highlighted in the previous section. However, the business model of an institution can also be relevant in terms of proportionality. Deposit-taking institutions like commercial banks, for instance, are in principle less prone to compensate their executives and employees through variable remuneration than investment banks and asset managers, which traditionally make wide recourse to bonus payments and other pecuniary incentives.
11.64 Nonetheless, proportionality has been interpreted differently across jurisdictions. US regulation, in particular, is more flexible than CRD IV. The US Federal Supervisory Agencies jointly exercised their mandate under the Dodd-Frank Wall Street Reform and Consumer Protection Act of 21 July 2010 by approving a Proposed Rule on incentive-based compensation arrangements for ‘covered financial institutions’ in February 2011.84 These are institutions under the supervision of the respective Federal Regulator, with total consolidated assets of one billion US dollars or more. As required by Section 956 of the Act, the proposed Rule prohibits ‘excessive’ compensation, that is, compensation that is ‘unreasonable or disproportionate to the services performed by a covered person’. Moreover, compensation should not encourage the taking of ‘inappropriate risks’ by the covered financial institution, by providing executives or employees with incentives that could lead to (p. 281) a ‘material financial loss’ to the institution.85 The use of standards, which are general in character, rather than rules analytically defining the compensation structure, reflects the US regulators’ willingness to keep the needed flexibility in compensation arrangements.86 Nonetheless, specific rules apply to ‘larger covered financial institutions’, such as bank holding companies with consolidated assets of more than fifty billion US dollars, which are required to defer at least 50 per cent of the incentive-based compensation payments to executive officers over a period of at least three years, with the release of the deferred amount to occur no faster than on a pro-rata basis. A ‘malus’ mechanism also applies, in that the deferred amount should be adjusted for actual losses incurred by the institution or other measures of performance during the deferral period.87
11.65 The proposed Rule is currently being reviewed by the Federal Regulators in light of their supervisory experience since 2011.88 Like the 2011 Rule, the new Proposed Rule would apply less prescriptive incentive-based compensation program requirements to the smallest covered institutions and progressively more rigorous requirements to the larger institutions. However, three categories of covered financial institutions (rather than two) would be identified based on average total consolidated assets: Level 1 (greater than or equal to 250 billion US dollars); Level 2 (greater than or equal to 50 billion and less than 250 billion US dollars); Level 3 (greater than or equal to one billion US dollars and less than 50 billion US dollars).89
11.66 The EU legislator has followed a different path. As already stated (Section III.2), institutions should comply with the CRD IV principles in a manner and to the extent that is appropriate to their size; internal organization; and the nature, scope, and complexity of their activities. Moreover, the provisions of this Directive on remuneration should reflect differences between different types of institutions in a proportionate manner, taking into account their size; internal organization; and the nature, scope, and complexity of their activities. However, both the EBA and (p. 282) the European Commission expressed the view that CRD IV does not permit exemptions or waivers to the application of the remuneration principles.90
11.67 The CRD IV requirements concerning the deferral of variable remuneration and the pay-out in instruments have been particularly criticized from the perspective of small institutions and staff with low variable remuneration. This has led most Member States to introduce waivers from such requirements based on proportionality, which were however found legally inadmissible by the EBA and the Commission. Indeed, as recognized by the Commission in a report to the European Parliament, small institutions face high compliance costs related to the fact that they have to make considerable investments in human resources, IT, and advisory services with respect to complex remuneration requirements.91 Moreover, small institutions encounter difficulties in creating appropriate instruments to comply with the pay-out requirement for variable remuneration, often due to their corporate statute or ownership structure.
11.68 As a result, the Commission presented a proposal for a directive amending CRD IV with respect to remuneration requirements too, some of which could be waved by Member States with respect to small and non-complex institutions and to staff members with low variable remuneration.92 Under the proposed new Article 94(3) the requirements as to pay-out in instruments and deferral of variable remuneration shall not apply to:
(a) an institution the value of the assets of which is on average equal to or less than EUR 5 billion over the four-year period immediately preceding the current financial year; (b) a staff member whose annual variable remuneration does not exceed EUR 50.000 and does not represent more than one fourth of the staff’s member’s annual total remuneration.
However, a competent authority could decide that institutions whose total asset value is below the threshold referred to in point (a) ‘are not subject to the derogation because of the nature and scope of their activities, their internal organisation or, if applicable, the characteristics of the group to which they belong’. Similarly, a competent authority may decide that staff members whose annual variable remuneration is below the threshold and are referred to in point (b) ‘are not subject to the derogation because of national market specificities in terms of remuneration practices or because of the nature of responsibilities and job profile of those staff members’.
