Regulatory Capital: Why Is It Different? Dirk Schoenmaker1 Duisenberg school of finance VU University Amsterdam

Draft version: 27 November 2014

Paper prepared for the Information for Better Markets Conference “Capital: reporting, regulating and resource allocation”

15-16 December 2014

Abstract The global financial crisis has highlighted that deviations between the accounting and regulatory concepts of equity capital have gone too far. Accounting standards have been overshooting in the application of fair value accounting. If there are no markets during times of crises, it does not make sense to mark-to-market. These exceptions have now been included in the accounting standards. At the same time, regulatory capital has gone astray by allowing debt elements, such as subordinated debt, to be incorporated, which did not absorb losses during the crisis. The new Basel III capital framework is rightfully reinforcing the central role of equity capital.

While the special liquidity function of banks may justify lower levels op capital than those of industrial firms, the social cost of bank failures (externalities) require higher levels than the extremely low levels of bank capital before the crisis. The level of regulatory capital has been increased, with a systemic surcharge for the large banks, to reduce the too-big-to-fail (TBTF) subsidy for these banks.

1

I would like to thank Elena Carletti for useful comments. Views in this paper are mine.

1

1. Introduction The main purpose of equity capital is to provide a first line of defence against bankruptcy. 2 The equity capital buffer has the capacity to absorb losses. In European industry, the average capital ratio (defined as equity capital divided by total assets) is about 70 per cent, and typically well above 50 per cent (Damodaran, 2014). As there is no capital regulation for industries, we are using here accounting definitions of capital.

Moving to the financial sector, the unweighted capital ratio is well below 10 per cent in banking and insurance. Some banks had capital ratios as low as 2 per cent before the global financial crisis struck in 2007/8. With such little loss absorbing capacity of the banking industry itself, the costs of bank bailouts has been squarely put on the taxpayers. Governments and supervisors have taken, and still are taking, action to change this game.

To an outsider, it may be surprising that the regulated financial sector has far lower levels of capital than the unregulated industrial sector. A first explanation is that finance, in particular banking with its liquidity providing services, is special. Nevertheless, there are strong arguments why the regulated capital levels should be beyond the levels that banks would keep themselves. The social costs of bank failures are higher than the private costs, due to externalities (also labelled systemic risk). The paper sets out the arguments for the regulation of capital in the banking sector. After the global financial crisis, there are calls for substantially higher levels of regulatory capital. The full range is from 3 per cent (the minimum leverage ratio of the new Basel III capital adequacy framework) up to 20 per cent (e.g. Admati and Hellwig, 2013; Miles et al., 2013).

The regulatory capital debate has progressively shifted toward the large banks. The global financial crisis brought into sharp focus the massive costs associated with the bail out of complex global systemically important banks (G-SIBs), which were perceived as too-big-tofail. The too-big-to-fail doctrine has been reinforced, if anything, by governments’ handling of the financial crisis. As a result, the most significant regulatory reform proposals have focused on the question of how to curtail the too-big-to-fail problem. Namely, how can one reduce moral hazard and rein back expectations of future bailouts of SIFIs?

The outcome of this regulatory reform are higher Basel III capital requirements, with a systemic capital surcharge for G-SIBs. In addition, these G-SIBs have to adopt the so-called 2

Another objective is to mitigate risk-taking incentives. Higher levels of equity capital reduces the put option value of equity and thus reduces the incentive to take risks for managers, who are assumed to work on behalf of the shareholders.

2

Living Wills. The objective is to put in place, ex ante, conditions that would allow a wider range of options other than having the whole bank rescued (Goodhart et al., 2013). A Living Will is a recovery and resolution plan drawn up ex ante with the purpose of using it if a bank gets into difficulties. The G20 requires Living Wills for the top 30 global banks (G-SIBs) and top 9 insurance companies (G-SIIs) (Claessens et al., 2010).

An interesting question is to what extent the accounting and regulatory definitions of capitals are diverging? Until the global financial crisis, the regulators, pushed by banks, had increasingly stretched the definition of regulatory capital by including capital ‘instruments’ such as subordinated term debt (Basel I) and even short-term subordinated debt (Basel II). As these classes of subordinated debt did not really absorb losses during the crisis, regulators, as well as rating agencies, are now stressing the use of Common Equity Tier 1 (CET1) capital, which consists of common shares issued (including share premium) and retained earnings. So, we are rightly returning to straight accounting equity capital for regulatory purposes.

The paper is organised as follows. Section 2 explains why capital requirements are needed. Financial stability concerns are the key driver of regulatory capital requirements. Next, Section 3 introduces the concept of regulatory capital and describes the history of capital regulation. The focus is on banking capital regulation. We also discuss the difference between the accounting and regulatory concept of capital. Section 4 explains in detail the new Basel III capital requirements for banks. The concept of bail-in debt and so-called cocos (contingent convertible bonds) are also discussed. Finally, Section 5 concludes and provides some policy recommendations.

