- Email: [email protected]
Systemic risk, governance and global financial stability
Journal of Banking & Finance xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Journal of Banking & Finance journal homepage: www.elsevier.com/locate/jbf
Systemic risk, governance and global financial stability Luci Ellis a, Andy Haldane b, Fariborz Moshirian c,⇑ a
Reserve Bank of Australia, Sydney, Australia Bank of England, London, United Kingdom c Institute of Global Finance, University of New South Wales, Sydney, Australia b
a r t i c l e
i n f o
Article history: Received 31 May 2013 Accepted 9 April 2014 Available online xxxx JEL classification: G15 G25 Keywords: Systemic risk Bank governance Financial stability
a b s t r a c t The paper argues that while attempts have been made to reform four of the five key pillars of banks’ operation, (i.e. competition, resolution, supervisory, and auditing and valuation policies), less attention has been paid to the role of bank governance and systemic risk, despite a strong link between governance and risk-taking. The paper offers four solutions to strengthen bank governance. First, the regulatory capital base of banks could be increased. Second, the compensation structure of managers could be reformed. Third, effort could be focussed on creating and implementing resolution regimes which offer the credible prospect of ‘‘bailing-in’’ creditors in the event of stress and fourth solution is to reform the structure of company law– for example, by extending control rights beyond shareholders. Furthermore, the paper argues that given the diversity of the whole financial system, it is expected that the risks individual financial institutions face are also diverse. It cannot be assumed that the appropriate capitalisation is constant across all risks. While leverage ratios are a useful backstop measure and guard against potential gaming of risk-weights, their appropriate role is as a backstop. The diversity within the financial system also supports the fact that a single measure of systemic risk is unlikely to be universally applicable, nor is a single instrument of financial stability policy. Ó 2014 Published by Elsevier B.V.
1. Introduction The recent global financial crisis highlighted the importance of addressing some of the incomplete financial reforms. One of these unaddressed reforms has been the contribution of systemic risk in destabilising financial markets. Recently, as reported by Bisias et al. (2012), international institutions such as the IMF, the BIS and the SFSB stated that systemic risk can be defined as a ‘‘risk of disruption to financial services that is caused by an impairment of all or parts of the financial system and has the potential to have serious negative consequences for the real economy.’’ One of the policy issues is how one might operationalize a policy to limit systemic risk given that a risk is inherently unobservable – only outcomes are observable. Unfortunately, there is no single operational definition of systemic risk so far. Bisias et al. (2012) identified 31 different measures of systemic risk, emphasising a range of aspects of systemic risk and potential channels to financial distress. Indeed, as Bisias et al. (2012) point out, it is probably not desirable to have
⇑ Corresponding author. Tel.: +61 612 93855859; fax: +61 61293854763. E-mail address: [email protected] (F. Moshirian).
a single measure of systemic risk as the focus of policy, as this may result in a ‘Maginot Line’ situation where vulnerabilities building in other parts of the financial system are missed. The recent global financial crisis has encouraged both policy makers and researchers to find ways by which one could mitigate the impact of systemic risk. To this end, a number of attempts have been made to identify policies, including forward-looking policies, that could potentially address systemic risk. As part of the efforts to address some of the underlying causes of the recent global financial crisis, the Financial Stability Board (‘FSB’) has been asked to act as the international standard-setter and attempt to address and coordinate issues underlying systemic risk. One of the key tasks of the FSB was to deal with the contributions of large financial institutions to global systemic risk. To this end, the FSB has identified global systemically important financial institutions (‘G-SIFIs’), and more recently domestic systemically important banks (‘D-SIBs’). These multinational and large national financial institutions have attracted more policy attention, partly due to their size, complexity, interconnectedness and their contributions to national and global financial systems (Moshirian, 2011, 2012). As part of this process and excessive risk taking by some of these institutions, themes related to too-big-to-fail (‘TBTF’) have also emerged for debate and analysis.
http://dx.doi.org/10.1016/j.jbankfin.2014.04.012 0378-4266/Ó 2014 Published by Elsevier B.V.
Please cite this article in press as: Ellis, L., et al. Systemic risk, governance and global financial stability. J. Bank Finance (2014), http://dx.doi.org/10.1016/ j.jbankfin.2014.04.012
2
L. Ellis et al. / Journal of Banking & Finance xxx (2014) xxx–xxx
Another major development in recent years has been a focus on developing macroprudential frameworks to complement the traditional microprudential approach to regulation. There is also a desire on the part of the policy makers and the market participants to see the financial system sharpen its ability to apply effective macroprudential policies without stifling economic growth and innovation. However, the identification and monitoring of systemic risk and optimal macroprudential policies are still in the early stage of development. In search for global financial stability, one area that has received less attention in recent years are the issues related to bank governance. More discussion and analysis of the role of governance within the operation of financial institutions and also the role of governance in contributing to global financial stability would be an essential part of any policy package aiming to address a number of the underlying causes of financial fragility over time. The paper argues that while there have been analyses of banks competition policies, resolution policies, supervisory policies and auditing and valuation policies, less attention has been paid to the role of bank governance and systemic risk, despite a strong link between governance and risk-taking. This is puzzling (Jensen and Meckling, 1976). Recent empirical evidence suggests this link may be especially strong for financial firms (Beltratti and Stulz, 2009; Ferreira et al., 2012; Laeven and Levine, 2009). Certainly, the global financial crisis unearthed bank governance failures operating at multiple levels: managers failing to control risk-takers, boards failing to control managers and investors failing to discipline either managers or boards. The purpose of this paper is to discuss some of the issues related to systemic risk, governance and financial stability. To this end, Section 2 of the paper discusses issues related to systemic risk, the identification of G-SFIS and D-SIBs and the recent policies related to macroprudential policies and the importance of implementing globally a number of regulatory policies that have been proposed by various international institutions; Section 3 analyses issues related to measurement of systemic risk. It argues that the diversity within the financial system also supports the fact that a single measure of systemic risk is unlikely to be universally applicable. Furthermore, this section discusses how, due to the diversity of the whole financial system, the risks individual financial institutions face are also diverse. The section also discusses why one may not assume that the appropriate capitalisation is constant across all risks. Section 4 discusses bank governance, including how this important aspect of global financial stability has been overlooked for vigorous debate and analysis in recent times and the importance of having specific policies to address the current structural weaknesses of bank governance. This section offers solutions regarding how to strengthen bank governance, including the regulatory capital base of banks, could be increased, the compensation structure of managers could be reformed and efforts could be made by putting resolution regimes which offer the credible prospect of ‘‘bailing-in’’ creditors in the event of stress, into place and Section 5 concludes.
have been identified by the FSB that could qualify certain financial institutions to be part of G-SIFIS. These attributes include: size, lack of substitutability and interconnectedness. In addition, the Basel Committee for Banking Supervision (Basel Committee) identified cross-jurisdictional activity and complexity as other attributes of financial institutions which could fall into the category of G-SIFIs (see BIS, 2013). As of November 2013, there are 29 designated G-SIFIs by the FSB. As part of an overall strategy to ensure more stable G-SIFs, these institutions are now required to maintain greater loss absorbency capabilities under Basel III. As highlighted by the FSB’s various publications, this additional requirement, which applies in addition to the Basel III capital requirements, is intended to reduce the ‘‘cross-border negative externalities’’ of the global financial system and to reduce systemic risk. Some regional and domestic banks could also pose systemic risk to the national and regional financial systems and national economies, (over and above those institutions identified as G-SIFIs). To this end, the Basel Committee has developed a framework and asked local authorities to identify whether a large domestic bank is systemically important from a domestic perspective (domestic systemically important bank (D-SIB). A number of national authorities have already identified their D-SIBs. One should also note that there are also some regional systemically important banks that could be operating in different continents whose operations could pose systemic risk. As part of an overall strategy to increase global financial stability, international institutions such as the FSB now have a strategy in place to ensure that national and international policy makers work together and ensure that the G-SIFIs and D-SIBs are better supervised and these institutions, as state above, are required to put aside more capital as part of their operation. This is because the current strategy is to insulate the global financial system from systemic shocks on top of the Basel III capital and liquidity requirements.1 As Moshirian (2012) stated, the Great Depression in the 1930s created incentive for authorities to collect national economic data so policy makers could measure the magnitude of economic downturn in the economy accurately. This attempt in the 1930s led to the emergence of relevant data that are used to generate what is now referred to as Gross Domestic Products (GDP). Similarly the recent global financial crisis has created a new impetus to policy makers and market participants to improve and also share some financial data, particularly when the financial market is becoming increasingly interconnected. The attempt by the FSB, in collaboration with central banks and supervisors, to create a mechanism that could facilitate the international sharing of firm-level data on systemically important financial institutions, is yet another step in the process of facilitating globalisation and another attempt to enhance the capacity of the global financial system to be more informed about the overall state of large financial institutions and the nature of their aggregate risk taking and business models.
2. Large financial institutions and propagation of systemic risk
The FSB has been working on issues related to macroprudential policies over the last few years. While a number of proposals have emerged as part of the tools available to enhance the effectiveness of policy recommendations with respect to macroprudential strategies, this area of policy development is still in its infancy and requires more work (Arnold et al., 2012). Nevertheless, some good progress has been made in recent times including the work of the FSB itself. At the present time, there are four main areas of
2.1. Identification of systemically important financial institution According to Financial Stability Board, a financial institution could be defined as a G-SIFI, if its failure or fragility could impact other financial institutions, the wider financial system and the domestic and international economies (FSB, 2011a, 2011b). Given the size and complex nature of these large financial institutions with other institutions, their failure or fragility could negatively impact the overall global financial system. A number of attributes
2.2. Macroprudential policies and a global approach to financial stability
1 This section of the paper benefited from a number of information and issues that can be found in the FSB publications and website.
