3.1 INTRODUCTION

In Chapter 2 we often referred to financial systemic risk and the related meltdown in the case of default by one or more banks considered “too interconnected” to be allowed to fail.

This topic deserves great consideration because it is strictly linked to the issue of moral hazard. As stated previously, without symmetric and complete information on the financial system, moral hazard can arise as insolvent banks may act as merely illiquid ones by underinvesting in liquid assets and gambling for assistance. In the literature many authors have written about the moral hazard and the social cost related to saving insolvent institutions, and made clear they were against any form of support or rescue for them.

We pointed out that in practice it is really difficult for central banks and supervisors to distinguish ex ante between illiquid and insolvent institutions. Nevertheless, we have already verified what can happen when a systemically important financial institution (SIFI) is allowed to fail: systemic meltdown and financial turmoil are not remote warnings to list in academic papers, they may become the worst and most dramatic side of a financial crisis.

3.2 WHEN GIANTS FALL

Let us consider the largest bankruptcy in history, the fall of Lehman Brothers. It was Monday, September 15, 2008: on that day the financial world realized that anyone, even the US's fourth largest investment bank, could fail if it was unable to counterbalance a liquidity crisis.

Once again, the root causes of the Lehman Brothers crisis lay in the huge exposure to real-estate mortgage and structured product markets, and to an excessive leverage ratio. Over the previous months, only heavy recourse to the Fed's Primary Dealer Credit Facility eased Lehman's liquidity problems, but the bank failed to strengthen its balance sheet by issuing new equity.

Lehman Brothers felt that stepping forward as a single bank to issue enough new shares, without a concerted effort among all banks, would be very expensive, because it would be perceived as a sign of desperation. Lehman's share price started to erode more and more, and the share plunged on September 9, 2008. This was followed by the bank announcing very disappointing third-quarter figures and talks with the state-controlled Korea Development Bank to sell Lehman Brothers suddenly halted. Subsequently, Lehman Brothers had to collateralize by cash some credit lines to fulfil significant margin calls: it experienced a lack of liquid, unencumbered assets to use as eligible collateral.

During a dramatic weekend the president of the Federal Reserve Bank of New York, Timothy Geithner, summoned all major banks' most senior executives to secure Lehman's future. Only two possible suitors were on the table: Bank of America and Barclays. The former decided to buy Merrill Lynch for USD50 billion, the latter refused to take over Lehman without a government guarantee. Eventually, Treasury and Fed officials decided not to offer this guarantee, because Lehman, as well as its customers and counterparties, had enough time to prepare for liquidity shortage: a bailout funded by taxpayers' money would have represented another case rewarding moral hazard. Lehman filed for Chapter 11 bankruptcy in the early morning of September 15, 2008. It was a similar case to that of Bear Stearns: both were investment banks, too interconnected to fail but with a dramatic liquidity shortage. They experienced different ends: the private sector was able to save Bear Stearns, but no rescuer (public or private) was found for Lehman Brothers.

The negative effects of Lehman's default immediately spread into the global financial markets: every trader was going to evaluate net exposure versus Lehman for both collateralized and unsecured transactions; a lot of collateral had to be sold in the market to minimize counterparty risk. Lehman had in place collateral agreements with many market counterparties: moreover, it was prime broker for many hedge funds.

All players were wondering about the effects of Lehman's “jump to default” on their own books and were trying to hedge the incoming closeout risk. Only the proper functioning of SwapClear, the clearing house for derivative products, whose members included Lehman among others, avoided further destabilizing effects on the financial system. Nevertheless, the Dow Jones index went into a savage downward, dropping 500 bp on the day: everyone was on edge, braced for a new shock, and there was not long to wait!

The day after, it was the turn of the world's biggest insurance corporation, American International Group, to drop sharply (more than 60%). AIG was one of the major players in the CDS market and was exposed to billions of US dollars against the failure of Lehman. Now the unthinkable had happened and AIG simply did not have the cash to make the payouts promised. Moreover, it had invested its insuring profits heavily in the CDO market, which was going to collapse with mass defaults by mortgage holders. Rating agencies immediately downgraded it, which required AIG to post other collateral with its trading counterparties. Facing a deadly liquidity shortage, AIG was on the brink of bankruptcy as well. But this time the authorities realized that AIG was too interconnected in the credit derivative business to allow it to fail: the Fed immediately announced the creation of a secured credit facility of USD85 billion, in exchange for an 80% equity stake. October and November figures called for another USD77 billion.

