Progress towards a Framework for Financial Stability Assessment
Speech by José Manuel González-Páramo,
Member of the Executive Board of the ECB
OECD World Forum on “Statistics, Knowledge and Policy”
Istanbul, 28 June 2007
I would like to thank the organisers for inviting me to participate in this conference covering such a wide breadth of fascinating and important issues, particularly those related to financial stability that we are focussing on in this session . After enduring significant financial turmoil at the start of the decade, it is probably fair to say that Turkey is one of the places in the world where the safeguarding of financial stability is more treasured by policy makers and by the general public.
It goes without saying that, while it is crucial for public authorities responsible for financial stability (including central banks, supervisory authorities and finance ministries) to develop mechanisms that contribute to the resolution of crises once they materialise, the primary objective of financial stability policies should be to prevent crises. This is obvious if we consider the potential costs of financial crises for the society in terms of foregone output and disruptions to the orderly functioning of the economy. At the same time, it should be noted that establishing an effective policy framework for crisis prevention by no means guarantees that financial crises will be eliminated forever. For that, systemic risk would need to be eliminated altogether, which is of course not possible.
Indeed, financial systems cannot be expected to be effective shock-absorbers under all conditions and at all times. In fact, history shows that vulnerabilities in financial systems and the accumulation of imbalances in other sectors of the economy can create sources of risk for macroeconomic and financial stability. And there have been occasions when, following some form of pro-cyclical financial behaviour or the accumulation of “excesses” during a boom, financial systems have amplified rather than mitigated the effects of shocks. However, through their commitment to the objective of preserving financial stability, I believe that public authorities can reduce the likelihood of crises and contain the potentially disruptive impact of the latter, should they emerge.
The maintenance of financial stability depends in the first instance on adequate financial regulation (aiming to provide the right structure of incentives for prudential risk behaviour by financial institutions) as well as effective prudential supervision. In my remarks today, I would like to focus on another important tool for crisis prevention: the systematic monitoring and assessment of financial stability conditions aiming at detecting key sources of risk and vulnerability for financial systems. This tool is at the core of the financial stability tasks pursued by central banks, including those which do not have supervisory responsibilities. It focuses on the financial system as a whole as well as on its main components and is normally reflected in the Financial Stability Reviews regularly published by central banks.
As you know, since December 2004 the ECB has published a semi-annual Financial Stability Review for the euro area. My main objective today is to briefly present the framework which underpins the analysis presented in this review. First, I will explain why central banks, including the ECB, have developed a financial stability assessment. Then I would like to focus on one important conceptual issue, notably the definition of financial stability. I will briefly sketch how we tackle the non-trivial task of measuring the degree of stability in the financial system at the ECB. I will finally point to some of the challenges that central banks around the world are facing in their financial stability assessment.
2. The need for financial stability assessment
Over the past decade a growing number of central banks around the world have begun to monitor and assess financial stability and many of them now periodically publish their findings in Financial Stability Reviews. An important reason behind this development has been the increased frequency of financial crises in the 1990s, both in emerging and mature economies.  These crises have shown that risks and vulnerabilities for the financial sector stem not only from developments within the sector itself but also - and to a rather large extent - from changes in the external macro-economic and financial environment in which the sector operates.
One of the main lessons drawn from these crises is that it is important to complement the supervisory monitoring of individual institutions with a regular monitoring of conditions in the system as a whole. In virtue of the expertise acquired in analysing macro-economic and financial developments for monetary policy purposes, central banks are the natural candidates to perform this task. Central banks have a special interest in maintaining financial stability, since a stable financial system is an important pre-condition for ensuring the smooth and efficient transmission of monetary policy, ultimately contributing to the achievement of the price stability goal.
In the euro area, the Eurosystem has undertaken financial stability monitoring and assessment activities for the area as a whole. In doing so, the Eurosystem aims to fulfil its statutory role of contributing to financial stability.  As regards the internal labour division within the Eurosystem, the ECB Financial Stability Review complements the reviews produced by the national central banks by providing a financial stability analysis from an area-wide perspective.