Please cite this article in press as: Ellis, L., et al. Systemic risk, governance and global financial stability. J. Bank Finance (2014), http://dx.doi.org/10.1016/ j.jbankfin.2014.04.012
L. Ellis et al. / Journal of Banking & Finance xxx (2014) xxx–xxx
development proceeding under the supervision of the FSB. The first is developing measures to identify and monitor systemic risk, including both regulated banking and shadow banking. As the FSB (2011a) stated, despite some progress in collecting new data, the challenge remains to be able to identify tools and instruments that could guide forward policy making and ensure its credibility. Therefore, to be able to identify and contain systemic risk through credible instruments and tools remains the second area of development. According to the FSB (2011a), there are currently a range of tools used by various countries to address systemic risk, and they fall roughly into three categories. These categories are: (1) tools to address financial stability risks arising from rapid credit expansion; (2) tools to address amplification mechanisms of systemic risk such as leverage and maturity mismatches and (3) tools to limit spillover effects from the failure of SIFIs. To be able to develop institutional arrangements for macroprudential regulators at both domestic and regional levels will be the third area of development. Not surprisingly, the fourth area identified by the FSB is the importance of achieving regional and international cooperation that is essential to address challenges of systemic risk that often become supernational in their dimension. The increasing interconnectedness of global finance implies that some of the financial risk may well spill over to other parts of the world. One of the challenges of our generation will remain how to ensure an effective global financial system that has the full support of national authorities and the private sector, that in turn could minimise the opportunity for regulatory arbitrage. Furthermore, consistency of action by all national authorities could also generate more confidence in the collective implementation of some of the current and future global agreements. Obviously, amongst the key protagonists of the financial market (i.e. policy makers, regulators, researchers and market participants) there should be constant dialogue, consultation, and reflection about some of these global agreements. It is noteworthy that the BIS is also keen to see certain standards globally accepted and implemented. For instance, Caruana (2013a) stated, if we do not control fragmentation tendency and also regulatory arbitrage, we could lose the benefits of globalisation and the pace of process of financial globalisation that has accelerated due to technological changes and the increasing interdependence of national economies. Caruana (2013a) also stated that fragmentation could prevent the proper transmission of monetary policy, particularly in the Euro-zone area., Furthermore, in addition to a structured supervisory framework (e.g. Peer reviews), greater cooperation is needed between domestic and foreign policy makers, where each are cooperating in order to resolve issues. Furthermore, consistency must be achieved across the international stage. Rules and minimum standards should not be different based on the basis of a bank’s nationality. As a result, if some banks are held to a lower standard than the internationally agreed rules, this will result in an uneven playing field. In this context, as the FSB (2011a) stated that there is a need to establish strong mechanisms to ensure consistent action by countries to contain risk, and to ensure national regulatory frameworks are consistent. With regard to regional cooperation, it should be noted that regional mechanisms, such as the college of supervisors, may work relatively well for some of the G-SIFs and also regional and domestic SIFS in the EU due to the high level of financial integration. However, such regional cooperation and information sharing may not necessarily work in other parts of the world such as Asia, where the political climate is different and the level of national autonomy much higher. One cannot dismiss a number of regional architectures that have been erected to bring the economies of the Asia Pacific region closer. For instance APEC is designed to promote free trade, while ASEAN and ASEAN plus 3 are more focused on free
3
trade and, economic and financial integration within the member countries of such blocks. Similarly ASEAN plus 6 and the East Asia Summit also have some economic, trade and finance objectives that are ultimately designed to bring a large number of countries economically and financially closer together. Given the importance of Asia and the massive amount of foreign capital flowing to this region and the increasing volume of trade, such cooperation and information sharing amongst policy makers with respect to the operations of some of the G-SIFs and regional and domestic SIFs will be crucial over time. Regarding the supervision and coordination of G-SIFs and some of the regional and domestic SIFs, Caruana (2013b) has also stated ‘‘this kind of supervision requires not only a new and different kind of expertise but, even more importantly, the capacity and willingness to act under significant uncertainty. These include the ability to conduct group-wide consolidated supervision and to challenge banks’ business models, their corporate strategy, governance, risk profiles, ROE targets and capital plans. Also necessary will be the willingness to exercise judgment and to act pre-emptively under conditions of uncertainty’’.
3. Considerations for modelling policies to address systemic risk 3.1. Systemic risk measurement Post-crisis, a large literature has started to build up, which attempts to incorporate macroprudential policy into canonical macro-models, generally of the micro-founded Dynamic Stochastic General Equilibrium modelling (DSGE) type or variants of these. These papers have therefore had to add some kind of financial sector, so that the policy has something to operate upon. This exercise necessarily requires taking a position on asset and credit dynamics, but these relationships are generally less well understood than the inflation and real-side dynamics that are more relevant for analysis of monetary policy questions. One of the problems with this line of research is that – unlike inflation – there is little welfare-theoretic basis for using typical measures of asset prices and credit as targets of macroprudential or any other policy. In the monetary policy literature, inflation (or inflation variability) maps into social welfare, either directly because agents are presumed to dislike inflation, or indirectly because price volatility is costly for output. For example, earlier generations of literature on monetary policy assumed an objective function for the policy maker that was a weighted average of inflation volatility and output volatility; this formulation, while not exactly corresponding to social welfare, turns out to bear close resemblance to it (Woodford, 2010). In contrast, asset prices and credit do not enter into welfare functions in a way that suggests that their growth should be resisted by some arm of public policy. To the extent that wealth enters into welfare, one could argue that policies that increase asset prices are desirable. Indeed, some micro-founded models produce the result that under-borrowing, rather than over-borrowing, is the more pertinent issue, and that macroprudential policies could in fact be welfare-reducing (Benigno et al., 2013). More fundamentally, in the presence of credit constraints coming from information asymmetries, it is far from clear that ‘normal’ levels of credit are in fact optimal. This observation in no way contradicts the empirical observation that credit booms can be harmful for the subsequent evolution of output, because they can spark a financial crisis (Dell’Ariccia et al., 2012). Even the internationally agreed prudential rules acknowledge this: rather than mandating a particular level of credit or asset prices or growth rate of these series, the Basel Committee’s guidance on application of the countercyclical capital buffer promoted
Please cite this article in press as: Ellis, L., et al. Systemic risk, governance and global financial stability. J. Bank Finance (2014), http://dx.doi.org/10.1016/ j.jbankfin.2014.04.012
4
L. Ellis et al. / Journal of Banking & Finance xxx (2014) xxx–xxx
(but did not require) the use of a ‘buffer guide’ derived from deviation from trend of the ratio of credit to GDP (BCBS, 2010). There are many legitimate reservations about this guide and how it is formulated (see, for example Edge and Meisenzahl, 2011 and Box C in APRA and RBA, 2012), but the important point in this context is that it is understood even by the proponents of the buffer guide that there is no welfare-theoretic basis for a fixed level of the ratio of credit to GDP. This ratio can trend over long time horizons, and it can have different sustainable levels in different countries, depending on such factors as the degree of financial development and demographic structure. Even substantial deviations from trend do not necessarily warrant a costly policy to prevent them. Dell’Ariccia et al. (2012) found that around one in three credit booms were followed by a banking crisis, which necessarily implies that around two in three were not. This disconnect between these variables and welfare-theoretic measures indicates that there is a deeper goal being pursued. What financial stability policies – whether labelled ‘macroprudential’ or otherwise – are intended to do is limit systemic risk. Systemic risk is generally defined as the risk that the financial system might experience a generalised collapse or other form of distress. The reason for limiting this risk is to avoid the attendant harm on the real economy. The ultimate target is therefore output. In other words, asset prices and credit growth are useful information variables and possibly even intermediate targets of policy, much as monetary aggregates were in an earlier era. They are not, however, ultimate targets in same the way as inflation and output are targets of macroeconomic policy. The above line of argument naturally raises the question of how one might operationalize a policy to limit systemic risk given that a risk is inherently unobservable – only outcomes are observable. Unfortunately, there is no single operational definition of systemic risk and possibly never will be. Bisias et al. (2012) identified 31 different measures of systemic risk, emphasising a range of aspects of systemic risk and potential channels to financial distress. Among these are measures that capture deviations from trend in asset prices or credit, much as Dell’Ariccia et al. (2012) do, and measures that stress test the effects of large declines in these and related variables. Other measures focus on the propagation of distress from one financial institution to another, such that an idiosyncratic failure ends up causing a systemic problem. Dimensions of vulnerability that are emphasised by the various measures include correlated exposures, network effects where the default of one party stresses its creditors, illiquidity and leverage. Indeed, as Bisias et al. (2012) point out, it is probably not desirable to have a single measure of systemic risk as the focus of policy, as this may result in a ‘Maginot Line’ situation where vulnerabilities building in some other part of the financial system are missed. This has obvious implications both for the approach to actual financial stability policy, and how that policy is captured in theoretical models. In either situation, a single instrument, single target analogue to monetary policy is unlikely to be able to generate useful results. 3.2. Systemic risk and diversity As is well known, the financial system performs a number of functions that are essential to the workings of a modern economy. As well as intermediating between savers and borrowers, the financial sector provides a conduit for payments and enables risk transfer. These are a diverse set of functions and it should therefore be no surprise that the activities of the financial system give rise to a diverse set of risks. For example, banks and similar entities intermediate between borrowers and savers; fund managers also help savers deploy their funds, but without interposing their own
creditworthiness in place of that of borrowers or equity issuers, as is the case for banks. Insurance companies allow customers to transfer diversifiable risks to them, such as property losses from fire, flood and theft, as well as life insurance. Many of the insured risks that the financial system helps to transfer are themselves risks created within the financial system. This is consistent with the principle that financial losses are easier or more attractive to insure because they can be fully compensated by a financial payout. (An example of this is that people are more likely to insure their homes against loss from fire or flood than they are to insure their pets’ lives against the same risks: a financial payout may not be seen as a true compensation for such a loss.) Even within the banking industry, banks serve different functions and therefore face different risks. Another way of saying this is that banks’ business models can vary. One way to think about bank business models is as a spectrum with investment banks at one end, commercial or retail banks at the other, and so-called ‘universal’ banks in the middle. Those labels correspond to different balance sheet structures: investment banks have larger holdings of securities and other trading assets, while commercial banks’ assets are mainly loans. As Fig. 1 shows, the importance of trading assets varies greatly within the largest global banking groups, and even within the 29 globally systemically important banks (G-SIBs) identified by the Basel Committee on Banking Supervision and the Financial Stability Board (FSB, 2013a, 2013b, 2013c), shown here as darker bars. While there are a cluster of banks with investment-banking business models in this group of G-SIBs, G-SIBs are represented across the whole range of observed balance sheet structures. In contrast, the four large Australian banks, shown as lighter bars in Fig. 1, for example, are all clearly commercial banks, with relatively little trading business. Several conclusions can be reasonably drawn from these observations. One is that if even the banking industry, let alone the whole financial system, is relatively diverse, it should be expected that the risks individual institutions face are also diverse. It cannot be assumed that the appropriate capitalisation is constant across all risks. This is a powerful argument against a single leverage ratio as the predominant instrument of prudential regulation. Even though there are concerns about the complexity of the current prudential framework, this suggests that risk-weighting is and should be here to stay. While leverage ratios are a useful backstop measure and guard against potential gaming of risk-weights, their appropriate role is as a backstop. The diversity within the financial system also supports the point made by Bisias et al. (2012) and mentioned earlier, that a single measure of systemic risk is unlikely to be universally
Share of assets, June 2013 % 60
G-SIBs Four major Australian banks
% 60
45
45
30
30
15
15
0
Banking groups
0
Fig. 1. Trading assets and securities of the largest banking groups (includes derivative assets; sample of 100 banking groups; latest available ratios have been used where June 2013 data are unavailable). Sources: FSB; RBA; SNL Financial; The Banker; banks’ annual and interim reports.