The huge recourse to taxpayers' money saved AIG, but over the following months everyone was wondering about the opportunity to save Lehman in order to avoid the consequent meltdown: How much did the Lehman failure cost the US government in terms of massive additional bailout funds for AIG and other financial sectors? If Lehman was saved, would the cost to taxpayers have been lower?

3.3 A HARD LESSON

However, in spite of the actions of the Fed and the US Treasury, the financial situation was going to be worse every passing day. Mass uncertainty and a fall in confidence, with no foreseeable solutions except government intervention, were hampering market sentiment. On September 23, 2008, Ben Bernanke, chairman of the Fed, testified before the Joint Economic Committee of the US Congress. He put forward a clear representation of the crisis and strongly supported the request of the Secretary of the Treasury, Hank Paulson, to use USD700 billion of taxpayers' money to buy toxic assets from the banks (the so-called TARP).

First of all, he came up with a list of the theoretical options available to a financial firm to address its difficulties: through private sector arrangements (e.g., by raising new equity capital), or by negotiations leading to a merger or acquisition, or by an orderly winddown. In his opinion (see [20]), “Government assistance should be given with the greatest of reluctance and only when the stability of the financial system, and, consequently, the health of the broader economy, is at risk.” This was the case with Fannie Mae and Freddie Mac, for which capital raises of sufficient size appeared infeasible and the size and government-sponsored status of the two companies had precluded a merger with or an acquisition by another company.

He then analysed the course of decisions that led to a different kind of output for the AIG and Lehman cases. In his words (see [20]), “… The Federal Reserve and the Treasury attempted to identify private-sector solutions for AIG and Lehman Brothers, but none was forthcoming. In the case of AIG, the Federal Reserve, with the support of the Treasury, provided an emergency credit line to facilitate an orderly resolution. The Federal Reserve took this action because it judged that, in light of the prevailing market conditions and the size and composition of AIG's obligations, a disorderly failure of AIG would have severely threatened global financial stability and, consequently, the performance of the U.S. economy. To mitigate concerns that this action would exacerbate moral hazard and encourage inappropriate risk-taking in the future, the Federal Reserve ensured that the terms of the credit extended to AIG imposed significant costs and constraints on the firm's owners, managers, and creditors.”

The outcome for Lehman was different (see [20]): “the Federal Reserve and the Treasury declined to commit public funds to support the institution. The failure of Lehman posed risks. But the troubles at Lehman had been well known for some time, and investors clearly recognized—as evidenced, for example, by the high cost of insuring Lehman's debt in the market for credit default swaps—that the failure of the firm was a significant possibility. Thus, we judged that investors and counterparties had had time to take precautionary measures ….”

Ben Bernanke recognized that, while perhaps manageable in itself, Lehman's default, combined with the unexpectedly rapid collapse of AIG, exacerbated the extraordinarily turbulent conditions in the global financial market. They led to a sharp decline in equity prices, a spike in the cost of short-term credit, where available, and a lack of liquidity in many markets. Losses at a large money market mutual fund sparked extensive withdrawals from a number of such funds. To avoid the broad repercussions of a run on US money market funds, the US Treasury set aside USD80 billion to guarantee brokers' money market funds. Prices paid for CDSs on financial names climbed, as each bank tried to protect itself against counterparty credit risk. Financial non-asset-backed commercial paper experienced a sharp decline, which led to the introduction of the Commercial Paper Funding Facility by the FED.

Unfortunately, these actions were not enough to stabilize markets and to halt the incumbent meltdown. Congress was asked to vote on state aid for the banking system in order to avoid the catastrophe. At one point during the audience, Hank Paulson, frustrated at the lawmakers themselves for asking why, in essence, taxpayers should be hit up for cash to cover Wall Street's mistakes, answered (see [73]): “When you ask about taxpayers being put on the hook, guess what? They're already on the hook.” Ben Bernanke gave dark predictions about what would happen if the market were left to its own devices and how hard times would spread farther across the country. He said (see [73]): “The financial markets are in quite fragile condition and I think absent a plan they will get worse. I believe if the credit markets are not functioning, that jobs will be lost, that our credit rate will rise, more houses will be foreclosed upon, GDP will contract, that the economy will just not be able to recover in a normal, healthy way.”