2. From private to regulatory capital In the absence of market failures, the appropriate amount of capital can be left to private market forces. That is what happens in most industries, as discussed in the introduction. De Haan et al. (2012) review the reasons for regulation of financial services. The case for government intervention is based on market failures. A market failure occurs when the private sector if left to itself (i.e. without government intervention) would produce a suboptimal outcome. Goodhart et al. (1998) identify three main reasons for government intervention in the financial sector:

1. Asymmetric information: customers are less informed than financial institutions. Financial supervision aims to protect customers against this information asymmetry.

3

2. Externalities: the failure of a financial institution may affect the stability of the financial system. Systemic supervision aims to foster financial stability and to contain the effects of systemic failure. 3. Market power: financial institutions or financial infrastructures, such as payment systems, may exert undue market power. Competition policy aims to protect consumers against monopolistic exploitation. This is beyond the scope of this paper.

Asymmetric information arises in two cases. First, customers are generally unable to properly assess the safety and soundness of a financial institution, because that requires extensive effort and technical knowledge (Dewatripont and Tirole, 1994). Establishing some sort of oversight may be needed, as financial institutions have an incentive to take too much risk. This is because high-risk investments generally bring in more revenues that accrue to the institution, while in case of failure a substantial part of the losses will be borne by the depositors. The information asymmetry creates problems of adverse selection (a riskier financial institution may make a more attractive offer to potential customers) as well as moral hazard (a financial institution may increase its risk after it has collected funds from customers). Prudential supervision aims to protect customers by ensuring the soundness of financial institutions. Capital adequacy regulations, the topic of this paper, form a key element of prudential supervision. In the new twin peaks regulatory structure, the Prudential Regulation Authority (PRA) at the Bank of England is responsible for prudential supervision in the United Kingdom (UK) (Huang and Schoenmaker, 2014).

Second, customers may not be in a position to assess properly the behaviour of a financial institution. This problem is common in professional services (Goodhart et al., 1998). In most cases, private-sector mechanisms are used to mitigate this principal-agent problem. A disciplinary body of a privately run medical association can, for example, expel a member when it finds that this member has (repeatedly) failed to meet the minimum standards of the medical profession. Why, then, is government supervision of financial services needed? An important explanation draws on the fiduciary nature of financial services. A customer hands over his money today, while the service is rendered in the (sometimes far) future. For example, only after retirement does it become clear whether the advised pension savings scheme is appropriate to meet the financial needs of the retirees. Moreover, the amount of money at risk is typically larger in financial services than in other professional services. Conduct-of-business supervision focuses on how financial institutions conduct business with their customers and how they behave in markets (see Llewellyn (1999) for an excellent overview). The Financial Conduct Authority (FCA) is responsible for conduct-of-business supervision in the UK.

4

The second market failure that may give rise to government regulation is externalities. There is a risk that a sound financial institution may fail when another financial institution goes bankrupt (spill-over effects). This externality is not incorporated in the decision making of the financial institution. The social costs of the failure of a financial institution thus exceed the private costs. In particular, banks are subject to spill-over effects as their balance sheet contains illiquid assets financed by redeemable deposits. When rumours about the quality of a bank’s assets spread, depositors may withdraw their deposits. The liquidity and subsequently the solvency of a bank will be threatened when it has to liquidate its assets at fire sale prices (i.e. prices well below prices under normal market conditions). The failure of multiple banks may lead to a banking crisis. Macroprudential supervision aims to foster financial stability and to contain the effects of systemic failure. The task of maintaining financial stability is usually assigned to the central bank. The Financial Policy Committee (FPC) at the Bank of England is responsible for macroprudential supervision in the UK.

The financial cycle Because of these externalities, the social optimal level of capital is higher than the private level. Nevertheless, financial stability related externalities were largely neglected before the financial crisis. The global financial crisis has revived interest in Minsky’s ‘financial-instability’ hypothesis (Minsky, 1986). In the Minsky model the events leading up to the crisis start with a ‘displacement’ -some exogenous, outside shock to the macroeconomic system- an invention or an abrupt change of economic policy about which investors get excited. Subsequently there are five stages to the boom and eventual bust:

1. credit expansion, characterised by rising assets prices; 2. euphoria, characterised by overtrading; 3. distress, characterised by unexpected failures; 4. discredit, characterised by liquidation; and 5. panic, characterised by the desire for cash.

The displacement sets in a boom fuelled by credit. As a boom leads to euphoria, banks extend credit to ever more dubious borrowers, often creating new financial instruments to do the job. Then, at the top of the market, some smart traders start to cash in their profits. The onset of panic is usually heralded by a dramatic event, such as a bank not being able to meet its obligations. Losses on loans begin to mount, and the drop in the value of the loans falls relative to liabilities, driving down the capital of financial institutions. With less capital, financial institutions cut back on their lending (deleveraging).