Please cite this article in press as: Ellis, L., et al. Systemic risk, governance and global financial stability. J. Bank Finance (2014), http://dx.doi.org/10.1016/ j.jbankfin.2014.04.012
L. Ellis et al. / Journal of Banking & Finance xxx (2014) xxx–xxx
applicable, and neither is a single instrument of financial stability policy. It will not be appropriate to conceive of macroprudential policy (or financial stability policy more generally) in the single instrument, single target framework that is generally used to analyse monetary policy. Different aspects of prudential regulation will bear more effectively on particular risks and aspects of systemic risk than others. Rather, issues of macroprudential policy may be best thought of as a design problem in calibrating the entire prudential framework. Separating out particular aspects of prudential regulation into specific macroprudential ‘tools’, possibly with different governance from the prudential framework more broadly, may lead to coordination problems. A more subtle issue arises in considering the niches that banks with different business models serve. Although there has been some debate about the social utility of some financial activity, and attempts such as the Volcker Rule to limit proprietary trading by banks, it is not the case that the optimal share of trading assets in bank balance sheets is zero. Neither is it the case that the optimal share of these assets would be the same across all banks. It depends on their business models and the particular risks they take on. Commercial banks are principally involved in intermediation between savers and borrowers from the non-financial sectors. In doing so, though, they may incur risks that may be prudently laid off to other parts of the financial sector that may be able to diversify those risks more effectively. Among these risks are interest rate risk and FX risk: even quite plain-vanilla retail banking can involve an array of market risks depending on how diversified the bank’s funding base is and how global the business of the bank’s commercial customers. At least some of the trading assets of banks with more investment-banking business models arise through their interbank exposures built up as counterparties to commercial banks. These counterparty relationships can cross borders. Thus even in a national financial system like Australia’s, where the domestically owned banks are primarily commercial or retail oriented, the gamut of financial services can be provided, and risks transferred, because there are other, generally foreign banks, that can serve as counterparties to local banks’ hedging activity. This implies that, although interconnectedness within the financial system can increase systemic risk, there is a limit below which this interconnectedness cannot prudently fall. In particular, although cross-border contagion is frequently identified as a risk to national financial stability, and inter-bank borrowing often a precursor to financial crisis, a position of financial autarky is not the desirable equilibrium point, either. This has implications for measures of systemic risk and crisis vulnerability, which should perhaps not be seen as linear indicators.
4. Bank governance and systemic risk 4.1. Five planks of effective reform There are five key planks of any well-defined regulatory regime: entry – that is, competition policy; exit – that is resolution policy: regulation – that is, supervisory policy; accounting – that is, auditing and valuation policy; and governance. In the light of the crisis, there have been intense efforts to reform the first four of these pillars. For example, new regimes are being put in place for competition (e.g. Parliamentary Commission on Banking Standards, 2013), regulation (e.g. Basel Committee on Banking Supervision, 2010a, 2010b), resolution (e.g. Financial Stability Board, 2011a, 2011b) and accounting (e.g. International Accounting Standards Board, 2013) within banking firms, nationally and internationally. Yet the fifth pillar – governance – has been largely untouched. That is puzzling. Academic theory has long suggested a strong link between governance and risk-taking (Jensen and Meckling, 1976).
5
And recent empirical evidence suggests this link may be especially strong for financial firms (Beltratti and Stulz, 2009; Ferreira et al., 2012; Laeven and Levine, 2009). Certainly, the financial crisis unearthed bank governance failures operating at multiple levels: managers failing to control risk-takers, boards failing to control managers and investors failing to discipline either managers or boards. Governance is also a potentially potent risk-mitigation tool. It has the potential to shape the risk-taking incentives of owners and managers. If reform can act on these incentives at source, it reduces the chances of regulators chasing risk around the system (Haldane, 2012). In other words, governance reform may be among the most effective ways of curbing risk-taking without engendering regulatory arbitrage. 4.2. Risk-taking incentives in banking Although all firms face governance problems, there are three good reasons for believing these may be more acute in banking than in other sectors. First, public companies have, for well over a century, benefitted from limited liability. This means that payoffs to equity-holders can mimic an out of money call option on its assets, with a strike price given by its debt liabilities (Merton, 1974). This payoff asymmetry shapes risk-taking incentives in potentially important ways. For example, the value of equity can then be boosted by increasing the variability of equity payoffs, either by investing in riskier assets or by leveraging those assets. This shifts risk up the capital structure to the detriment of debt-holders (Jensen and Meckling, 1976). In other words, limited liability gives rise to a principalagent problem between shareholders and debt-holders. These incentive problems are likely to be more acute in banking than in other sectors: first, because risk is a deliberate choice variable for banks by dint of their choice of assets; and second, because their much greater leverage increases balance sheet risk and thereby the payoff asymmetry. These problems have long been recognised. They are the reason why limited liability came later to banking than other sectors (Crick and Wadsworth, 1936). And they are why double or even treble liability persisted in banking long after unlimited liability had been abolished (Haldane, 2012). There is a second, well-known principal-agent problem within many firms, namely between shareholders and managers (Jensen and Meckling, 1976). This might arise because managers put their own objectives over those of shareholders. One means of aligning incentives between the two is to remunerate bank managers in equity. That has become a widespread practice during recent years, particularly within the financial sector. To take a striking example, in 2006 the typical bank CEO’s wealth rose by $1 million for every 1% increase in the value of their firm (Fahlenbrach and Stulz, 2011). Compensating managers in equity is not, however, costless. By aligning managerial incentives with shareholders, the risk-shifting problem is potentially exacerbated. This includes incentives to ‘‘gamble for resurrection’’ when firms are nearing insolvency. Evidence during the crisis is revealing here. In 2006, the largest percentage equity stakes by US bank CEOs were held in Lehman Brothers, Bear Stearns, Merrill Lynch, Morgan Stanley and Countrywide (Fahlenbrach and Stulz, op. cit.). In each case, this may have induced managerial incentives to gamble for resurrection – in each case unsuccessfully. In short, in solving one principal/agent problem (between managers/shareholders), equity-based pay may have worsened another (between shareholders/debt-holders). A third incentives issue is moral hazard. In principle, if risk is shifted to debt-holders they ought to seek compensation through higher yields. That, in turn, would impose a degree of discipline on shareholder/manager incentives to risk-shift (Haldane, 2012). In practice, evidence of debt-holders having exercised discipline
Please cite this article in press as: Ellis, L., et al. Systemic risk, governance and global financial stability. J. Bank Finance (2014), http://dx.doi.org/10.1016/ j.jbankfin.2014.04.012
6
L. Ellis et al. / Journal of Banking & Finance xxx (2014) xxx–xxx
