While there are a number of justifications for establishing regulatory capital requirements, including the need to safeguard publicly funded depositor protection schemes as well as the desire to maintain competitive equality between firms operating in international markets (otherwise referred to as the 'level playing field'), the Basel Committee has always emphasised that the primary objective for its capital standard is to protect against systemic risk in international financial markets. We agree that this should remain the primary objective of internationally agreed capital standards for financial firms. However, if this is indeed the primary objective of the proposed regulatory capital regime, we believe that it is essential to achieve a broad level of consensus about the nature of international systemic risk and the way in which capital can act as a protection against it. In this same context it is also essential that there is agreement about the definition of capital itself.
The term 'systemic risk' is widely used and has a number of different definitions. In particular, the term is used to refer both to a sudden crisis of confidence and withdrawal of liquidity from a market as in the case of the collapse of Bankhaus Herstatt in 1974, or the classic bank run of the 19th century as well as to a longer term solvency crisis affecting the banking sector in a country such as the S&L crisis in the US or the current problems facing the Japanese banking system. However, in order to meet the Basel Committee definition the risk has to be international in character and systemic in nature. The most appropriate description of this risk is that of Sir Andrew Large, who when Chairman of the UK Securities and Investments Board described this risk in a speech given in 1996 as, 'The risk that something which goes wrong in one firm or market will, because of the close linkages which now exist between firms and markets, spill over to affect other firms and markets.'
Based on this definition of international systemic risk, we need to understand why maintaining a particular level of capital at individual firms has become generally recognised as the primary regulatory tool to protect against international systemic risk. We also need to consider how this risk has changed as a result of the developments in international markets since the original Basel Accord was agreed in 1988.
In this context it is necessary to recall that at the time the original Basel Accord was drafted the prevalent international regulatory concern was about the undercapitalisation of the Japanese banks. It was widely believed that this was allowing these banks to compete for international lending business at margins which did not adequately compensate for the risks in the business and that the downward pressure on margins which this created threatened the solvency of all other banks involved in international lending. In these circumstances, a common capital standard which increased the capital underpinning required for international lending business was indeed an appropriate regulatory response, since it raised the cost of operating in the market for the previously undercapitalised banks and provided an increased capital buffer against insolvency. Indeed, one measure of the success of the 1988 Accord is the extent to which it achieved an increase in capital in those countries and banks where the regulatory concern was the greatest at the time.
However, when considering the appropriate regulatory approach which the Basel Committee should adopt in order to protect against international systemic risk at the beginning of the 21st century, we need to recognise two significant developments since the introduction of the 1988 Basle Accord both of which have profound implications for the role which capital should play in any future regulatory regime. These are:
- the application of modern risk management and risk measurement techniques to banks' credit portfolios; and
- the increasing importance of non banking activities in international finance
- Modern credit risk management and the distinction between expected and unexpected loss
The introduction of modern risk management techniques in relation to banks' credit portfolios is a comparatively recent development, but is already having a significant effect on the way that banks understand and manage their credit risk and capital. The principal development in this context has been the recognition that credit risk can be regarded as a statistical phenomenon, and consequently that banks can estimate the likely losses associated with a portfolio of credit exposures - an amount referred to as the 'Expected Loss (EL) of the portfolio. This in turn allows the bank to price the cost of its credit exposure more accurately into its transaction pricing, and also to estimate the potential variability of those losses from year to year - a measure often referred to as the Unexpected Loss (UL) of the portfolio, (a term which includes the potential for stress loss events as well as the more normal volatility of credit results from year to year).
This approach to credit risk management in turn alters the way that banks think about the required capital underpinning for their credit business. Under the approach which existed at the time when the 1988 Accord was introduced, banks generally regarded credit losses as unfortunate but largely random. As a result, loans were held on the balance sheet at face value and were provided against only when the borrower actually encountered difficulties in repayment. In this environment it was clearly necessary for banks to maintain a suitable capital buffer to enable these losses to be absorbed as and when they arose.
By contrast under the modern approach to credit risk management, banks recognise that they inevitably face some level of EL associated with their portfolio, and should seek to make appropriate accounting adjustments to recognise this fact. This can be done either by means of a fair value or mark to market adjustment to the value of the asset or, where an accrual accounting convention is being followed, by setting up a suitable provision against the value of the asset which is released as and when the losses actually occur. In both cases, however, the result is the same. Since the EL in the portfolio is regarded as an unavoidable cost of doing business, it needs to be recognised as a business cost and charged against revenues.
Under this new approach, capital is no longer regarded as the primary source of protection against EL, but is available instead to protect against the UL of the portfolio. This recognition that capital should not be used to protect a firm against the EL in its portfolio is a comparatively recent development, but is an unavoidable consequence of adopting modern credit risk management techniques. However, this development has yet to be reflected in the Basel Committee approach, since the current 8% standard and the stated wish to maintain a given level of capital in the financial system is still based on the assumption that capital exists to cover a bank against the EL as well as the UL in its credit portfolio.
- The growing systemic importance of liquidity risk
The second important development which affects any consideration of the appropriate regulatory approach to address international systemic risk is the fact that both the institutions and the types of risks which now characterise international financial markets are substantially different from those which existed at the time of the introduction of the 1988 Accord. In particular, traditional international bank lending is a much smaller part of cross border finance, with traded products including securities and OTC derivatives playing a much more significant role than previously. In response to these trends, international banks have diversified their activities into new areas such as investment banking, fund management and insurance where the risks are materially different from those of the traditional credit business. At the same time, major new non bank competitors have emerged which compete with banks in a number of these activities and which face similar risks, but are not subject to a credit based approach to setting capital adequacy standards.
Indeed, as recent events have underlined, the changes in international finance mean that it is increasingly likely that international systemic risk will derive from events in securities and derivatives markets rather than in the market for bank credit. The Asian and Russian crises in particular demonstrated that changes in liquidity preferences and previously unidentified risk concentrations can give rise to new sources of systemic risk. However, these types of systemic event are materially different from a traditional banking type of systemic crisis because of the speed with which the problems impact the financial system as a result of the interrelationships between the different parties involved.
Thus, in comparison with the situation when the 1988 Accord was established, there is little if any evidence that undercapitalisation of internationally active banks is currently a likely source of systemic risk, or that the primary systemic concern is one of solvency rather than liquidity. By contrast the primary public policy concern in dealing with a potentially systemic crisis in recent years has been the need to maintain adequate levels of liquidity while the crisis is resolved. Moreover, as the case of LTCM showed clearly, the sources of international systemic risk are not limited to firms which are regulated as banks but can derive from any firm which plays a sufficiently significant part in international markets.