5

Minsky’s financial-instability hypothesis highlights the pro-cyclicality of the financial system. Several factors contribute to this pro-cyclicality. First, the role of risk assessment is important. While risk tends to be underestimated in good times (euphoria with ‘low risk’), it is overestimated in bad times (distress with ‘high risk’). Moreover, risk can be endogenous. For example, when financial institutions sell a particular asset to reduce risk, the price of that asset may fall further. Second, the amount of debt (leverage) is a key factor explaining the depth of the financial crisis. The more debt is built up in the upswing, the more severe is the deleveraging in the downswing. This is not only an argument for more equity financing in general, but also for more equity capital for banks (see section 4). Adrian and Shin (2008) show that banks have contributed to the upswing prior to the crisis, by increasing their leverage (more debt; less equity). This resulted in a declining leverage ratio, defined here as equity divided by total assets (see Figure 3 in section 4).

Third and last, capital requirements play a role. Banks have to keep minimum capital against new loans (see section 3). In good times, retained earnings boost capital, which enables banks to increase lending. In bad times, capital shrinks through losses, which may hamper the granting of new credit. Figure 1 illustrates how the financial cycle (measured by credit and house prices) can amplify the business cycle (measured by GDP). The amplitude of the financial cycle over the 1970-2011 period is five times that of the business cycle in the United States (US) (Borio, 2014). Moreover, the duration of the financial cycle tends to be longer than that of business cycle. While similar patterns can be found across Europe, Figure 2 represents the financial cycle in the UK. This pro-cyclical role of capital requirements is behind the new countercyclical capital buffer in Basel III (see section 4). Figure 1. The Financial and Business Cycles in the US

Note: Orange and green bars indicate peaks and troughs of the financial cycle measured by the combined behaviour of the component series (credit, the credit to GDP ratio and house prices). The blue line traces the financial cycle measured as the average of the medium-term cycle. The red line traces the GDP cycle. Source: Borio (2014)

6

Figure 2. The Financial Cycle in the UK

Note: Orange and green bars indicate peaks and troughs of the financial cycle measured by the combined behaviour of the component series (credit, the credit to GDP ratio and house prices). The blue line traces the financial cycle measured as the average of the medium-term cycle. Source: Borio (2014)

3. Regulatory capital and accounting Capie and Wood (2013) provide an excellent overview of bank capital in the UK. During the early-mid nineteenth century, there were a number of banking crises. Banks responded to these crises by holding higher levels of capital. An analysis of bank capital shows that banks adjusted their capital ratios according to the risks
that they were taking and that they were well capitalised in comparison with the later standards set by the Basel regulators. Capie and Wood (2013) argue that these capital rules led banks trying to game the rules.

The concept of regulatory capital, often described as capital adequacy requirements, was only introduced in the 1970s. The UK moved to a risk-weighted capital ratio in 1975 (Goodhart, 2011). The aim of risk-weighted assets (RWAs) is to move from a static capital requirement to a requirement based on the riskiness of a bank’s asset classes i.

Risk

weights range from 0 per cent for assets that are deemed the ‘safest’ to a 100 per cent for the riskiest assets. For example, EU authorities allow for a zero risk weight being applied to EU sovereign exposures. The risk-weighted capital ratio (RWCR) is: (1)

∑

The level of 8 per cent is explained below. It is clear that such a framework invites regulatory arbitrage, but that is beyond the scope of this paper. During my spell at the Supervisory

7

Policy Division at the Bank of England in the 1990s, we got regular visits from parties arguing that a lower risk weight should apply to them.

The question of capital adequacy moved into the international policy arena in the early 1980s (Goodhart, 2011). There were concerns about the rate of growth of international lending. In particular, Japanese banks -which had low capital levels- were competing for larger market shares. To create a level playing, the Basel Committee on Banking Supervision introduced the Basel Capital Accord in 1988.

The basic purpose of regulatory capital is to absorb losses in order to protect other claimholders, especially deposit holders (Dewatripont and Tirole, 1994). Regulatory capital comes in different forms that serve different purposes. There are two types of capital: 

Going concern capital: this allows an institution to continue its activities and helps to prevent insolvency. Tier 1 capital is considered to be the going concern capital. The purest form is Common Equity Tier 1 (CET1) capital.



Gone concern capital: this helps ensuring that depositors and senior creditors can be repaid if the institution fails. This category of capital includes hybrid capital and subordinated debt. Gone concern capital is named Tier 2 capital.