consistently over bank risk-taking is scarce, including in the precrisis period (Flannery and Sorescu, 1996; Krishnan et al., 2005). One possible reason is that bank debt-holders’ risk senses tend to be dulled by insurance, either explicit in the case of bank depositors or implicit in the case of bank bondholders. During this and previous crises, both classes of debt were effectively underwritten by the state to protect the wider economy from disorderly bank runs (Haldane, 2010). But (implicit or explicit) insurance of bank debt has two unfortunate side-effects. First, it removes the market disciplining effect of debt which might otherwise damp risk-shifting incentives. Second, it shifts the burden of risk from debt-holders onto the state, and hence taxpayers, with a corresponding deadweight cost. In other words, a second principal-agent problem emerges – between bank investors and society as a whole (Haldane, 2010). This is not an incentive problem felt as acutely outside banking where the collateral costs of firm failure, and hence the probability of state insurance, are lower. 4.3. Reforming risk-taking incentives in banking If these risk-taking incentives are potent, which crisis experience suggests may have been the case, what might be done to counter them? There are various reform options on the table, which speak to each of the different dimensions of the incentive problem. First, the regulatory capital base of banks could be increased. This reduces incentives to generate a leverage-induced rise in equity returns (Admati and Hellwig, 2013). Or, put differently, it reduces the amount of risk that will be shifted onto debt-holders. Regulatory reform initiatives, such as Basel III, augment regulatory capital standards and so would tend to reduce the principal-agent problem between shareholders and debt-holders. Nonetheless, there is an open debate about whether levels of capital in the system, even after Basel III, will be adequate to reshape fundamentally risk-shifting incentives (Admati and Hellwig, op.cit.). Second, the compensation structure of managers could be reformed. For example, incentives to risk-shift to debt-holders would be diluted or internalised by paying managers in long-term debt, rather than cash or equity (Edmans and Liu, 2011). And incentives to gamble for resurrection could be reduced by having a lengthy period over which rewards are deferred (ex-ante) or clawed-back (ex-post) in the event of risks subsequently materialising. Some progress has been made in enacting these reforms by regulators since the crisis, under the auspices of the Financial Stability Board (2009, 2012a, 2012b, 2012c). But, as with bank capital, there is an open debate about whether existing international rules go sufficiently far in reshaping risk-taking incentives (Squam Lake Working Group on Financial Regulation, 2010; Parliamentary Commission on Banking Standards, 2013). The extent to which individual countries apply these rules consistently has also been queried (FSB, 2012a, 2012b, 2012c). Third, efforts could be made to encourage debtor discipline, and hence discourage risk-shifting, by reducing the probability of governments needing to provide support to banks during crisis. One way this could be done is by putting in place resolution regimes which offer the credible prospect of ‘‘bailing-in’’ creditors in the event of stress (Coeuré, 2013). A number of countries have, or are in the process, of putting in place such regimes, again under the auspices of the Financial Stability Board (2011a, 2012a, 2012b, 2012c). As with the other initiatives, the jury remains out on the practical impact this will have on risk-taking incentives. There is a final way in which incentives of bank stakeholders, broadly-defined, might be better aligned at source with societal preferences. That would be by reforming the structure of company law – for example, by extending control rights beyond
shareholders. The case for doing so seems especially strong in banking where the imbalance between shareholder control (100% of the balance sheet) and their stake (typically less than 5% of the balance sheet) is so significant. Some reform suggestions have been made in this direction (Parliamentary Commission on Banking Standards, 2013), but so far these have made relatively little headway. 5. Conclusion The aim of this paper is to analyse some of the issues with respect to systemic risk, governance and global financial stability. The paper discusses the work of the Financial Stability Board (FSB) with respect to global, systemically important financial institutions (G-SIFs) and domestic systemically important financial institutions (D-SIBs). Given the increasingly interconnected nature of the global financial system, this paper welcomes the attempt by the FSB and central banks and supervisors to facilitate the sharing of firm level data on systemically important financial institutions. The paper argues that one of the issues, with respect to the global regulatory framework, is to ensure that certain standards are globally accepted and implemented. This is because, as Caruana (2013a) stated, if we do not control the tendency to fragmentation and also regulatory arbitrage, we could lose the benefits of globalisation and the pace of process of financial globalisation that has accelerated due to technological changes and the increasing interdependence of national economies. Furthermore, consistency must be achieved across the international stage. Rules and minimum standards should not be different based on a bank’s nationality. If some banks are held to a lower standard than those internationally agreed upon, this will result in an uneven playing field. The paper also argues that given the diversity of the whole financial system, it should be expected that the risks individual financial institutions face are also diverse. It cannot be assumed that the appropriate capitalisation is constant across all risks. While leverage ratios are a useful backstop measure and guard against potential gaming of risk-weights, their appropriate role is as a backstop. The diversity within the financial system also supports the fact that a single measure of systemic risk is unlikely to be universally applicable, and neither is a single instrument of financial stability policy. The paper argues that while there have been analyses of banks competition policies, resolution policies, supervisory policies and auditing and valuation policies, less attention has been paid to the role of bank governance and systemic risk, despite a strong link between governance and risk-taking. The paper offers four solutions to strengthen bank governance. First, the regulatory capital base of banks could be increased. This reduces incentives to generate a leverage-induced rise in equity returns. Second, the compensation structure of managers could be reformed. For example, incentives to risk-shift to debt-holders would be diluted or internalised by paying managers in long-term debt, rather than in cash or equity. Third, efforts could be made to put in place resolution regimes which offer the credible prospect of ‘‘bailing-in’’ creditors in the event of stress. There is a final way in which incentives of bank stakeholders, broadly-defined, might be better aligned at source with societal preferences. This would be by reforming the structure of company law – for example, by extending control rights beyond shareholders. Disclaimer The views expressed by Luci Ellis and Andy Haldane in this article are theirs and not necessarily those of the Reserve Bank of Australia and the Bank of England respectively. Luci Ellis would like to thank Elliott James for his valuable research assistance work.
Please cite this article in press as: Ellis, L., et al. Systemic risk, governance and global financial stability. J. Bank Finance (2014), http://dx.doi.org/10.1016/ j.jbankfin.2014.04.012
L. Ellis et al. / Journal of Banking & Finance xxx (2014) xxx–xxx
Fariborz Moshirian would like to acknowledge the support of the Australian Research Council (RG124356) for this project. Fariborz Moshirian would like to thank Christopher Wong and Jeffrey Chen for their research assistance work on this project. All errors remain the authors responsibility.
References Admati, A., Hellwig, M., 2013. The Bankers’ New Clothes: What’s Wrong with Banking and What to Do About it. Princeton University Press. Arnold, B., Borio, C., Ellis, L., Moshirian, F., 2012. Systemic risk, macroprudential policy frameworks, monitoring financial systems and the evolution of capital adequacy. Journal of Banking and Finance 36, 3125–3132. Australian Prudential Regulation Authority and Reserve Bank of Australia, 2012. Macroprudential Analysis and Policy in the Australian Financial Stability Framework. September, Australia., . Bank for International Settlements, 2013. Global Systemically Important Banks: Updated Assessment Methodology and the Higher Loss Absorbency Requirement. BIS, Basel. Basel Committee on Banking Supervision, 2010a. Basel III: A Global Regulatory Framework for More Resilient Banks and Banking Systems. BCBS, Basel. Basel Committee on Banking Supervision, 2010b. Guidance for National Authorities Operating the Countercyclical Capital Buffer. BCBS, Basel. . Beltratti, A., Stulz, R.M., 2009. Why did Some Banks Perform Better during the Credit Crisis? A Cross-country Study of the Impact of Governance and Regulation. NBER Working Paper Series No. 15180. Benigno, G., Chen, H., Otrok, C., Rebucci, A., Young, E.R., 2013. Financial crises and macro-prudential policies. Journal of International Economics 89, 453–470. Bisias, D., Flood, M., Lo, A.W., Valavanis, S., 2012. A Survey of Systemic Risk Analytics. Office of Financial Research Working Paper #0001. Caruana, J., 2013a. Regulatory landscapes: consistent regulatory implementation to keep markets integrated. In: Speech at the 2013 International Monetary Conference, Shanghai, 3 June. Caruana, J., 2013b. Addressing risks to financial stability. In: Speech to the 49th SEACEN Governors’ Conference and High-level Seminar, Kathmandu, 21 November. Coeuré, B., 2013. The implications of bail-in rules for bank activity and stability. Speech at ‘Financing the Recovery after the Crisis – The Roles of Bank Profitability, Stability and Regulation. Bocconi University, Milan, 30 September. Crick, W.F., Wadsworth, J.E., 1936. A Hundred Years of Joint Stock Banking. Hodder & Stoughton. Dell’Ariccia, G., Igan, D., Laeven, L., Tong, H., 2012. Policies for Macrofinancial Stability: How to Deal with Credit Booms. IMF Staff Discussion Note SDN 12/06. Edge, R.M., Meisenzahl, R.R., 2011. The unreliability of credit-to-GDP ratio gaps in real-time: implications for countercyclical capital buffers. Federal Reserve Board Finance and Economics Discussion Series, 2011–2037. Edmans, A., Liu, Q., 2011. Inside debt. Review of Finance 15, 75–102.