(p. 283) 11.69 The new proposal would no doubt improve on the present treatment of small institutions, but the comparison with the US regulation of bankers’ pay would still show a big divide, raising the question whether the EU views on proportionality in remuneration matters are appropriate. Building on the US model, particularly on the recent proposals of the Federal Agencies, EU institutions could also be grouped into more than one category and different requirements should apply to them depending on their category. Only the largest institutions should be subject to all CRD IV requirements, which reflect the international principles (save for the bonus cap) and have therefore been designed with respect to significant financial institutions. The threshold of five billion euros still appears too low for identifying significant institutions (which in the United States must possess more than fifty billion US dollars in consolidated assets), while it is appropriate for defining the smallest ones.
1 On the role of culture and motivation in setting managerial incentives, see Shanshan Zhu and Guido Ferrarini, Chapter 16 in this volume.
2 See The High-level Group on Financial Supervision in the EU, chaired by Jacques de Larosière, 25 February 2009 (De Larosiére Report), 30 (the excessive level of remuneration and remuneration structure induced too high risk-taking and encouraged short-termism); Commission Recommendation of 30 April 2009 on remuneration policies in the financial services sector (2009/384/EC),  OJ L120/22 (whilst not the main cause of the financial crisis, inappropriate remuneration practices in the financial services industry induced excessive risk-taking and thus contributed to significant losses of major financial undertakings); Commission Staff Working Document, ‘Corporate Governance in Financial Institutions: Lessons to be drawn from the current financial crisis’, accompanying document to the Green Paper, ‘Corporate governance in financial institutions and remuneration policies’, Brussels 2 June 2010, SEC(2010) 669, 9; ‘A review of corporate governance in UK banks and other financial industry entities’ (the Walker Review), 16 July 2009, 90ff.
3 See the de Larosière Report, 29ff. (corporate governance is one of the most important failures of the present crisis); Commission Green Paper, ‘Corporate governance in financial institutions and remuneration policies’, 2 June 2010, COM(2010) 284 final, 2 (boards of directors rarely comprehended either the nature or scale of the risks they were facing); Basel Committee on Banking Supervision (BCBS), ‘Principles for enhancing corporate governance’, October 2010; OECD, ‘Corporate Governance and the Financial Crisis. Conclusions and emerging good practices to enhance implementation of the Principles (2010).
5 ibid, 12. According to this study, CEOs had substantial wealth invested in their banks, with the median CEO portfolio including stocks and options in the relevant bank worth more than eight times the value of the CEO’s total compensation in 2006. Similar equity holdings should have led CEOs to focus on the long term, avoiding too much risk and excessive leverage for their banks. Instead, the study shows that a bank’s stock return performance in 2007–2008 was negatively related to the dollar value of its CEO’s holdings of shares in 2006, and that a bank’s return on equity in 2008 was negatively related to its CEO’s holdings in shares in 2006.
7 ibid, 260. Indeed, performance-based compensation paid to top executives at Bear Stearns and Lehman Brothers substantially exceeded the value of their holdings at the beginning of the period. Bebchuk et al argue that this provides a basis for concern about the incentives of the two banks’ executives. Rather than producing a ‘tight alignment’ of their interests with long-term shareholder value, the design of performance-based compensation provided executives of the relevant firms with substantial opportunities ‘to take large amounts of compensation based on short-term gains off the table and retain it even after the drastic reversal of the two companies’ fortunes’ (ibid, 274).
9 ibid, 142.
10 ibid, 143, making reference to a previous study by the same author: Ranghuram Rajan, ‘Why Bank Credit Policies Fluctuate: A Theory and Some Evidence’, Quarterly Journal of Economics (1994), 109, 399.
11 Ing-Haw Cheng, Harrison Hong, and Jose Scheinkman, ‘Yesterday’s Heroes: Compensation and Creative Risk Taking’, Journal of Finance (2014) (pay and risk are correlated not because misaligned pay leads to creative risk-taking; rather, as principal-agent theory predicts, riskier firms have to pay more total compensation to provide the same incentives for a risk-averse manager than less risky firms).