Tier 1 capital is seen as the best form of capital, with Common Equity Tier 1 (CET 1) instruments, mainly common shares and retained earnings, being the predominant form of Tier 1 capital. Tier 2 capital is known as supplementary capital. The latter consists of undisclosed reserves, revaluation reserves, general provisions, hybrid instruments (which combine certain characteristics of equity and certain characteristics of debt) and subordinated debt. At the same time, there are certain deductions from regulatory capital, such as goodwill and investments in bank and other financial subsidiaries (Basel Committee on Banking Supervision, 1988). Equity capital consists of issued and fully paid ordinary shares/common stock and non-cumulative perpetual preferred stock (but excluding cumulative preferred stock), and disclosed reserves (including retained earnings). This key element of regulatory capital is the only element common to all countries' banking systems and wholly visible in the published accounts.

Basel I introduces subordinated term debt (as well as the hybrid instruments) in addition to core equity capital (Basel Committee on Banking Supervision, 1988). The Basel Committee recognises that subordinated term debt instruments have significant deficiencies as constituents of regulatory capital in view of their fixed maturity and inability to absorb losses except in a liquidation. The Committee puts therefore restrictions on the amount of such debt

8

capital. Only subordinated term debt instruments with a minimum original term to maturity of over five years may be included within the supplementary elements of capital, up to a maximum of 50 per cent of the core equity capital. More generally, Basel I sets the target standard ratio of capital to risk-weighted assets at 8 per cent, of which the core equity capital was at least 4 per cent (Basel Committee on Banking Supervision, 1988)

Basel II introduces a third tier of capital (Tier 3) in the form of short-term subordinated debt to meet a proportion of the capital requirements for market risks (Basel Committee on Banking Supervision, 2006). While subordinated term debt has an original maturity of at least five years, this new short-term subordinated debt needs only an original maturity of at least two years. It is subject to a lock-in clause, which stipulates that neither interest nor principal may be paid (even at maturity) if such payment means that the bank falls below or remains below its minimum capital requirement.

Differences between accounting and regulatory capital Having discussed the regulatory capital concept, the accounting concept of capital refers to equity capital. There are several differences between accounting and regulatory capital. We discuss here two major differences. A first difference is the use of debt in regulatory capital. The tax-deductibility of interest has spurred the push towards the increasing use of debt as regulatory capital, in an attempt to have the best of both worlds: Telling the regulator that these ‘capital instruments’ can absorb losses, like equity, while at the same telling the taxman that these instruments are debt, so that interest payments can be deducted for corporate tax. Nevertheless, the crisis has led to a rethink of the use of debt as regulatory capital. Some forms of regulatory capital did not absorb losses. Notably subordinated debt would only absorb losses in case of insolvency, but not in the case of bailout, which was the commonly used instrument of government support during the crisis.

In this case, the regulatory deviation from straight equity capital should be adjusted. Basel III is reinforcing the central role of equity (CET1) in regulatory capital (see section 4).

A second difference is the use of fair value accounting for financial instruments. The accounting profession, represented in the International Accounting Standards Board, promotes fair value accounting. Since 2005, firms have to apply the International Financial Reporting Standards (IFRS), which require fair value accounting for financial instruments (IAS 39 and IFRS 7). Allen and Carletti (2008a and 2008b) have reviewed the merits of mark-to-market accounting for financial institutions. Many argue that market prices provide the best estimate of value available and should always be used. However, others suggest

9

that in times of crisis market prices are not a good reflection of value and their use can lead to serious distortions. Allen and Carletti (2008b) distinguish between the circumstances where market prices do reflect future earning power and those where market imperfections imply that they do not. They suggest that in financial crisis situations where liquidity is scarce and prices are low as a result, market prices should be supplemented with both model-based and historic cost valuations. The rest of the time and in particular when asset prices are low because expectations of future cash flows have fallen, mark-to-market accounting should instead be used.

Similarly, the Basel Committee on Banking Supervision (2009) has drawn some accounting lessons from the financial crisis. The Basel Committee argues that the IAS 39 standard for financial instruments should: (a) reflect the need for earlier recognition of loan losses to ensure robust provisions; (b) recognise that fair value is not effective when markets became dislocated or are illiquid; (c) permit reclassifications from the fair value to the amortised cost category; this should be allowed in rare circumstances following the occurrence of events having clearly led to a change in the business model; and (d) promote a level playing field across jurisdictions.

More generally, the Basel Committee (2009) argues that fair value should not be required for items, which are managed on an amortised cost basis in accordance with a bank’s business model. In this case, the accounting treatment has rightly been changed. Amending IAS 39 and IFRS 7, the International Accounting Standards Board (IASB, 2008) has permitted the reclassification of securities out of the trading category in rare circumstances. The IASB has also permitted the reclassification to loan category (cost basis) if intention and ability to hold for the foreseeable future (loans) or until maturity (debt securities). These reclassifications of some financial instruments, which were already permitted under US GAAP, allow financial institutions to discontinue fair value accounting for these instruments.

4. Basel III capital framework The global financial crisis did painfully show that large parts of the regulatory capital could not absorb losses on a going concern basis. Moreover, the widespread use of internal models led to very low levels of regulatory capital. These models were used to reduce the risk weights in the risk weighted capital requirements. Figure 3 highlights this massive reduction in risk weights. While risk weighted capital requirements stayed flat, the actual level

10

of capital -appropriately measured by the (unweighted) leverage ratio- declined sharply from 1994 up to 2007.