7
Fahlenbrach, R., Stulz, R.M., 2011. Bank CEO incentives and the credit crisis. Journal of Financial Economics 99, 11–26. Ferreira, D., Kershaw, D., Kirchmaier, T., Schuster, E.-P., 2012. Shareholder Empowerment and Bank Bailouts. London School of Economics Financial Markets Group Discussion Paper No. 714. Financial Stability Board, 2009. Principles for Sound Compensation Practices. BIS, Basel. Financial Stability Board, 2011a. Policy Measures to Address Systemically Important Financial Institutions. BIS, Basel. Financial Stability Board, 2011b. Macroprudential Policy Tools and Frameworks Progress Report to the G20. BIS, Basel. Financial Stability Board, 2012a. First Progress Report on Compensation Practices. BIS, Basel. Financial Stability Board, 2012b. Identifying the Effects of Regulatory Reforms on Emerging Market and Developing Economies: A Review of Potential Unintended Consequences. BIS, Basel. Financial Stability Board, FSB 2012c. Update of Group of Global Systemically Important Banks (G-SIBS). BIS, Basel. Financial Stability Board, 2013a. 2013 Update of Group of Global Systemically Important Banks (G-SIBs).. Financial Stability Board, 2013b. Global Systemically Important Insurers (G-SIIs) and the Policy Measures that will Apply to Them. BIS, Basel. Financial Stability Board, 2013c. Thematic Review on Resolution Regimes. BIS, Basel. Flannery, M.J., Sorescu, S.M., 1996. Evidence of bank market discipline in subordinated debenture yields: 1983–1991. Journal of Finance 51, 1347–1377. Haldane, A.G., 2010. The debt hangover. In: Speech Given at a Professional Liverpool Dinner, 27 January, Liverpool. Haldane, A.G., 2012. Discussion of Financial instruments, financial reporting, and financial stability by Christian Laux. Accounting and Business Research 42, 261– 266. International Accounting Standards Board, 2013. Financial Instruments: Expected Credit Losses. ED/2013/3. Jensen, M.C., Meckling, W.H., 1976. The theory of the firm: managerial behaviour, agency costs and ownership structure. Journal of Financial Economics 3, 305– 360. Krishnan, C.N.V., Ritchken, P.H., Thompson, J.B., 2005. Monitoring and controlling bank risk: does risky debt help? Journal of Finance 60, 343–378. Laeven, L., Levine, R., 2009. Bank governance, regulation and risk taking. Journal of Financial Economics 93, 259–275. Merton, R.C., 1974. On the pricing of corporate debt: the risk structure of interest rates. The Journal of Finance 29, 449–470. Moshirian, F., 2011. The global financial crisis and the evolution of markets, institutions and regulation. Journal of Banking and Finance 35, 502–511. Moshirian, F., 2012. The future and dynamics of global systemically important banks. Journal of Banking and Finance 36, 2675–2679. Parliamentary Commission on Banking Standards, 2013. Changing Banking for Good – Summary, Conclusions and Recommendations, vol. I. June. Squam Lake Working Group on Financial Regulation, 2010. Regulation of Executive Compensation in Financial Services. February. Woodford, M., 2010. Optimal monetary stabilization policy. In: Friedman, B.M., Woodford, M. (Eds.), Handbook of Monetary Economics, vol. 3, pp. 723–828 (Chapter 14).
Please cite this article in press as: Ellis, L., et al. Systemic risk, governance and global financial stability. J. Bank Finance (2014), http://dx.doi.org/10.1016/ j.jbankfin.2014.04.012
Contents lists available at ScienceDirect
Journal of Banking & Finance journal homepage: www.elsevier.com/locate/jbf
Systemic risk, governance and global financial stability Luci Ellis a, Andy Haldane b, Fariborz Moshirian c,⇑ a
Reserve Bank of Australia, Sydney, Australia Bank of England, London, United Kingdom c Institute of Global Finance, University of New South Wales, Sydney, Australia b
a r t i c l e
i n f o
Article history: Received 31 May 2013 Accepted 9 April 2014 Available online xxxx JEL classification: G15 G25 Keywords: Systemic risk Bank governance Financial stability
a b s t r a c t The paper argues that while attempts have been made to reform four of the five key pillars of banks’ operation, (i.e. competition, resolution, supervisory, and auditing and valuation policies), less attention has been paid to the role of bank governance and systemic risk, despite a strong link between governance and risk-taking. The paper offers four solutions to strengthen bank governance. First, the regulatory capital base of banks could be increased. Second, the compensation structure of managers could be reformed. Third, effort could be focussed on creating and implementing resolution regimes which offer the credible prospect of ‘‘bailing-in’’ creditors in the event of stress and fourth solution is to reform the structure of company law– for example, by extending control rights beyond shareholders. Furthermore, the paper argues that given the diversity of the whole financial system, it is expected that the risks individual financial institutions face are also diverse. It cannot be assumed that the appropriate capitalisation is constant across all risks. While leverage ratios are a useful backstop measure and guard against potential gaming of risk-weights, their appropriate role is as a backstop. The diversity within the financial system also supports the fact that a single measure of systemic risk is unlikely to be universally applicable, nor is a single instrument of financial stability policy. Ó 2014 Published by Elsevier B.V.
1. Introduction The recent global financial crisis highlighted the importance of addressing some of the incomplete financial reforms. One of these unaddressed reforms has been the contribution of systemic risk in destabilising financial markets. Recently, as reported by Bisias et al. (2012), international institutions such as the IMF, the BIS and the SFSB stated that systemic risk can be defined as a ‘‘risk of disruption to financial services that is caused by an impairment of all or parts of the financial system and has the potential to have serious negative consequences for the real economy.’’ One of the policy issues is how one might operationalize a policy to limit systemic risk given that a risk is inherently unobservable – only outcomes are observable. Unfortunately, there is no single operational definition of systemic risk so far. Bisias et al. (2012) identified 31 different measures of systemic risk, emphasising a range of aspects of systemic risk and potential channels to financial distress. Indeed, as Bisias et al. (2012) point out, it is probably not desirable to have
⇑ Corresponding author. Tel.: +61 612 93855859; fax: +61 61293854763. E-mail address: [email protected] (F. Moshirian).
a single measure of systemic risk as the focus of policy, as this may result in a ‘Maginot Line’ situation where vulnerabilities building in other parts of the financial system are missed. The recent global financial crisis has encouraged both policy makers and researchers to find ways by which one could mitigate the impact of systemic risk. To this end, a number of attempts have been made to identify policies, including forward-looking policies, that could potentially address systemic risk. As part of the efforts to address some of the underlying causes of the recent global financial crisis, the Financial Stability Board (‘FSB’) has been asked to act as the international standard-setter and attempt to address and coordinate issues underlying systemic risk. One of the key tasks of the FSB was to deal with the contributions of large financial institutions to global systemic risk. To this end, the FSB has identified global systemically important financial institutions (‘G-SIFIs’), and more recently domestic systemically important banks (‘D-SIBs’). These multinational and large national financial institutions have attracted more policy attention, partly due to their size, complexity, interconnectedness and their contributions to national and global financial systems (Moshirian, 2011, 2012). As part of this process and excessive risk taking by some of these institutions, themes related to too-big-to-fail (‘TBTF’) have also emerged for debate and analysis.
http://dx.doi.org/10.1016/j.jbankfin.2014.04.012 0378-4266/Ó 2014 Published by Elsevier B.V.
Please cite this article in press as: Ellis, L., et al. Systemic risk, governance and global financial stability. J. Bank Finance (2014), http://dx.doi.org/10.1016/ j.jbankfin.2014.04.012
2
L. Ellis et al. / Journal of Banking & Finance xxx (2014) xxx–xxx
Another major development in recent years has been a focus on developing macroprudential frameworks to complement the traditional microprudential approach to regulation. There is also a desire on the part of the policy makers and the market participants to see the financial system sharpen its ability to apply effective macroprudential policies without stifling economic growth and innovation. However, the identification and monitoring of systemic risk and optimal macroprudential policies are still in the early stage of development. In search for global financial stability, one area that has received less attention in recent years are the issues related to bank governance. More discussion and analysis of the role of governance within the operation of financial institutions and also the role of governance in contributing to global financial stability would be an essential part of any policy package aiming to address a number of the underlying causes of financial fragility over time. The paper argues that while there have been analyses of banks competition policies, resolution policies, supervisory policies and auditing and valuation policies, less attention has been paid to the role of bank governance and systemic risk, despite a strong link between governance and risk-taking. This is puzzling (Jensen and Meckling, 1976). Recent empirical evidence suggests this link may be especially strong for financial firms (Beltratti and Stulz, 2009; Ferreira et al., 2012; Laeven and Levine, 2009). Certainly, the global financial crisis unearthed bank governance failures operating at multiple levels: managers failing to control risk-takers, boards failing to control managers and investors failing to discipline either managers or boards. The purpose of this paper is to discuss some of the issues related to systemic risk, governance and financial stability. To this end, Section 2 of the paper discusses issues related to systemic risk, the identification of G-SFIS and D-SIBs and the recent policies related to macroprudential policies and the importance of implementing globally a number of regulatory policies that have been proposed by various international institutions; Section 3 analyses issues related to measurement of systemic risk. It argues that the diversity within the financial system also supports the fact that a single measure of systemic risk is unlikely to be universally applicable. Furthermore, this section discusses how, due to the diversity of the whole financial system, the risks individual financial institutions face are also diverse. The section also discusses why one may not assume that the appropriate capitalisation is constant across all risks. Section 4 discusses bank governance, including how this important aspect of global financial stability has been overlooked for vigorous debate and analysis in recent times and the importance of having specific policies to address the current structural weaknesses of bank governance. This section offers solutions regarding how to strengthen bank governance, including the regulatory capital base of banks, could be increased, the compensation structure of managers could be reformed and efforts could be made by putting resolution regimes which offer the credible prospect of ‘‘bailing-in’’ creditors in the event of stress, into place and Section 5 concludes.
have been identified by the FSB that could qualify certain financial institutions to be part of G-SIFIS. These attributes include: size, lack of substitutability and interconnectedness. In addition, the Basel Committee for Banking Supervision (Basel Committee) identified cross-jurisdictional activity and complexity as other attributes of financial institutions which could fall into the category of G-SIFIs (see BIS, 2013). As of November 2013, there are 29 designated G-SIFIs by the FSB. As part of an overall strategy to ensure more stable G-SIFs, these institutions are now required to maintain greater loss absorbency capabilities under Basel III. As highlighted by the FSB’s various publications, this additional requirement, which applies in addition to the Basel III capital requirements, is intended to reduce the ‘‘cross-border negative externalities’’ of the global financial system and to reduce systemic risk. Some regional and domestic banks could also pose systemic risk to the national and regional financial systems and national economies, (over and above those institutions identified as G-SIFIs). To this end, the Basel Committee has developed a framework and asked local authorities to identify whether a large domestic bank is systemically important from a domestic perspective (domestic systemically important bank (D-SIB). A number of national authorities have already identified their D-SIBs. One should also note that there are also some regional systemically important banks that could be operating in different continents whose operations could pose systemic risk. As part of an overall strategy to increase global financial stability, international institutions such as the FSB now have a strategy in place to ensure that national and international policy makers work together and ensure that the G-SIFIs and D-SIBs are better supervised and these institutions, as state above, are required to put aside more capital as part of their operation. This is because the current strategy is to insulate the global financial system from systemic shocks on top of the Basel III capital and liquidity requirements.1 As Moshirian (2012) stated, the Great Depression in the 1930s created incentive for authorities to collect national economic data so policy makers could measure the magnitude of economic downturn in the economy accurately. This attempt in the 1930s led to the emergence of relevant data that are used to generate what is now referred to as Gross Domestic Products (GDP). Similarly the recent global financial crisis has created a new impetus to policy makers and market participants to improve and also share some financial data, particularly when the financial market is becoming increasingly interconnected. The attempt by the FSB, in collaboration with central banks and supervisors, to create a mechanism that could facilitate the international sharing of firm-level data on systemically important financial institutions, is yet another step in the process of facilitating globalisation and another attempt to enhance the capacity of the global financial system to be more informed about the overall state of large financial institutions and the nature of their aggregate risk taking and business models.