“We have a serious ‘too big to fail’ problem in this economy,” he concluded. “If the crisis has a single lesson, it is that the too-big-to-fail problem must be solved … A promise not to intervene in and of itself will not solve the problem.”

3.4 CLOSER SUPERVISION

The large-ripple effects of Lehman's default affected financial stability for a long time: many months were needed for authorities to address effectively the open issues and to fix the situation. The vicious circle was only broken when the US government decided to convert TARP into a binding capital injection for all major US banks. By so doing, the dilemma between illiquid and insolvent institutions was ultimately resolved, because no longer could anyone doubt the solvency of the banking system after this recapitalization.

Since then, no other large bank has been allowed to fail in order to prevent relevant spillover effects to the real economy. However, it was clear that mistakes made in the past should not be repeated in the future. For all the disagreements among legislators, regulators, technicians, policy officials and the public over the right set of financial reform measures, on one point there was near unanimity: no one wanted another TARP, not those who thought TARP was the best of the bad set of options available in 2008, certainly not those who opposed it, not American citizens many of whom saw the injection of billions of dollars of government capital into financial firms as more a bailout of large banks than an imperative to stabilize the financial system, nor even most of the large financial firms that were obliged to receive government aid and tried to pay it back as soon as possible.

The main lesson learnt about Lehman is not that financial institutions that are too interconnected do not have to fail, but that they require more supervision, in order to tackle the root causes of the problems that can push them to the verge of collapse. Before this crisis a systemic bank had many incentives to become larger and more interconnected, because it could have maximized the bailout probability (“hang on to others and take positions that drag others down when you are in trouble”). In a boom, a SIFI may play a role in the buildup of leverage and wider maturity mismatches, while at the same time fostering recourse to complex and opaque forms of financial innovation. This mechanism is reversed during a downswing, when a SIFI has a disproportionate effect on the deleveraging process. The intensity of deleveraging, liquidity hoarding and asset fire sales is proportional to the size and interconnectedness of the balance sheet of a SIFI. Furthermore, the economic losses and the deterioration of confidence triggered by the distress of a SIFI are likely to generate ripple effects that dwarf those stemming from a non-systemic institution.

A too-big-to-fail firm produces two different types of negative externalities (see Brunnermeier [39]). First, its own maturity mismatch easily affects the worth of the balance sheet of other players, because fire sales needed to adjust its liquidity situation are likely to reduce the assets value of other counterparties. Second, it is likely to take on an opaque connected position whose counterparty risk adversely affects other players. The latter, often known as “network externality”, can be effectively tackled by the introduction of clearing house arrangements, which would allow netting among opposite exposures and would reduce counterparty credit risk. These measures could be powered by imposing higher capital charges on OTC contracts not centralized to a clearing house. In this way the CDS spiral, ignited by the failure of Lehman, should be avoided in the future.

It is worth pointing out that centralizing the larger part of derivative transactions in one or more clearing houses is an effective measure for reducing counterparty risk, but it simply substitutes this kind of risk with the liquidity risk related to higher initial and additional margins required by clearing houses to cover the default of major counterparties for exposures. It seems that the priority for regulators now is the reduction of counterparty risk, as if liquidity risk can always be coped with by means of liquidity provisions or, as a last resort, by central bank liquidity. As already explained, central bank liquidity can only be supposed to be always available during crisis times after having posted eligible collateral: it is not unlikely in the near future that there may be a shortage of collateral to enhance systemic liquidity provisions. In this case, the only way to pump central bank liquidity into the system is to reduce eligibility criteria, with the consequence that some components of counterparty risk could ultimately affect the soundness of a central bank's balance sheet.

The so-called fire sale/maturity mismatch externality can be managed only by regulation, by imposing stricter liquidity ratios on a SIFI. Stricter capital ratios or tailor-made leverage ratios for a SIFI are useful tools to cover all the dimensions of a financial crisis only if they are combined with effective liquidity ratios. Indeed, capital or leverage ratios are not able to capture the strong reliance on shorter term borrowing though repos and CP. In that case more capital can only reduce the negative effects of bad management, but it may not prevent future crises.