The main purpose of Basel III is to improve both the quality and the level of capital (Basel Committee on Banking Supervision, 2011). On the quality side, the key role of Common Equity Tier 1 (CET1) is reinforced. Rating agencies are also looking at CET1 ratios to assess the soundness of banks. Looking more closely, there is distinction between the accounting concepts of tangible equity and intangible equity. The latter concept includes intangible assets. Basel III (just like Basel I and II) requires that goodwill and all other intangibles must be deducted in the calculation of CET1. In other words, Basel III uses the stringent definition of tangible equity. In addition, Basel III will require -as a new element- that deferred tax assets (DTAs) that rely on future profitability of the bank to be realised are to be deducted in the calculation of CET1. Moreover, a bank’s holding of its own common shares are deducted from CET1 (unless already derecognised under the relevant accounting standards).

Figure 3. Average risk weights and leverage ratio of large banks

Note: The series represent the weighted averages across a sample of 17 global banks (Bank of America, Barclays, BNP Paribas, Bank of New York Mellon, Citigroup, Commerzbank, Deutsche Bank, HSBC, ING, JPMorgan, Lloyds Banking Group, Royal Bank of Scotland, Santander, State Street, UBS, UniCredit and Wells Fargo). Leverage ratio measured as Tier 1 capital/total assets. Source: Bank of England (2014)

11

Higher levels of capital In addition to the quality, Basel III also substantially increases the level of capital. Previously, banks had to hold total capital of at least 8 per cent of RWAs. But the minimum requirement for CET1 capital was only 2 per cent. CET1 has been increased from 2 per cent to 4.5 per cent, while total Tier 1 capital should be 6 per cent. This produces additional Tier 1 capital of 1.5 per cent, which may include cocos (see below). Tier 2 capital should be at least 2 per cent. Furthermore, several new buffers have been introduced. Figure 4 provides an overview, including the new buffers.

Figure 4. The new Basel III capital framework

The capital conservation buffer is a capital buffer of 2.5 per cent of total exposures of a bank that needs to be met with an additional amount of CET1 capital. It sits on top of the 4.5 per cent CET1 capital requirement. As its name indicates, the buffer’s objective is to conserve a bank’s capital. When a bank breaches the buffer, i.e. when its CET1 capital ratio falls below 7 per cent, automatic safeguards kick in and limit the amount of dividend and bonus payments a bank can make. The further the bank “eats” into the buffer, the stricter the limits become. This prevents the bank’s capital to be further eroded by such payments.

12

Next, the purpose of the countercyclical buffer is to counteract the effects of the economic cycle on banks’ lending activity, thus making the supply of credit less volatile and possibly even reduce the probability of credit bubbles or crunches, as discussed in section 2. In good times, i.e. where an economy is booming and credit growth is strong, it requires a bank to have an additional amount of CET1 capital between 0 and 2.5 per cent. This prevents that credit becomes too cheap and that banks lend too much. If a bank does not have enough capital to fill this buffer, the same restrictions as in the case of the capital conservation buffer kick in. When the economic cycle turns, and economic activity slows down or even contracts, this buffer can be “released” (i.e. the bank is no longer required to have the additional capital). This allows the bank to keep lending to the real economy or at least reduce its lending by less than would otherwise be the case. Moreover, the countercyclical buffer can also fulfil its function of absorbing losses and can thus be depleted.

Finally, the Financial Stability Board (FSB) introduces a mandatory systemic risk buffer of CET1 capital for banks that are identified as globally systemically important banks (G-SIBs). The identification criteria and the allocation into categories of systemic importance include size, cross border activities and interconnectedness (Basel Committee, 2013). The mandatory surcharge will be between 1 and 3.5 per cent CET1 capital of RWAs.

For a major bank, total CET1 can move from 2 per cent to 11.5 per cent (assuming a full countercyclical buffer of 2.5 per cent and an average G-SIB surcharge of 2 per cent). The total capital requirement (RWCR) will then move to 15 per cent of risk-weighted assets.

A new leverage ratio In addition to risk-weighted capital ratio (RWCR), Basel III introduces a minimum leverage ratio, i.e. the ratio between a bank’s non-risk-weighted assets and its Tier 1 capital, as a supplementary measure to the risk-based framework of Basel II. The leverage ratio (LR) is defined as follows: (2)

∑

The leverage ratio is intended to be a hard backstop against the risk-based capital requirements (Equation 1) and is also designed to constrain excess leverage, which was common amongst many banks before the crisis. This excess leverage also contributed to the depth of the financial crisis, as explained in section 2. The capital measure is made up of Tier 1 capital and is currently set at a minimum of 3 per cent. But more data and experience will

13

be gathered before an effective leverage ratio is introduced as a binding requirement in each jurisdiction (Basel Committee, 2011).