2. Large financial institutions and propagation of systemic risk
The FSB has been working on issues related to macroprudential policies over the last few years. While a number of proposals have emerged as part of the tools available to enhance the effectiveness of policy recommendations with respect to macroprudential strategies, this area of policy development is still in its infancy and requires more work (Arnold et al., 2012). Nevertheless, some good progress has been made in recent times including the work of the FSB itself. At the present time, there are four main areas of
2.1. Identification of systemically important financial institution According to Financial Stability Board, a financial institution could be defined as a G-SIFI, if its failure or fragility could impact other financial institutions, the wider financial system and the domestic and international economies (FSB, 2011a, 2011b). Given the size and complex nature of these large financial institutions with other institutions, their failure or fragility could negatively impact the overall global financial system. A number of attributes
2.2. Macroprudential policies and a global approach to financial stability
1 This section of the paper benefited from a number of information and issues that can be found in the FSB publications and website.
Please cite this article in press as: Ellis, L., et al. Systemic risk, governance and global financial stability. J. Bank Finance (2014), http://dx.doi.org/10.1016/ j.jbankfin.2014.04.012
L. Ellis et al. / Journal of Banking & Finance xxx (2014) xxx–xxx
development proceeding under the supervision of the FSB. The first is developing measures to identify and monitor systemic risk, including both regulated banking and shadow banking. As the FSB (2011a) stated, despite some progress in collecting new data, the challenge remains to be able to identify tools and instruments that could guide forward policy making and ensure its credibility. Therefore, to be able to identify and contain systemic risk through credible instruments and tools remains the second area of development. According to the FSB (2011a), there are currently a range of tools used by various countries to address systemic risk, and they fall roughly into three categories. These categories are: (1) tools to address financial stability risks arising from rapid credit expansion; (2) tools to address amplification mechanisms of systemic risk such as leverage and maturity mismatches and (3) tools to limit spillover effects from the failure of SIFIs. To be able to develop institutional arrangements for macroprudential regulators at both domestic and regional levels will be the third area of development. Not surprisingly, the fourth area identified by the FSB is the importance of achieving regional and international cooperation that is essential to address challenges of systemic risk that often become supernational in their dimension. The increasing interconnectedness of global finance implies that some of the financial risk may well spill over to other parts of the world. One of the challenges of our generation will remain how to ensure an effective global financial system that has the full support of national authorities and the private sector, that in turn could minimise the opportunity for regulatory arbitrage. Furthermore, consistency of action by all national authorities could also generate more confidence in the collective implementation of some of the current and future global agreements. Obviously, amongst the key protagonists of the financial market (i.e. policy makers, regulators, researchers and market participants) there should be constant dialogue, consultation, and reflection about some of these global agreements. It is noteworthy that the BIS is also keen to see certain standards globally accepted and implemented. For instance, Caruana (2013a) stated, if we do not control fragmentation tendency and also regulatory arbitrage, we could lose the benefits of globalisation and the pace of process of financial globalisation that has accelerated due to technological changes and the increasing interdependence of national economies. Caruana (2013a) also stated that fragmentation could prevent the proper transmission of monetary policy, particularly in the Euro-zone area., Furthermore, in addition to a structured supervisory framework (e.g. Peer reviews), greater cooperation is needed between domestic and foreign policy makers, where each are cooperating in order to resolve issues. Furthermore, consistency must be achieved across the international stage. Rules and minimum standards should not be different based on the basis of a bank’s nationality. As a result, if some banks are held to a lower standard than the internationally agreed rules, this will result in an uneven playing field. In this context, as the FSB (2011a) stated that there is a need to establish strong mechanisms to ensure consistent action by countries to contain risk, and to ensure national regulatory frameworks are consistent. With regard to regional cooperation, it should be noted that regional mechanisms, such as the college of supervisors, may work relatively well for some of the G-SIFs and also regional and domestic SIFS in the EU due to the high level of financial integration. However, such regional cooperation and information sharing may not necessarily work in other parts of the world such as Asia, where the political climate is different and the level of national autonomy much higher. One cannot dismiss a number of regional architectures that have been erected to bring the economies of the Asia Pacific region closer. For instance APEC is designed to promote free trade, while ASEAN and ASEAN plus 3 are more focused on free
3
trade and, economic and financial integration within the member countries of such blocks. Similarly ASEAN plus 6 and the East Asia Summit also have some economic, trade and finance objectives that are ultimately designed to bring a large number of countries economically and financially closer together. Given the importance of Asia and the massive amount of foreign capital flowing to this region and the increasing volume of trade, such cooperation and information sharing amongst policy makers with respect to the operations of some of the G-SIFs and regional and domestic SIFs will be crucial over time. Regarding the supervision and coordination of G-SIFs and some of the regional and domestic SIFs, Caruana (2013b) has also stated ‘‘this kind of supervision requires not only a new and different kind of expertise but, even more importantly, the capacity and willingness to act under significant uncertainty. These include the ability to conduct group-wide consolidated supervision and to challenge banks’ business models, their corporate strategy, governance, risk profiles, ROE targets and capital plans. Also necessary will be the willingness to exercise judgment and to act pre-emptively under conditions of uncertainty’’.
3. Considerations for modelling policies to address systemic risk 3.1. Systemic risk measurement Post-crisis, a large literature has started to build up, which attempts to incorporate macroprudential policy into canonical macro-models, generally of the micro-founded Dynamic Stochastic General Equilibrium modelling (DSGE) type or variants of these. These papers have therefore had to add some kind of financial sector, so that the policy has something to operate upon. This exercise necessarily requires taking a position on asset and credit dynamics, but these relationships are generally less well understood than the inflation and real-side dynamics that are more relevant for analysis of monetary policy questions. One of the problems with this line of research is that – unlike inflation – there is little welfare-theoretic basis for using typical measures of asset prices and credit as targets of macroprudential or any other policy. In the monetary policy literature, inflation (or inflation variability) maps into social welfare, either directly because agents are presumed to dislike inflation, or indirectly because price volatility is costly for output. For example, earlier generations of literature on monetary policy assumed an objective function for the policy maker that was a weighted average of inflation volatility and output volatility; this formulation, while not exactly corresponding to social welfare, turns out to bear close resemblance to it (Woodford, 2010). In contrast, asset prices and credit do not enter into welfare functions in a way that suggests that their growth should be resisted by some arm of public policy. To the extent that wealth enters into welfare, one could argue that policies that increase asset prices are desirable. Indeed, some micro-founded models produce the result that under-borrowing, rather than over-borrowing, is the more pertinent issue, and that macroprudential policies could in fact be welfare-reducing (Benigno et al., 2013). More fundamentally, in the presence of credit constraints coming from information asymmetries, it is far from clear that ‘normal’ levels of credit are in fact optimal. This observation in no way contradicts the empirical observation that credit booms can be harmful for the subsequent evolution of output, because they can spark a financial crisis (Dell’Ariccia et al., 2012). Even the internationally agreed prudential rules acknowledge this: rather than mandating a particular level of credit or asset prices or growth rate of these series, the Basel Committee’s guidance on application of the countercyclical capital buffer promoted
Please cite this article in press as: Ellis, L., et al. Systemic risk, governance and global financial stability. J. Bank Finance (2014), http://dx.doi.org/10.1016/ j.jbankfin.2014.04.012
4
L. Ellis et al. / Journal of Banking & Finance xxx (2014) xxx–xxx