3.5 G-SIFI REGULATIONS

One of the first responses from regulators was the important package of reforms in capital regulation known as Basel III. The Basel III requirements for better quality of capital, improved risk weightings, higher minimum capital ratios, liquidity and leverage ratios, and a capital conservation buffer comprise a key component of the post-crisis reform agenda. Although a few features of Basel III reflect macroprudential concerns, in the main it is a microprudential exercise: thus, a macroprudential perspective on capital requirements is required to complement the microprudential orientation of Basel III.

There would be very large negative externalities, as stated before, associated with the disorderly failure of any SIFI, distinct from the costs incurred by the firm and its stakeholders. According to Daniel Tarullo, governor of the Fed's board (see [110]): “The failure of a SIFI, especially in a period of stress, significantly increases the chances that other financial firms will fail, for two reasons. First, direct counterparty impacts can lead to a classic domino effect. Second, because losses in a tail event are much more likely to be correlated for firms deeply engaged in trading, structured products, and other capital market instruments, all such firms are vulnerable to accelerating losses as troubled firms sell their assets into a declining market. A SIFI has no incentive to carry enough capital to reduce the chances of such systemic losses. The micro-prudential approach of Basel III does not force them to do so. The rationale for enhanced capital requirements for SIFI is to take these costs into account, make SIFI less prone to failure, and thereby to make the financial system safer. An ancillary rationale is that additional capital requirements could help offset any funding advantage derived from the perceived status of such institutions as too-big-too-fail.”

On November 4, 2011, the Financial Stability Board took the first step toward stricter regulation for G-SIFIs (global systemically important financial institutions), by giving the first list of 29 globalized banks that are required to have loss absorption capacity tailored to the impact of their default, rising from 1 to 2.5% of risk-weighted assets (with an empty bucket of 3.5%, to discourage further systemicness), to be met with common equity by end-2012. This list will be updated annually and published by the FSB each November. In the communiqué (see [30]) the FSB and BCBS stated they had assessed the macroeconomic impact of higher loss absorbency requirements for G-SIFIs: “the enduring global economic benefits of greater resilience of these institutions far exceed the modest temporary decline of GDP over the implementation horizon.”

SIFIs are defined as “financial institutions whose distress or disorderly failure, because of their size, complexity and systemic interconnectedness, would cause significant disruption to the wider financial system and economic activity” (see [30]). It is crucial to gauge the systemic relevance of a financial institution on the basis of a combination of several factors, instead of some specific ones. The size, either in absolute or in relative terms, can be reflected by a dominant position in a specific market or product or service. Interconnectedness stands for the linkage with the rest of the system, mostly via interbank lending or a special position as counterparty in key markets, as critical participant in various payment systems and as provider of major functions related to the risk management of market infrastructures. Another factor to gauge with care is substitutability (i.e., the extent to which other components of the system can provide the same service in the event of a failure).

These points have to be evaluated in combination of other relevant characteristics of the firm, as the institution's specific risk profile (leverage, liquidity, maturity mismatches and concentration of asset/liabilities) and its organisational and legal structure. The assessment of systemic importance is obviously a dynamic, time-varying and forward-looking process, depending, inter alia, on the particular conditions of financial markets. The ultimate aim of the classification should be to achieve a continuous or at least a finely granular ranking, as opposed to a simple division of firms into either systemically relevant or not, that can easily lead to some arbitrage practices for firms that can switch from one category to the other according to some quantitative ratio or indicator.

They recognize that addressing the too-big-to-fail problem requires a multi-pronged and integrated set of policies. In addition to capital requirements, the policy measures comprise:

• a new international standard, as a point of reference for reform of national resolution regimes, setting out the responsibilities, instruments and powers that all national resolution regimes should have to enable authorities to resolve failing financial firms in an orderly manner and without exposing the taxpayer to the risk of loss (the so-called “effective resolution regime”);
• requirements for resolvability assessments and for recovery and resolution planning for G-SIFIs, and for the development of institution-specific cross-border cooperation agreements so that home and host authorities of G-SIFIs are better equipped for dealing with crises and have clarity on how to cooperate in a crisis;
• more intensive and effective supervision of all SIFIs, including through stronger supervisory mandates, resources and powers, and higher supervisory expectations for risk management functions, data aggregation capabilities, risk governance and internal controls.