In general, the financial stability assessment of central banks is confronted with two main difficulties. The first difficulty relates to the increasing complexity and interdependencies of financial systems due to rapid pace of innovation and the ongoing globalisation and integration of markets. These phenomena may complicate the interpretation of the existing indicators, thereby putting a premium on the need of improving and expanding the statistical framework in order to ensure that it remains fit for purpose. The second difficulty refers to the lack of an analytical framework for financial stability assessment as developed as the one available for monetary policy purposes. This reflects the fact that analytical efforts in the domain of financial stability are still in their infancy, though they have made marked progress in recent times. But it also mirrors the undeniable conceptual difficulties inherent in this field, starting with the definition itself of financial stability. It is encouraging that many academic institutions, international organizations and central banks, including the ECB, are currently investing resources in developing a sound analytical framework.
3. The definition of financial stability
Perhaps the most important building block for an analytical framework designed to measure, monitor and safeguard financial stability is to formulate a working definition of financial stability. For policy objectives such as maintaining price stability or fiscal stability, this is a more straightforward task. For financial stability it is less, since there is not even a precise and universally recognised definition of “financial stability”. We have worked hard on this at the ECB and the definition we have adopted reads as follows:
“ Financial stability can be defined as a condition in which the financial system – comprising of financial intermediaries, markets and market infrastructure – is capable of withstanding shocks and the unravelling of financial imbalances, thereby mitigating the likelihood of disruptions in the financial intermediation process which are severe enough to significantly impair the allocation of savings to profitable investment opportunities.” 
This definition encompasses three key elements: financial system, financial stability and a notion of what we mean by financial instability. All of these are necessary to set the boundaries of the analysis.
Going through each of them in detail, the financial system comprises of three distinct but closely-linked components:
the financial intermediaries, i.e. the institutions that pool funds and risks and allocate them to competing uses, including banks, insurance companies and pension funds, but also other institutions (e.g. hedge funds);
the financial markets which enable the matching of savers and investors and/or risk sellers and risk buyers, thereby serving as an alternative source of finance to financial institutions and allowing risks to be re-distributed among financial intermediaries;
the financial infrastructure, including clearance, payment and settlement systems.
As both the individual components and the linkages among them constitute the financial system, the maintenance of financial stability requires their collective stability. Vulnerabilities in one component may affect the smooth functioning of others, thereby posing a threat to the smooth functioning of the financial system as a whole.
The ECB’s understanding of financial stability originates from the notion that there are three distinct aspects of a stable financial system.  First, the function of inter-temporal allocation of resources from savers to investors should be facilitated efficiently and smoothly. Second, financial risks should be assessed and priced with a reasonable degree of accuracy and also efficiently managed using a forward-looking approach. Third, the financial system should be in a position to comfortably absorb both financial and real economic shocks. Should any of these conditions not be satisfied, the financial system would most likely become unstable.
The consequences of defining financial stability in this particular way are very interesting. Among them, the inter-temporal and forward-looking aspects of the definition imply that threats to financial stability can arise not only from shocks but also from disorderly adjustments of imbalances built up in the past, perhaps as a result of over-optimistic expectations about future returns or the mis-pricing of risks.
4. Assessment of financial stability
Let me now turn to the issue of the rigorous and accurate assessment of the degree of financial stability. The main objective of this assessment is the early identification of risks and vulnerabilities that could threaten the smooth functioning of the financial system, interrupting financial intermediation and imposing real economic costs. A list of the potential sources of risks to financial stability could be quite long.