12 Rajan, n 8, 144–5.
15 Guido Ferrarini and Maria Cristina Ungureanu, ‘Economics, Politics, and the International Principles for Sound Compensation Practices: An Analysis of Executive Pay at European Banks’, Vanderbilt Law Review (2011), 64, 431–502.
16 Blinder, n 14, 84 and 284.
17 Luc Laeven and Lev Ratnovski, ‘Corporate Governance of Banks and Financial Stability’, Vox, 21 July 2014, available at http:voxeu.org (accessed 30 September 2018).
18 See, however, for heavy criticism of post-crisis regulation, Luca Enriques and Dirk Zetzsche, ‘Quack Corporate Governance, Round III? Bank Board Regulation Under the New European Capital Requirement Directive’, ECGI Law Working Paper No 249/2014.
19 See the studies by Andrea Beltratti and René Stulz, ‘The Credit Crisis around the Globe: Why Did Some Banks Perform Better?’, Journal of Financial Economics (2012), 105, 1–17and Renée Adams, ‘Corporate Governance and the Financial Crisis’, International Review of Finance (2012), 12, 7–38.
20 See Laeven and Ratnovski, n 17 (better corporate governance, by itself, is unlikely to make banks safer); Marco Becht, Patrick Bolton, and Ailsa Roell, ‘Why Bank Governance is Different’, Oxford Review of Economic Policy (2012), 27, 437–63, 445 (not only shareholders, but also depositors, other creditors, transaction counterparties, and the taxpayers are at risk from banks’ activities; therefore, not just the interests of shareholders, but also those of other constituencies should be protected).
21 See Lawrence White, ‘Corporate Governance and Prudential Regulation of Banks: Is there Any Connection?’, in James Barth, Chen Lin, and Clas Wihlborg (eds), Research Handbook for Banking and Governance, Elgar, 2012, 344–59, 344 (senior managers who properly respond to the interest of shareholders ought—unless restrained by the debt holders or by prudential regulators—to be undertaking activities that might otherwise appear to be excessively risky); Hamid Mehran, Alan Morrison, and Joel Shapiro, ‘Corporate Governance and Banks: What Have We Learned from the Financial Crisis?’, Federal Reserve Bank of New York, Staff Report No 502/2011.
22 See Andrew Ellul and Vijay Yerramilli, ‘Stronger Risk Controls, Lower Risk: Evidence from U.S Bank Holding Companies’, Journal of Finance (2013), 68, 1757–803; Senior Supervisors Group, ‘Observations on risk management practices during the recent market turbulence, 2008.
24 See, for the role of pay disclosure and its regulation, Guido Ferrarini and Maria Cristina Ungureanu, ‘Executive Remuneration. A Comparative Overview’, in Jeffrey Gordon and Georg Ringe (eds), Oxford Handbook of Corporate Law and Governance, Oxford University Press, 2018.
25 See Ferrarini and Ungureanu, n 15.
26 See n 4.
27 FSB, ‘Thematic Review on Compensation: Peer Review Report’ 10-11, 2010, available at http://www.fsb.org/wp-content/uploads/r_100330a.pdf, accessed 30 September 2018.
29 Commission Recommendation on remuneration policies in the financial sector, C (2009) 3159, (April 2009). In June 2010 the Commission also published a Green Paper on corporate governance in financial institutions and remuneration policies, which analysed the deficiencies in corporate governance arrangements in the financial services industry and proposed possible ways forward; Commission Green Paper on corporate governance in financial institutions and remuneration policies (May 2011).
31 Commission Report on the application by Member States of the EU of the Commission 2009/384/EC Recommendation on remuneration policies in the financial services sector; CEBS, ‘Report on national implementation of CEBS High-level principles for Remuneration Policies’ (June 2010).
32 Directive 2010/76/EU of the European Parliament and of the Council of 24 November 2010 Amending Directives 2006/48/EC and 2006/49/EC As Regards Capital Requirements for the Trading Book and for Re-Securitisations, and the Supervisory Review of Remuneration Policies,  OJ L329/3.
33 Directive 2013/36/EU of the European Parliament and of the Council of 26 June 2013 on access to the activity of credit institutions and the prudential supervision of credit institutions and investment firms, amending Directive 2002/87/EC and repealing Directives 2006/48/EC and 2006/49/EC, OJ , L176/338.