(but did not require) the use of a ‘buffer guide’ derived from deviation from trend of the ratio of credit to GDP (BCBS, 2010). There are many legitimate reservations about this guide and how it is formulated (see, for example Edge and Meisenzahl, 2011 and Box C in APRA and RBA, 2012), but the important point in this context is that it is understood even by the proponents of the buffer guide that there is no welfare-theoretic basis for a fixed level of the ratio of credit to GDP. This ratio can trend over long time horizons, and it can have different sustainable levels in different countries, depending on such factors as the degree of financial development and demographic structure. Even substantial deviations from trend do not necessarily warrant a costly policy to prevent them. Dell’Ariccia et al. (2012) found that around one in three credit booms were followed by a banking crisis, which necessarily implies that around two in three were not. This disconnect between these variables and welfare-theoretic measures indicates that there is a deeper goal being pursued. What financial stability policies – whether labelled ‘macroprudential’ or otherwise – are intended to do is limit systemic risk. Systemic risk is generally defined as the risk that the financial system might experience a generalised collapse or other form of distress. The reason for limiting this risk is to avoid the attendant harm on the real economy. The ultimate target is therefore output. In other words, asset prices and credit growth are useful information variables and possibly even intermediate targets of policy, much as monetary aggregates were in an earlier era. They are not, however, ultimate targets in same the way as inflation and output are targets of macroeconomic policy. The above line of argument naturally raises the question of how one might operationalize a policy to limit systemic risk given that a risk is inherently unobservable – only outcomes are observable. Unfortunately, there is no single operational definition of systemic risk and possibly never will be. Bisias et al. (2012) identified 31 different measures of systemic risk, emphasising a range of aspects of systemic risk and potential channels to financial distress. Among these are measures that capture deviations from trend in asset prices or credit, much as Dell’Ariccia et al. (2012) do, and measures that stress test the effects of large declines in these and related variables. Other measures focus on the propagation of distress from one financial institution to another, such that an idiosyncratic failure ends up causing a systemic problem. Dimensions of vulnerability that are emphasised by the various measures include correlated exposures, network effects where the default of one party stresses its creditors, illiquidity and leverage. Indeed, as Bisias et al. (2012) point out, it is probably not desirable to have a single measure of systemic risk as the focus of policy, as this may result in a ‘Maginot Line’ situation where vulnerabilities building in some other part of the financial system are missed. This has obvious implications both for the approach to actual financial stability policy, and how that policy is captured in theoretical models. In either situation, a single instrument, single target analogue to monetary policy is unlikely to be able to generate useful results. 3.2. Systemic risk and diversity As is well known, the financial system performs a number of functions that are essential to the workings of a modern economy. As well as intermediating between savers and borrowers, the financial sector provides a conduit for payments and enables risk transfer. These are a diverse set of functions and it should therefore be no surprise that the activities of the financial system give rise to a diverse set of risks. For example, banks and similar entities intermediate between borrowers and savers; fund managers also help savers deploy their funds, but without interposing their own
creditworthiness in place of that of borrowers or equity issuers, as is the case for banks. Insurance companies allow customers to transfer diversifiable risks to them, such as property losses from fire, flood and theft, as well as life insurance. Many of the insured risks that the financial system helps to transfer are themselves risks created within the financial system. This is consistent with the principle that financial losses are easier or more attractive to insure because they can be fully compensated by a financial payout. (An example of this is that people are more likely to insure their homes against loss from fire or flood than they are to insure their pets’ lives against the same risks: a financial payout may not be seen as a true compensation for such a loss.) Even within the banking industry, banks serve different functions and therefore face different risks. Another way of saying this is that banks’ business models can vary. One way to think about bank business models is as a spectrum with investment banks at one end, commercial or retail banks at the other, and so-called ‘universal’ banks in the middle. Those labels correspond to different balance sheet structures: investment banks have larger holdings of securities and other trading assets, while commercial banks’ assets are mainly loans. As Fig. 1 shows, the importance of trading assets varies greatly within the largest global banking groups, and even within the 29 globally systemically important banks (G-SIBs) identified by the Basel Committee on Banking Supervision and the Financial Stability Board (FSB, 2013a, 2013b, 2013c), shown here as darker bars. While there are a cluster of banks with investment-banking business models in this group of G-SIBs, G-SIBs are represented across the whole range of observed balance sheet structures. In contrast, the four large Australian banks, shown as lighter bars in Fig. 1, for example, are all clearly commercial banks, with relatively little trading business. Several conclusions can be reasonably drawn from these observations. One is that if even the banking industry, let alone the whole financial system, is relatively diverse, it should be expected that the risks individual institutions face are also diverse. It cannot be assumed that the appropriate capitalisation is constant across all risks. This is a powerful argument against a single leverage ratio as the predominant instrument of prudential regulation. Even though there are concerns about the complexity of the current prudential framework, this suggests that risk-weighting is and should be here to stay. While leverage ratios are a useful backstop measure and guard against potential gaming of risk-weights, their appropriate role is as a backstop. The diversity within the financial system also supports the point made by Bisias et al. (2012) and mentioned earlier, that a single measure of systemic risk is unlikely to be universally
Share of assets, June 2013 % 60
G-SIBs Four major Australian banks
% 60
45
45
30
30
15
15
0
Banking groups
0
Fig. 1. Trading assets and securities of the largest banking groups (includes derivative assets; sample of 100 banking groups; latest available ratios have been used where June 2013 data are unavailable). Sources: FSB; RBA; SNL Financial; The Banker; banks’ annual and interim reports.
Please cite this article in press as: Ellis, L., et al. Systemic risk, governance and global financial stability. J. Bank Finance (2014), http://dx.doi.org/10.1016/ j.jbankfin.2014.04.012
L. Ellis et al. / Journal of Banking & Finance xxx (2014) xxx–xxx
applicable, and neither is a single instrument of financial stability policy. It will not be appropriate to conceive of macroprudential policy (or financial stability policy more generally) in the single instrument, single target framework that is generally used to analyse monetary policy. Different aspects of prudential regulation will bear more effectively on particular risks and aspects of systemic risk than others. Rather, issues of macroprudential policy may be best thought of as a design problem in calibrating the entire prudential framework. Separating out particular aspects of prudential regulation into specific macroprudential ‘tools’, possibly with different governance from the prudential framework more broadly, may lead to coordination problems. A more subtle issue arises in considering the niches that banks with different business models serve. Although there has been some debate about the social utility of some financial activity, and attempts such as the Volcker Rule to limit proprietary trading by banks, it is not the case that the optimal share of trading assets in bank balance sheets is zero. Neither is it the case that the optimal share of these assets would be the same across all banks. It depends on their business models and the particular risks they take on. Commercial banks are principally involved in intermediation between savers and borrowers from the non-financial sectors. In doing so, though, they may incur risks that may be prudently laid off to other parts of the financial sector that may be able to diversify those risks more effectively. Among these risks are interest rate risk and FX risk: even quite plain-vanilla retail banking can involve an array of market risks depending on how diversified the bank’s funding base is and how global the business of the bank’s commercial customers. At least some of the trading assets of banks with more investment-banking business models arise through their interbank exposures built up as counterparties to commercial banks. These counterparty relationships can cross borders. Thus even in a national financial system like Australia’s, where the domestically owned banks are primarily commercial or retail oriented, the gamut of financial services can be provided, and risks transferred, because there are other, generally foreign banks, that can serve as counterparties to local banks’ hedging activity. This implies that, although interconnectedness within the financial system can increase systemic risk, there is a limit below which this interconnectedness cannot prudently fall. In particular, although cross-border contagion is frequently identified as a risk to national financial stability, and inter-bank borrowing often a precursor to financial crisis, a position of financial autarky is not the desirable equilibrium point, either. This has implications for measures of systemic risk and crisis vulnerability, which should perhaps not be seen as linear indicators.
4. Bank governance and systemic risk 4.1. Five planks of effective reform There are five key planks of any well-defined regulatory regime: entry – that is, competition policy; exit – that is resolution policy: regulation – that is, supervisory policy; accounting – that is, auditing and valuation policy; and governance. In the light of the crisis, there have been intense efforts to reform the first four of these pillars. For example, new regimes are being put in place for competition (e.g. Parliamentary Commission on Banking Standards, 2013), regulation (e.g. Basel Committee on Banking Supervision, 2010a, 2010b), resolution (e.g. Financial Stability Board, 2011a, 2011b) and accounting (e.g. International Accounting Standards Board, 2013) within banking firms, nationally and internationally. Yet the fifth pillar – governance – has been largely untouched. That is puzzling. Academic theory has long suggested a strong link between governance and risk-taking (Jensen and Meckling, 1976).