This list would include some risks that are endogenous to the financial system, i.e. related to one of its components. When talking about endogenous risks, we should consider those related to: (1) financial institutions, including credit, liquidity or reputation risks; (2) financial markets, such as counterparty risk, contagion or asset price misalignments; and (3) infrastructures, i.e. legal or regulatory risk, domino effect or systemic risk in payment, clearance or settlement systems. Sources of risks which are exogenous to the financial system, particularly those related to macroeconomic disturbances as well as policy actions or disaster events, are of course also very important.
In practice, the process of assessing financial stability can be divided into three steps. In the first step, the overall robustness of the financial system is assessed, taking into account the endogenous sources of risk. In the second step, the main sources of exogenous risks and vulnerabilities to financial stability should be identified and their probabilities should be evaluated. The third and most challenging step, based on the outcome of the assessment undertaken under the first two steps, is to evaluate the ability of the financial system to absorb shocks, should the identified risks materialise.
Indicators for each of these three steps can be identified. However, the choice of the appropriate indicators may depend on the composition of the financial system, especially the relative importance of markets and institutions, as well as the institutional set-up of the system. Indeed, these are significant differences in financial system structures across countries. For instance, in a bank-based financial system, the banks solvency ratios would be of much greater importance than in a market-dominated financial system, in which market stability indicators would seem to be crucial.
Because of this multidimensional and multifaceted nature of financial stability, there is as yet no widely accepted set of measurable indicators of financial stability. For the same reason, an absolute “single number” for the assessment of financial stability, comparable to the rate of increase of a consumer price index for the assessment of price stability, appears to be elusive for the time being. For example, if measures of the stability of the banking sector indicated underlying strength, but at the same time there were signs of strain in financial markets, the overall assessment of the stability of financial system would be ex-ante ambiguous.
Although relatively simple indicators for assessing the soundness of banks have been developed and are widely used, great care is needed in interpreting them.  Many financial stability indicators are designed to detect sectoral imbalances and are often calculated as ratios between different balance sheet items. Because movements in such indicators often have different interpretations, they can be misleading. For instance, increasing bank solvency may indicate improved shock absorption capacity, but it may also reflect risk aversion and the foregoing of profitable lending opportunities. Hence, rather than indicating stability, high solvency could be a signal of weaknesses developing in a bank in a competitive environment, that may potentially entail loss of market share and profit erosion in the future.
A similar duality of interpretation applies to indicators of the robustness of financial markets. Low asset price volatility may be indicative of stable conditions and effective market functioning, but it might also be a signal of failure in the price discovery process. Narrow credit default swap (CDS) spreads, for instance, could reflect a perception of low credit risk but may also result from the under-pricing of risk.
The solution to this identification problem is to cross-check different indicators of the same phenomenon or to combine various measures in composite indicators. However, this may still not be enough to assess financial stability unambiguously, without looking beneath the surface.
Let me give one recent concrete example to explain the issue a bit further. In order to identify the drivers behind recent asset price movements, ECB staff has recently constructed a composite indicator of financial market liquidity in the euro area, based on a methodology developed by the Bank of England. This indicator combines several measures covering four different markets and three different dimensions of market liquidity (tightness, depth and resilience) as well as liquidity premia.
Although the indicator has confirmed earlier market intelligence information that financial market liquidity rose substantially in 2003 and has remained high since then, it does not address the question of what are the implications of abundant liquidity for financial stability. One cannot assess the influence and potential implications of current market liquidity conditions without looking into the sources of the current abundant liquidity. If liquidity in financial markets is largely a reflection of greater risk appetite as opposed to potentially more lasting factors, such as structural changes in financial markets – and here I am thinking about the development of new investment products, mostly credit derivatives, or the entry by hedge funds into traditionally illiquid markets – then it could suddenly fade away in the event of unexpected market stress.
All in all, excessive reliance on any single indicator without taking into account the need for a broader assessment of economic and financial conditions on the basis of a comprehensive set of measures may lead to a potentially unsound assessment of financial stability.