Nevertheless, some interesting applications of the leverage ratio are emerging. The first is on the rationale for the systemic surcharge for G-SIBs. If one increases the risk-weighted capital ratio, then one should also increase the leverage ratio (Schoenmaker, 2013). Several countries, such as Switzerland, Sweden, the Netherlands, the UK, and the US, are applying this logic, with a systemic surcharge on the leverage ratio. The second is a time-varying component, which will produce a countercyclical leverage ratio buffer. The Financial Policy Committee (FPC) at the Bank of England has recently proposed this innovation, which is subsequently approved by HM Treasury (Bank of England, 2014).3

This new approach towards the leverage ratio is to be applauded. First, it uses Tier 1 capital (i.e. equity) and thus excludes Tier 2 debt elements. In that way, there is a close alignment between accounting and regulatory capital. Second, several studies find that the leverage ratio is a better predictor of bank failure than the risk-weighted capital ratio (e.g. IMF, 2009). Third, it incorporates the macroprudential lessons: 1. a countercyclical buffer to dampen credit growth; and 2) a systemic risk surcharge to reduce the TBTF subsidy.

The new leverage ratio will buck the trend of declining capital levels in banking. The FPC estimates that the calibration of the new leverage ratio could be up to 4 per cent for the major UK banks and to almost 5 per cent in case of full application of the countercyclical buffer (Bank of England, 2014). Figure 5 illustrates the long-term trend of capital in banking, measured as the leverage ratio (Tier 1 capital/total assets). Figure 5. The leverage ratio of UK banks To be included

There is a heated debate in academia on the appropriate level of regulatory capital in banks. Admati and Hellwig (2013) and Miles et al. (2013), for example, argue for unweighted capital ratios of up to 20 per cent. Their argument for higher (equity) capital requirements is twofold: 1) to reduce the probability of a public bailout and thus reduce the exposure of taxpayers; and 2) to reduce incentives for managers to take excessive risks (and play games with risk

3

The FPC uses a 35 per cent conversion factor from RWCR to LR buffers. Using Tier 1, the RWCR (the minimum Tier 1 requirement and the conservation buffer) is 8.5 per cent and the LR is 3 per cent. The conversion factor is then 8.5/3=0.35.

14

weights). On the first, they make a clear distinction between costs to individual institutions (private costs) and overall economic (or social) costs (see section 2). Miles et al. (2013) estimate the long-run costs and benefits of having banks fund more of their assets with lossabsorbing equity capital. The benefits are related to the reduction of the chance of banking crises, which generate substantial economic costs. The offset to any such benefits comes in the form of potentially higher costs of bank intermediation (e.g. higher lending rates or lower savings rates). They find that the amount of equity capital that is likely to be desirable for banks to use is very much larger than banks have used in recent years and also higher than targets agreed under the Basel III framework.

By contrast, DeAngelo and Stulz (2013) stress the liquidity function of banks and argue for lower levels of equity capital. High leverage is optimal for banks in order to have a meaningful role in liquid-claim production (i.e. demandable deposits). Their model has a market premium for (socially valuable) safe/liquid debt in the form deposits, but no taxes or other traditional motives to lever up. Because only safe debt commands a liquidity premium, banks with risky assets use risk management to maximise their capacity to include such safe debt in the capital structure. In that way, the model of DeAngelo and Stulz (2013) explains that banks have higher leverage than most industrial firms, and also that leverage limits for regulated banks impede their ability to compete with unregulated shadow banks.

Finally, it may be useful to look beyond capital ratios based on book values and look at market based ratios, using stock market data (Acharya et al., 2012). A market based leverage ratio can be calculated as a bank’s market capitalisation divided by that bank’s total assets. Developments in this market based leverage ratio provide additional information to supervisors and investors about the soundness of a bank. The book and market value approaches are complementary.

Bail-in of debt and cocos The regulatory capital debate has moved beyond Basel III. To reduce the exposure of taxpayers, ministers of finance have promoted the concept of bail-in debt. The Bank Recovery and Resolution Directive (BRRD Directive 2014/59/EU) permits recapitalisation through the write-down of liabilities and/or their conversion to equity. This would allow the bank to continue as a going concern, while avoiding the disruption to the financial system caused by stopping or interrupting its critical services, and giving the authorities time to reorganise it or wind down parts of its business in an orderly manner. If a bank needs to resort to bail-in, authorities would first write down all shareholders and would then follow a

15

pre-determined order in bailing in other liabilities. Shareholders and other holders of instruments such as convertible bonds and junior bonds would bear losses first.

In addition, Basel III allows cocos, which are contingent convertible bonds that convert to equity if the regulatory capital ratio drops below a certain pre-determined threshold. Cocos can be issued as additional Tier 1 capital (as well as Tier 2 capital), and can thus contribute up to 1.5 per cent of the risk weighted capital ratio (Basel Committee, 2011).