About the last point it is crucial not to scatter possible synergies between supervision and monetary policy. The systemic regulatory authority at the national level should be the national central bank or an institution closely associated to it. The UK example, with competencies shared between the FSA and the Bank of England, could not optimize the effectiveness of supervision because of lack of a continuous information flow between bank supervision and the central bank. For instance, having direct access to bank supervision information is essential to making a speedy bailout versus no bailout decision, to use the “lender of last resort” functions in a timely manner.

3.6 THE NEXT STEPS

Widespread discussion is well under way concerning possible rules to reduce the risk of failure of a SIFI or to mitigate the consequences of such failure for the financial system as a whole. Proposals as a result of the current debate can be classified in two main categories: (i) ex ante measures aimed at reducing the probability and impact of a SIFI's default; (ii) ex post measures aimed at ensuring that the failure of a financial institution can be solved in an orderly fashion. In the first group we can list the regulations for G-SIFIs defined by the FSB and the so-called “Volcker rule”. This proposal aims to limit proprietary trading and investment in hedge funds or private equity funds, as well as the excessive growth of leverage of the largest financial institutions relative to the financial system as a whole.

The new challenge now is to define effective ex post measures able to decrease the expected burden on taxpayers in case of a SIFI crisis. Authorities have to be endowed with appropriate mechanisms to resolve the failure of a SIFI in an orderly and prompt manner, with the cost of default or restructuring falling on equity holders and bondholders without “socialization” of losses. The major step is the development of credible plans for recovery and resolution, often known as “living wills” [29]. Recovery or going concern plans include contingency funding and derisking plans and should be prepared and periodically updated by a SIFI and reviewed by competent authorities. Resolution or gone concern plans should identify the actions to be taken once the going concern plans have proven insufficient without taking into account the possibility of public support.

In the 2011 consultation paper issued by the FSB [28] the following resolution tools were proposed:

• Sale of the business:

  The authorities would be able to effect the sale of a financial institution or all or part of its assets and liabilities to one or more purchasers on commercial terms, without requiring the consent of the shareholders or complying with procedural requirements that would otherwise apply.

  Box 3.7. The Liikanen report

  In October 2012 the High-level Expert Group, chaired by governor of the Bank of Finland Erkki Liikanen, recommended (see [84]) a set of measures to augment and complement the set of regulatory reforms already enacted or proposed by the EU, the Basel Committee and national governments.

  First, proprietary trading and other significant trading activities should be assigned to a separate legal entity if the activities to be separated amount to a significant share of a bank's business. This would ensure that trading activities beyond the threshold are carried out on a standalone basis and separate from the deposit bank. As a consequence, deposits, and the explicit and implicit guarantee they carry, would no longer directly support risky trading activities.

  Second, the Group emphasizes the need for banks to draw up and maintain effective and realistic recovery and resolution plans, as proposed in the Commission's Bank Recovery and Resolution Directive.

  Third, the Group strongly supports the use of designated bailin instruments. Banks should build up a sufficient large layer of bailinable debt that should be clearly defined, so that its position within the hierarchy of debt commitments in a bank's balance sheet is clear and investors understand the eventual treatment in case of resolution. Such debt should be held outside the banking system!

  Fourth, the Group proposes to apply more robust risk weights in the determination of minimum capital standards and more consistent treatment of risk in internal models.

• A bridge bank:

  The authorities would be able to transfer all or part of a credit institution to a bridge bank. This would be a publicly owned institution and the ultimate objective would be to facilitate the sale of the bridge bank. The operations of the bridge bank should be temporary and should be terminated within one year, although this may be extended by up to one year.

• Asset separation:

  The authorities would able to transfer certain assets of a SIFI to an asset management vehicle for the purpose of facilitating the use or ensuring the effectiveness of another resolution tool. In order to address concerns about moral hazard associated with the use of this tool that might otherwise arise, it should only be used in conjunction with another resolution tool. Then, the selected assets should be transferred to an asset management vehicle for fair consideration. Lastly, the resolution authority should appoint asset managers to manage the assets with the objective of maximizing their value through eventual sale.