However, it would be unwise to conclude from these observations that financial stability may not be a measurable concept after all. Indeed, even if financial stability is unobservable, we still have statistical techniques at our disposal which can provide quantitative assessments of unobservable phenomena. Although we are far from agreeing on one single measure of financial stability, for the reasons I have just listed, we have made recently progress in quantifying some unobservable variables, which could be included in the information set available to policy-makers for assessing financial stability. For instance, in the most recent issue of the ECB Financial Stability Review, which was published earlier this month, we have introduced a single indicator of investors’ risk appetite derived from a set of market-based and model-based measures of risk aversion.
An important part of any financial stability assessment is to gauge the shock absorption capacity of the financial system, i.e. its ability to withstand shocks, should they materialise. Following the introduction of the IMF Financial Sector Assessment Program (FSAP), stress-testing has become a standard part of the financial stability assessment tool-kit among central banks, supervisors and financial institutions themselves. Although important advances have been made in this field, it is important to be aware of the limitations of stress-testing. For instance, the approaches commonly applied usually do not include the second round effects of shocks to the financial system resulting, for instance, from the probable tightening of lending terms and conditions by banks. Moreover, the models are usually based on log-linear relationships, whereas situations of financial stress are usually created by so-called tail events. In such situations, the relationship between variables may be non-linear.
Notwithstanding these caveats, stress-testing is an important tool in the process of risk management at a micro-level, particularly in banks. By encouraging financial institutions to carry out stress-testing exercises - also by publishing potential risk scenarios in their Financial Stability Review - central banks promote the ability of the institutions to withstand shocks and thus prevent financial instability.
5. The approach followed at the ECB
I would like to briefly sketch how we go about assessing financial stability at the ECB. It is very broad in scope and encompasses a broad set of indicators, covering the external environment, the euro area economic outlook, the balance sheet conditions of non-financial corporation and households, financial markets, banking sector and other financial institutions as well as financial system infrastructures.
In particular, a first set of indicators assesses financial sector conditions focusing on banking sector’s balance sheets, profitability, asset quality and capital adequacy. These indicators are largely derived from publicly available financial statements and from (micro) supervisory sources.
A second set of indicators aims to identify the major sources of risk to the financial sector and covers the competitive conditions in the banking sector, credit growth and banking exposure concentrations and more generally, asset price developments, business cycle and monetary conditions as well as indicators of financial fragility in the counterpart sectors to banks (primarily households and non-financial corporations). This set of indicators should benefit in the future from the release for the first time in June 2007 of quarterly euro area accounts that will help to shed light on the financial soundness of the main institutional sectors, particularly households and non-financial corporations, by allowing to compute a large variety of indicators, including measures of household indebtedness and corporate leverage as well as income-based measures of sectoral fragility. These accounts will be released according to a timeline (initially, a 120 day lag relative to the reference period) that will ensure that they are of much greater use for financial stability analysis.
A third set of indicators aims to assess the resilience of the financial sector based on its risk-absorption capacity given its financial condition as well as the appreciation by market participants of banks’ risks and the main sources of risk outlined in the first two sets of indicators. For this purpose, a number of these indicators have been developed based mostly on market based information sources. These indicators are assessed together with additional qualitative information in order to develop an overall assessment of financial stability.
Nevertheless, because the euro area financial system remains on average more bank-based than market-based in terms of risk-taking activity, we devote quite a bit of attention to the banking sector. For this, we have developed a methodology for identifying large and complex banking groups (LCBGs) whose size and nature of business is such that their failure or inability to operate would most likely influence adversely the smooth functioning of financial markets or other financial institutions operating within the system. Having identified these LCBGs, we periodically monitor their main profitability, solvency and capital-adequacy ratios in order to assess the ability of these institutions to generate profits and withstand shocks.
Other indicators such as returns on risk-weighted assets, trading income relative to Tier 1 capital, loan impairment charges or value at risk (VaR) are calculated on the basis of individual institutions’ financial reports and usually presented in the form of frequency distributions. This is because we are particularly interested in how the financially weakest institutions - the potentially most vulnerable links in the system - are performing.