It looks like history is repeating itself. Before the crisis, subordinated debt was promoted by academics because of its disciplinary function (e.g. Flannery, 2001). As banks with higher asset risk have to pay higher interest rates on subordinated debt, such debt can induce banks to lower asset risk in order to reduce interest payments. Next, indirect discipline may happen when regulators take prompt corrective actions against banks with high subordinated debt yields or banks unable to roll over subordinated debt. These corrective actions may not only prevent further losses of problem banks, but also stop bank managers from pursuing unsound risk.

But it did not work as envisaged. First, subordinated debt yields were only partly rising, as investors (with hindsight rightly) expected to be bailed out. Next, subordinated debt (and future bail-in debt) may work in the case of an idiosyncratic failure, but not during widespread banking crisis as authorities may not want to spread contagion by writing down subordinated / bail-in debt.4

Several academics question the financial stability implications of bail-in debt and cocos. Goodhart and Avgouleas (2014) call for a closer examination of the bail-in process, if it is to become a successful substitute to the unpopular bailout approach. They argue that bail-in regimes will fail to eradicate the need for an injection of public funds where there is a threat of systemic collapse, because a number of banks have simultaneously entered into difficulties, or in the event of the failure of a large complex cross-border bank, except in those cases where failure was clearly idiosyncratic.

4

A good example is ING: doubt arose about the interest payment on the subordinated debt of ING at the height of the financial crisis in Autumn 2008. Some investors questioned publicly whether ING would meet the upcoming interest payments on its supervisor did not permit ING to say so as payments on subordinated debt are conditional on meeting certain capital ratios at the time of payment. After a sharp drop in the share price, the supervisor gave special permission to ING to announce that it was planning to meet its upcoming interest payments (Avgouleas et al., 2013).

16

Similarly, Chan and Van Wijnbergen (2014) show that while the coco conversion of the issuing bank may bring the bank back into compliance with capital requirements, it will nevertheless raise the probability of the bank being run, because conversion is a negative signal to depositors about asset quality. Moreover, conversion imposes a negative externality on other banks in the system in the likely case of correlated asset returns, so bank runs elsewhere in the banking system become more probable too and systemic risk will actually go up after conversion. Cocos thus lead to a direct conflict between micro- and macroprudential objectives.

So, there is a limit to the extent that bail-in debt or contingent convertible capital can replace real upfront equity capital.

5. Conclusions and policy recommendations The global financial crisis has highlighted that deviations between the accounting and regulatory concepts of capital have gone too far, on both sides. Accounting standards have been overshooting in the application of fair value accounting. If there are no markets during times of crises, it does not make sense to mark-to-market. These exceptions -in the form of reclassification under IAS 39 and IFRS 7- have now been included in the accounting standards.

Regulatory capital has gone astray by allowing Tier 2 (subordinated term debt) and Tier 3 (subordinated short-term debt), which did not absorb losses during the crisis. Basel III is rightfully reinforcing the central role of Common Equity Tier 1 and abolishing Tier 3. Moreover, the smart use of risk weight lowered actual levels of capital. Basel III is thus complementing the risk weighted capital ratio with an unweighted leverage ratio. More generally, the use of complementary approaches is useful to gain more information for supervisor and investors.

It is important to define the objective(s) of regulatory capital very carefully. A stated objective is to absorb losses, while preventing failure and thus protecting depositors (and policyholders). We support the use of high quality of capital in Basel III. It is better to have real equity. At the same time, we question the popularity of cocos, which are debt instruments that can convert to equity. This conversion may or may not happen. In times of crisis, supervisors may be reluctant to allow conversions and more broadly to bail-in debt to avoid sending a bad signal to markets and depositors. If that were to happen, we would be back at square one: only real equity absorbs losses and all kind of fancy ‘debt’ instruments,

17

from subordinated debt to cocos, escape losses. This reinforces the point of having real equity to absorb losses.

A neutral tax treatment of debt and equity would reduce the incentive for more debt and less equity. Allen et al. (2011) recommend to abolish the interest rate deductibility for corporate taxes. That would reduce the incentive for debt financing. As the Group of Twenty (G-20) ministers of finance are in charge of the regulatory reform agenda at the FSB, they would be able to implement this recommendation in their tax codes.

From the macro-prudential side, there are two main concerns arising from the financial crisis. The first is cyclical. Minsky (1986) argues that the boom-bust cycle of credit lies at the heart of financial instability. An important innovation of the new Basel III framework is the introduction of the countercyclical buffer and the leverage ratio, which will help to counter the building up of the credit cycle. The second concern is addressing too-big-to-fail (TBTF). Basel III is increasing capital levels, with a systemic surcharge for the large global banks. This will help to reduce the TBTF subsidy to large banks.