5
And recent empirical evidence suggests this link may be especially strong for financial firms (Beltratti and Stulz, 2009; Ferreira et al., 2012; Laeven and Levine, 2009). Certainly, the financial crisis unearthed bank governance failures operating at multiple levels: managers failing to control risk-takers, boards failing to control managers and investors failing to discipline either managers or boards. Governance is also a potentially potent risk-mitigation tool. It has the potential to shape the risk-taking incentives of owners and managers. If reform can act on these incentives at source, it reduces the chances of regulators chasing risk around the system (Haldane, 2012). In other words, governance reform may be among the most effective ways of curbing risk-taking without engendering regulatory arbitrage. 4.2. Risk-taking incentives in banking Although all firms face governance problems, there are three good reasons for believing these may be more acute in banking than in other sectors. First, public companies have, for well over a century, benefitted from limited liability. This means that payoffs to equity-holders can mimic an out of money call option on its assets, with a strike price given by its debt liabilities (Merton, 1974). This payoff asymmetry shapes risk-taking incentives in potentially important ways. For example, the value of equity can then be boosted by increasing the variability of equity payoffs, either by investing in riskier assets or by leveraging those assets. This shifts risk up the capital structure to the detriment of debt-holders (Jensen and Meckling, 1976). In other words, limited liability gives rise to a principalagent problem between shareholders and debt-holders. These incentive problems are likely to be more acute in banking than in other sectors: first, because risk is a deliberate choice variable for banks by dint of their choice of assets; and second, because their much greater leverage increases balance sheet risk and thereby the payoff asymmetry. These problems have long been recognised. They are the reason why limited liability came later to banking than other sectors (Crick and Wadsworth, 1936). And they are why double or even treble liability persisted in banking long after unlimited liability had been abolished (Haldane, 2012). There is a second, well-known principal-agent problem within many firms, namely between shareholders and managers (Jensen and Meckling, 1976). This might arise because managers put their own objectives over those of shareholders. One means of aligning incentives between the two is to remunerate bank managers in equity. That has become a widespread practice during recent years, particularly within the financial sector. To take a striking example, in 2006 the typical bank CEO’s wealth rose by $1 million for every 1% increase in the value of their firm (Fahlenbrach and Stulz, 2011). Compensating managers in equity is not, however, costless. By aligning managerial incentives with shareholders, the risk-shifting problem is potentially exacerbated. This includes incentives to ‘‘gamble for resurrection’’ when firms are nearing insolvency. Evidence during the crisis is revealing here. In 2006, the largest percentage equity stakes by US bank CEOs were held in Lehman Brothers, Bear Stearns, Merrill Lynch, Morgan Stanley and Countrywide (Fahlenbrach and Stulz, op. cit.). In each case, this may have induced managerial incentives to gamble for resurrection – in each case unsuccessfully. In short, in solving one principal/agent problem (between managers/shareholders), equity-based pay may have worsened another (between shareholders/debt-holders). A third incentives issue is moral hazard. In principle, if risk is shifted to debt-holders they ought to seek compensation through higher yields. That, in turn, would impose a degree of discipline on shareholder/manager incentives to risk-shift (Haldane, 2012). In practice, evidence of debt-holders having exercised discipline
Please cite this article in press as: Ellis, L., et al. Systemic risk, governance and global financial stability. J. Bank Finance (2014), http://dx.doi.org/10.1016/ j.jbankfin.2014.04.012
6
L. Ellis et al. / Journal of Banking & Finance xxx (2014) xxx–xxx
consistently over bank risk-taking is scarce, including in the precrisis period (Flannery and Sorescu, 1996; Krishnan et al., 2005). One possible reason is that bank debt-holders’ risk senses tend to be dulled by insurance, either explicit in the case of bank depositors or implicit in the case of bank bondholders. During this and previous crises, both classes of debt were effectively underwritten by the state to protect the wider economy from disorderly bank runs (Haldane, 2010). But (implicit or explicit) insurance of bank debt has two unfortunate side-effects. First, it removes the market disciplining effect of debt which might otherwise damp risk-shifting incentives. Second, it shifts the burden of risk from debt-holders onto the state, and hence taxpayers, with a corresponding deadweight cost. In other words, a second principal-agent problem emerges – between bank investors and society as a whole (Haldane, 2010). This is not an incentive problem felt as acutely outside banking where the collateral costs of firm failure, and hence the probability of state insurance, are lower. 4.3. Reforming risk-taking incentives in banking If these risk-taking incentives are potent, which crisis experience suggests may have been the case, what might be done to counter them? There are various reform options on the table, which speak to each of the different dimensions of the incentive problem. First, the regulatory capital base of banks could be increased. This reduces incentives to generate a leverage-induced rise in equity returns (Admati and Hellwig, 2013). Or, put differently, it reduces the amount of risk that will be shifted onto debt-holders. Regulatory reform initiatives, such as Basel III, augment regulatory capital standards and so would tend to reduce the principal-agent problem between shareholders and debt-holders. Nonetheless, there is an open debate about whether levels of capital in the system, even after Basel III, will be adequate to reshape fundamentally risk-shifting incentives (Admati and Hellwig, op.cit.). Second, the compensation structure of managers could be reformed. For example, incentives to risk-shift to debt-holders would be diluted or internalised by paying managers in long-term debt, rather than cash or equity (Edmans and Liu, 2011). And incentives to gamble for resurrection could be reduced by having a lengthy period over which rewards are deferred (ex-ante) or clawed-back (ex-post) in the event of risks subsequently materialising. Some progress has been made in enacting these reforms by regulators since the crisis, under the auspices of the Financial Stability Board (2009, 2012a, 2012b, 2012c). But, as with bank capital, there is an open debate about whether existing international rules go sufficiently far in reshaping risk-taking incentives (Squam Lake Working Group on Financial Regulation, 2010; Parliamentary Commission on Banking Standards, 2013). The extent to which individual countries apply these rules consistently has also been queried (FSB, 2012a, 2012b, 2012c). Third, efforts could be made to encourage debtor discipline, and hence discourage risk-shifting, by reducing the probability of governments needing to provide support to banks during crisis. One way this could be done is by putting in place resolution regimes which offer the credible prospect of ‘‘bailing-in’’ creditors in the event of stress (Coeuré, 2013). A number of countries have, or are in the process, of putting in place such regimes, again under the auspices of the Financial Stability Board (2011a, 2012a, 2012b, 2012c). As with the other initiatives, the jury remains out on the practical impact this will have on risk-taking incentives. There is a final way in which incentives of bank stakeholders, broadly-defined, might be better aligned at source with societal preferences. That would be by reforming the structure of company law – for example, by extending control rights beyond
shareholders. The case for doing so seems especially strong in banking where the imbalance between shareholder control (100% of the balance sheet) and their stake (typically less than 5% of the balance sheet) is so significant. Some reform suggestions have been made in this direction (Parliamentary Commission on Banking Standards, 2013), but so far these have made relatively little headway. 5. Conclusion The aim of this paper is to analyse some of the issues with respect to systemic risk, governance and global financial stability. The paper discusses the work of the Financial Stability Board (FSB) with respect to global, systemically important financial institutions (G-SIFs) and domestic systemically important financial institutions (D-SIBs). Given the increasingly interconnected nature of the global financial system, this paper welcomes the attempt by the FSB and central banks and supervisors to facilitate the sharing of firm level data on systemically important financial institutions. The paper argues that one of the issues, with respect to the global regulatory framework, is to ensure that certain standards are globally accepted and implemented. This is because, as Caruana (2013a) stated, if we do not control the tendency to fragmentation and also regulatory arbitrage, we could lose the benefits of globalisation and the pace of process of financial globalisation that has accelerated due to technological changes and the increasing interdependence of national economies. Furthermore, consistency must be achieved across the international stage. Rules and minimum standards should not be different based on a bank’s nationality. If some banks are held to a lower standard than those internationally agreed upon, this will result in an uneven playing field. The paper also argues that given the diversity of the whole financial system, it should be expected that the risks individual financial institutions face are also diverse. It cannot be assumed that the appropriate capitalisation is constant across all risks. While leverage ratios are a useful backstop measure and guard against potential gaming of risk-weights, their appropriate role is as a backstop. The diversity within the financial system also supports the fact that a single measure of systemic risk is unlikely to be universally applicable, and neither is a single instrument of financial stability policy. The paper argues that while there have been analyses of banks competition policies, resolution policies, supervisory policies and auditing and valuation policies, less attention has been paid to the role of bank governance and systemic risk, despite a strong link between governance and risk-taking. The paper offers four solutions to strengthen bank governance. First, the regulatory capital base of banks could be increased. This reduces incentives to generate a leverage-induced rise in equity returns. Second, the compensation structure of managers could be reformed. For example, incentives to risk-shift to debt-holders would be diluted or internalised by paying managers in long-term debt, rather than in cash or equity. Third, efforts could be made to put in place resolution regimes which offer the credible prospect of ‘‘bailing-in’’ creditors in the event of stress. There is a final way in which incentives of bank stakeholders, broadly-defined, might be better aligned at source with societal preferences. This would be by reforming the structure of company law – for example, by extending control rights beyond shareholders. Disclaimer The views expressed by Luci Ellis and Andy Haldane in this article are theirs and not necessarily those of the Reserve Bank of Australia and the Bank of England respectively. Luci Ellis would like to thank Elliott James for his valuable research assistance work.
Please cite this article in press as: Ellis, L., et al. Systemic risk, governance and global financial stability. J. Bank Finance (2014), http://dx.doi.org/10.1016/ j.jbankfin.2014.04.012
L. Ellis et al. / Journal of Banking & Finance xxx (2014) xxx–xxx
Fariborz Moshirian would like to acknowledge the support of the Australian Research Council (RG124356) for this project. Fariborz Moshirian would like to thank Christopher Wong and Jeffrey Chen for their research assistance work on this project. All errors remain the authors responsibility.
References Admati, A., Hellwig, M., 2013. The Bankers’ New Clothes: What’s Wrong with Banking and What to Do About it. Princeton University Press. Arnold, B., Borio, C., Ellis, L., Moshirian, F., 2012. Systemic risk, macroprudential policy frameworks, monitoring financial systems and the evolution of capital adequacy. Journal of Banking and Finance 36, 3125–3132. Australian Prudential Regulation Authority and Reserve Bank of Australia, 2012. Macroprudential Analysis and Policy in the Australian Financial Stability Framework. September, Australia.
7
Fahlenbrach, R., Stulz, R.M., 2011. Bank CEO incentives and the credit crisis. Journal of Financial Economics 99, 11–26. Ferreira, D., Kershaw, D., Kirchmaier, T., Schuster, E.-P., 2012. Shareholder Empowerment and Bank Bailouts. London School of Economics Financial Markets Group Discussion Paper No. 714. Financial Stability Board, 2009. Principles for Sound Compensation Practices. BIS, Basel. Financial Stability Board, 2011a. Policy Measures to Address Systemically Important Financial Institutions. BIS, Basel. Financial Stability Board, 2011b. Macroprudential Policy Tools and Frameworks Progress Report to the G20. BIS, Basel. Financial Stability Board, 2012a. First Progress Report on Compensation Practices. BIS, Basel. Financial Stability Board, 2012b. Identifying the Effects of Regulatory Reforms on Emerging Market and Developing Economies: A Review of Potential Unintended Consequences. BIS, Basel. Financial Stability Board, FSB 2012c. Update of Group of Global Systemically Important Banks (G-SIBS). BIS, Basel. Financial Stability Board, 2013a. 2013 Update of Group of Global Systemically Important Banks (G-SIBs).
Please cite this article in press as: Ellis, L., et al. Systemic risk, governance and global financial stability. J. Bank Finance (2014), http://dx.doi.org/10.1016/ j.jbankfin.2014.04.012