36 The impact of CRD III on remuneration practices has been substantial, even though empirical research shows that changes also occurred before the Directive’s adoption: Roberto Barontini et al, ‘Directors’ Remuneration before and after the Crisis: Measuring the Impact of Reforms in Europe’, in Massimo Belcredi and Guido Ferrarini (eds), Boards and Shareholders in European Listed Companies, Cambridge University Press, 2013), 251–314.
37 See, for the author’s previous work on this topic, Guido Ferrarini, ‘Regulating Bankers’ Pay in Europe: The Case for Flexibility and Proportionality’, in Festschrift für Theodor Baums, Mohr Siebeck, 2017, I, 401–16; Guido Ferrarini, ‘CRD IV and the Mandatory Structure of Bankers’ Pay’, ECGI Law Working Paper 289/2015.
38 Commission Delegated Regulation (EU) No 604/2014 of 4 March 2014, supplementing Directive 2013/36/EU of the European Parliament and of the Council with regard to regulatory technical standards with respect to qualitative and appropriate quantitative criteria to identify categories of staff whose professional activities have a material impact on an institution’s risk profile.
44 EBA, n 40, 13.
49 See European Parliament, ‘Report on the proposal for a directive of the European Parliament and of the Council on the access to the activity of credit institutions and the prudential supervision of credit institutions and investment firms etc’. (A 7-0170/2012) of 30 May 2012, Article 90(1)(f) (‘institutions shall set the appropriate ratios between the fixed and the variable component of the total remuneration where the variable component shall not exceed one time the fixed component of the total remuneration’).
50 See Kate Allen, ‘Goldman top bankers lead UK pay league with £3m packages’, Financial Times, 1 January 2015: in 2013 Goldman Sachs paid its senior staff bonuses worth 5.5 times their average salary, the highest ratio among the five top banks; Citigroup was the only top bank consistent with the EU cap in 2013, paying staff bonuses averaging 1.6 times salary.
52 Opinion of Advocate General Jääskinen delivered on 20 November 2014, Case C-507/13 United Kingdom of Great Britain and Northern Ireland v European Parliament and Council of the European Union, para 120.
… the material that was available to the decision-makers clearly demonstrated that restricting incentives to excessive risk taking by the management and staff of financial institutions was likely to reduce such risk taking and, in consequence, any risk for the stability of financial markets following therefrom. Under such circumstances, the question where and by whom the concrete limits to such initiatives were to be set concerned, in my opinion, the degree of regulation that was appropriate. This has clearly involved economic and political choices. However, such choices need to have been manifestly inappropriate before the legislative measure can be annulled.
58 Richard Posner, A Failure of Capitalism. The Crisis of ‘08 and the Descent into Depression, Harvard, 2009, 297; similar comments in Sanjai Bhagat, Brian Bolton, and Roberto Romano, ‘Getting Incentives Right: Is Deferred Bank Executive Compensation Sufficient?’, ECGI Law Working Paper No. 241/2014, February 2014, 35; and Rui Albuquerque, Luís Cabral, and José Corrêa Guedes, ‘Relative Performance, Banker Compensation, and Systemic Risk’, ECGI Finance Working Paper No 490/2016, who argue that, without a regulatory constraint on relative performance evaluation (RPE), some of the restrictive measures on executive compensation that are found in regulation are ineffective in reducing systemic risk.
59 See ‘Opinion of the European Banking Authority on the application of Directive 2013/36/EU (Capital Requirements Directive) regarding the principles on remuneration policies of credit institutions and investment firms and the use of allowances’, EBA/Op/2014/10, 15 October 2014, 2. See also the EBA Report, ‘On the application of Directive 2013/36/EU (CRD) regarding the principles on remuneration policies of credit institutions and investment firms and the use of allowances’, 15 October 2014.
60 EBA, n 41, 2.
62 Kevin Murphy, ‘Regulating Banking Bonuses in the European Union: a Case Study in Unintended Consequences’, European Financial Management (2013), 19, 631–57. A slightly different position is taken by John Thanassoulis, ‘The Case for Intervening in Bankers’ Pay’, Journal of Finance (2012), 67, 849–95, who argues that a modest bonus cap—‘modest in that it is close to the current equilibrium rate of bonuses’—lowers default risk; however, stringent bonus caps, such as those introduced by CRD IV, enhance default risk, and are value destroying.