• Debt writedown:

  The EC is considering a mechanism to write down the claims of some or all of the unsecured creditors of a failing institution and, possibly, to convert debt claims to equity, as a valuable additional resolution tool that would allow authorities greater flexibility. The paper makes it clear that this relates to possible future legislative changes which would be subject to full assessment and appropriate transition provisions and periods to avoid market instability or unintended consequences. It also notes that it is not envisaged to apply the measures adopted to debt currently in issue: wisely, this avoids automatically considering current unsecured debt as a new kind of subordinated debt.

  Box 8.8. How to implement the debt writedown tool?

  The consultation paper describes two different approaches to how a debt writedown tool might be implemented. The comprehensive approach provides statutory power to write down by a discretionary amount or convert all debt into equity to return a bank to solvency. It would only apply to new issued debt and this feature should be contractually recognized within bond documentation in the future.

  The paper notes that some exclusions might be necessary, such as swap, repo and derivatives counterparties and other trade creditors; short-term debt (defined by a specified maximum maturity); retail and wholesale deposits and secured debt (i.e., covered bonds). Without dwelling on the matter, we note that there could be several technical problems with this. For example, How would tapped bonds be treated? Would this not encourage banks to fund themselves with short maturities to avoid having to issue debt with writedown language? Is there not a risk of subverting the normal ranking of bank bonds?

  The targeted approach requires institutions to have a fixed volume of bailin debt in issue which would be written down or converted via a statutory trigger. This would need to include a contractual term that would specify that the relevant resolution authority could use statutory power to write down debt when an institution meets the trigger conditions for entry into resolution. This could include a fixed minimum for all institutions (e.g., as a percentage of total liabilities).

The fourth resolution tool aims to address the associated moral hazard to the notion of “too big or too interconnected to fail”. It arises from the general acknowledgment that governments and supervisory authorities would not let an ailing SIFI fail, given the significant damage to the financial system. In turn, this expectation of central bank or government support translates into a funding advantage compared with non-systemic banks.

When debtholders do not have to consider the risk of default on their investment, they will naturally tend to require a lower rate of return on the debt issued by systemic institutions. This lack of market discipline is by itself conducive to risk-taking: endowed with an implicit subsidy on its cost of funding, it is economically convenient for a SIFI to engage riskier strategies, expanding its balance sheets without appropriate price penalties.

To sum up, SIFIs benefit from double distortion to fair competition. In fact, the ex post subsidy embodied in implicit or explicit bailout guarantees translates into an ex ante funding advantage, which compounds the incentives to excessive risk-taking driven by pervasive moral hazard.

3.7 CONCLUSION

The scale of the potential fallout from the failure of a SIFI has been unveiled by the current financial crisis. In order to address, or mitigate, their potential contribution to financial instability, an overarching approach is strongly required of international policymakers and authorities. From a microprudential perspective, they are developing a strengthened regulatory and supervisory regime in order to reduce the risk contribution of the failure of a SIFI and to increase the overall resilience of the financial system. The collective behaviour of financial institutions and their interconnectedness stress the importance of recognizing the public aspect of financial stability, and underpin the recent emphasis on a joint micro and macroprudential approach to regulation and supervision.

The requirements defined by the FSB for G-SIFIs represent only the first steps to implement effective supervision and regulation to avoid new financial crises. As many banks are globalized it is very difficult for any national authority to perform these tasks successfully: for the eurozone it is strongly advised to fill as soon as possible one of the many holes in the single currency's original design. Since the euro's creation, integration of the finance industry has been rapid, with banks sprawling across national borders. The obvious answer is to move the supervision and eventual support, if required, for banks (or at least for big ones) away from national regulators to European ones.

The first and easier step was taken by the EC in the second half of 2012, placing the ECB in charge of the supervision of all European banks by the end of 2013. At a minimum there should be a eurozone-wide system of deposit insurance (to prevent loss of confidence that ignites bank runs) and oversight, with collective resources for the recapitalization of endangered institutions and regional rules for the resolution of truly failed banks. With a European rescue fund to recapitalize weak banks, and a common system of deposit insurance, politicians would no longer be able to force their banks to support national firms or buy their government bonds: banks would no longer be Italian, Spanish or German, but increasingly European. It should be the ultimate in European financial integration.