The overall conclusion is that the accounting and regulatory concepts of capital are converging. In particular, the increasingly prominent role of the leverage ratio aligns the two concepts. Both the Basel Committee and the International Accounting Standards Board have taken their responsibility. But the use of cocos, debt instruments that may (or may not) convert to equity, is again causing unwelcome deviations.

18

References Acharya, V., R. Engle, and M. Richardson (2012), Capital Shortfall: A New Approach to Ranking and Regulating Systemic Risks, American Economic Review 102, 59-64. Admati, A. and M. Hellwig (2013), The Bankers' New Clothes: What's Wrong with Banking and What to Do about It, Princeton University Press, Princeton. Adrian, T. and H. Shin (2010), Liquidity and leverage, Journal of Financial Intermediation 19, 418–437. Allen, F. and E. Carletti (2008a), Mark-to-Market Accounting and Liquidity Pricing, Journal of Accounting and Economics 45, 358-378. Allen, F. and E. Carletti (2008b), Should Financial Institutions Mark-to-Market?, Financial Stability Review 12, 1-6, Banque de France. Allen, F., T. Beck, E. Carletti, P. Lane, D. Schoenmaker, and W. Wagner (2011), CrossBorder Banking in Europe: Implications for Financial Stability and Macroeconomic Policies, CEPR Report, London. Avgouleas, E., C. Goodhart, and D. Schoenmaker (2013), Bank Resolution Plans as a catalyst for global financial reform, Journal of Financial Stability 9, 210-218 Bank of England (2014), The Financial Policy Committee’s Review of the Leverage Ratio, London. Basel Committee on Banking Supervision (1988), International convergence of capital measurement and capital standards, Basel, Switzerland. Basel Committee on Banking Supervision (2006), Basel II: International Convergence of Capital Measurement and Capital Standards: a Revised Framework, Comprehensive Version, Basel, Switzerland. Basel Committee on Banking Supervision (2009), Guiding principles for the replacement of IAS 39, Basel, Switzerland. Basel Committee on Banking Supervision (2011), Basel III: A global regulatory framework for more resilient banks and banking systems - revised version June 2011, Basel, Switzerland. Basel Committee on Banking Supervision (2013), Global systemically important banks: updated assessment methodology and the higher loss absorbency requirement, Basel, Switzerland. Borio, C. (2014), The Financial Cycle and Macroeconomics: What Have We Learnt?, Journal of Banking and Finance 45, 182-198. Capie, F. and G. Wood (2013), Do we need regulation of bank capital? Some evidence from the UK, IEA Current Controversies No. 40. Chan, S. and S. van Wijnbergen (2014), Cocos, Contagion and Systemic Risk, Duisenberg school of finance - Tinbergen Institute Discussion Paper, TI 14-110/ DSF 79. Damodaran, A. (2014), Applied Corporate Finance, 4th Edition, John Wiley and Sons, New York. DeAngelo, H. and R. Stulz (2013), Liquid-Claim Production, Risk Management, and Bank Capital Structure: Why High Leverage is Optimal for Banks, Fisher College of Business WP 2013-03-08. De Haan, J., S. Oosterloo, and D. Schoenmaker (2012), Financial Markets and Institutions: A European Perspective, Second Edition, Cambridge University Press, Cambridge, UK.

19

Dewatripont, M. and J. Tirole (1994), The Prudential Regulation of Banks, MIT Press, Cambridge (MA). Flannery, M. J. (2001), The faces of “market discipline”, Journal of Financial Services Research 20, 107-119. Goodhart, C. (2011), The Basel Committee on Banking Supervision: A History of the Early Years, 1974–1997, Cambridge University Press, Cambridge (UK). Goodhart, C., P. Hartmann, D. Llewellyn, L. Rojas-Suarez, and S. Weisbrod (1998), Financial Regulation: Why, How and Where Now?, Routledge, London. Goodhart, C. and E. Avgouleas (2014), A Critical Evaluation of Bail-ins as Bank Recapitalisation Mechanisms, CEPR Discussion Paper No 10065. Huang, R. H. and D. Schoenmaker (eds.) (2014), Institutional Structure of Financial Regulation: Theories and International Experiences, Routledge, London. International Accounting Standards Board (2008), Reclassification of Financial Assets (Amendments to IAS 39 Financial Instruments: Recognition and Measurement and IFRS 7 Financial Instruments: Disclosures), London. International Monetary Fund (2009), Global Financial Stability Report, April, Washington D.C. Llewellyn, D. (1999), The Economic Rationale for Financial Regulation, FSA Occasional Paper 1, Financial Services Authority, London. Miles, D., J. Yang and G. Marcheggiano (2013), Optimal Bank Capital, The Economic Journal 121, 1-37. Minsky, H. P. (1986), Stabilizing An Unstable Economy, Yale University Press. Schoenmaker, D. (2013), Governance of International Banking: The Financial Trilemma, Oxford University Press, New York.

20