63 The case of Deutsche Bank is interesting. At the 2014 AGM the bank proposed to raise the maximum bonus senior managers can receive to twice their fixed annual salary, double the current level. Deutsche Bank officials said the move was necessary so that the bank could comply with European rules on pay, while also competing for staff with US rivals. They said that if the bonus increase was rejected, the bank would need to raise base salaries to retain top talent. But opposition to the proposals (which was however approved) was mounting from shareholder groups who argued that the payment was excessive and fostered improper behaviour: see Eyk Henning, ‘Some Deutsche Bank Shareholders Plan to Protest Bonus Proposal’, Wall Street Journal, 16 May 2014, available at http://online.wsj.com/articles/SB10001424052702304908304579566140929304688 accessed 24 November 2014.
64 See also Bhagat, Bolton, and Romano, n 58, 36 (the EU cap will make pay even less sensitive to performance than it was before the crisis, which is the opposite of what is desirable in an incentive compensation plan).
65 Murphy, n 62.
67 Report from the Commission to the European Parliament and the Council, Assessment of the remuneration rules under Directive 2013/36/EU and Regulation (EU) No 575/2013, Brussels, 28 July 2016. See also Commission Staff Working Document, Detailed assessment of the remuneration rules under Directive 2013/36/EU and Regulation (EU) No 575/2013, Brussels, 28 July 2016.
68 Report, n 67.
71 Study on the remuneration provisions applicable to credit institutions and investment firms prepared by the Institute for financial services for European Commission’s DG JUST (Contract JUST/2015/MARK/PR/CIVI/0001), Final Report, January 2016, 93.
73 Report, n 67, 11.
75 See Michele Siri, ‘Corporate Governance of Insurance Firms after Solvency II’, ICIR Working Paper Series No 27/2017, Goethe University Frankfurt, International Center for Insurance Regulation (ICIR), available at https://ideas.repec.org/p/zbw/icirwp/2717.html, accessed 30 September 2018.
76 See Lodewijk van Setten and Danny Busch (eds), Alternative Investment Funds in Europe: Law and Practice Oxford University Press, 2014; Niamh Moloney, EU Securities and Financial Markets Regulation, Oxford University Press, 2014, 211, noting that it was the AIFM Directive which influenced the reform of the UCITS Directive as to remuneration.
77 See Jens-Heinrich Binder, ‘Governance of Investment Firms under MiFID II’, in Danny Busch and Guido Ferrarini (eds), Regulation of the EU Financial Markets: MiFID II and MiFIR, Oxford University Press, 2017, 72.
79 Alex Edmans and Qi Liu, ‘Inside Debt’, Review of Finance (2011), 15, 75–102; Patrick Bolton, Hamid Mehran, and Joel Shapiro, ‘Executive Compensation and Risk Taking’, Review of Finance (2015), 19, 2139–218.
80 ibid, 24.
82 See Misa Tanaka and John Thanassoulis, ‘Fixing bankers’ pay: punish bad risk management, not bad risk outcomes’, Bank Underground, 7 December 2015, available at https://bankunderground.co.uk/2015/12/07/fixing-bankers-pay-punish-bad-risk-management-not-bad-risk-outcomes/, accessed 30 September 2018.
84 Section 956 of the Dodd-Frank Wall Street Reform and Consumer Protection Act requires the Agencies (the Board of the Governors of the Federal Reserve System, the Federal Deposit Insurance Corporation, the National Credit Union Association, the Office of the Comptroller of the Currency, the Federal Housing Finance Agency, and the Securities and Exchange Commission) to jointly prescribe regulations or guidelines with respect to incentive-based compensation practices at certain financial institutions (referred to as ‘covered financial institutions’). In April 2011, the Agencies published a joint notice of proposed rule making that proposed to implement section 956 (2011 Proposed Rule).
85 See, for a short comment of the Rule from a comparative perspective, Guido Ferrarini and Maria Cristina Ungureanu, ‘Bankers’ Pay after the 2008 Crisis: Regulatory Reforms in the US and the EU’, Zeitschrift fur Bankrecht und Bankwirtschaft (2011), 23, 418–30, 422ff.
87 Moreover, if the covered financial institution has consolidated assets of 50 billion US dollars or more, the board of directors or a board committee shall identify those covered persons (other than executive officers) who individually have the ability to expose the institution to possible ‘substantial’ losses.
88 See the Board of the Governors of the Federal Reserve System, Incentive-based Compensation Arrangements, Notice of Proposed Rulemaking and Request for Comment, available at https://www.federalreserve.gov/newsevents/pressreleases/files/bcreg20160502a2.pdf, accessed 30 September 2018.
91 Report, n 67, 8.