As the balance sheet and profit and loss account based indicators are backward looking in nature, we support our assessment of the banking sector with forward-looking indicators derived from securities prices. Among these are option-implied risk-neutral distributions of the Dow Jones Euro STOXX bank index, which provide a quantification of the distribution of expected banks’ equity price outcomes, thereby providing a market assessment of downside risks. We also monitor some direct indicators of default risk of LCBGs, including Moody’s KMV expected default frequencies or credit default swaps spreads.
Financial markets are the source of large sets of high frequency data which can provide indications of market strain. I will briefly list only few among the many indicators we use. For instance, bid-ask spreads for EONIA swaps serve as an indicator of liquidity conditions in the euro area money market. The evolution of historical market prices as well as of both historical and implied volatilities can provide information on the overall conditions of different markets, particularly equity markets. As regards government bond markets, in addition to the shape of the yield curve, we monitor the term premia. We also use typically forward looking indicators, such as the options-implied skewness coefficient for German ten-year bonds in order to assess the likelihood of large future movements in long term bonds yield as perceived by the market participants.
6. Challenges for financial stability assessment
The ongoing process of financial development and integration and a wave of financial innovation, especially in credit markets, are posing challenges for the assessment of financial stability.
The development of new financial products based on credit derivatives enables financial institutions to transfer risk to more lightly-supervised agents. There are some concerns that this may, under certain conditions, effectively lead to new threats to financial stability. We should also be aware of the potential problems that credit derivatives can create. These include the possibility that banks might tend to soften their risk and pricing standards more structurally.
At the same time, the increasing degree of development and integration of financial markets is reinforcing the linkages among financial institutions, possibly exposing the financial system to new systemic risks due to rising correlation among asset prices. On the one hand, these developments are enabling financial institutions to achieve a better redistribution of risks, thereby improving their shock absorption capacities. On the other hand, it is possible that this could create the potential for an amplification of shocks at times of market stress.
Overall these elements make the financial stability assessment of domestic financial systems more complex and more dependent on external factors.
Let me now conclude. As I hope I have shown, the ECB has in place a robust approach to financial stability analysis based on an assessment of the condition of the financial system, the sources of risk that could arise and the ability of the system to handle these risks. The framework to undertake this assessment has at its centre a set of indicators which, in turn, draw on a variety of different data sources.
Nevertheless, the analytical foundations of financial stability are relatively new, as evidenced by the absence of a universally accepted definition of stability. Reflecting its novelty, work to develop an appropriate conceptual and statistical framework for financial stability analysis remains in its infancy. While much progress has been made in recent years, we must continue making efforts to improve the theoretical framework and the information dataset underpinning our financial stability assessment. In doing so, we must remain responsive to the impact of a number of factors related to globalisation, market development and integration as well as innovation on the financial systems.
Ladies and gentleman, I thank you for your attention.
 I am very grateful to Dawid Zochowski and Patrick Sandars for their valuable inputs and contributions.
 Examples include the ERM exchange rate rises of 1992, the bond market turbulence in G-10 countries in 1994, the Mexican (tesebono) crisis of 1994-95, the 1997-1998 Asian crises which affected Thailand Indonesia and the Republic of Korea, the Russian default in 1998 and the market turbulence associated with the near-failure of Long-Term Capital Management, the Argentina and Turkey crises of 1999, the global bursting of the equity price bubble in 2000, and corporate governance problems involving Enron, WorldCom, etc in 2002. See, for example, Schinasi, G. J. (2006), “Safeguarding Financial Stability: Theory and Practice”, International Monetary Fund.
 Indeed, Article 105(5) of the Treaty and Article 3.3 of the Statute assign to the Eurosystem the task of contributing to the “smooth conduct of policies pursued by the competent authorities relating to the prudential supervision of credit institutions and the stability of the financial system”.
 See, for instance, ECB (2007), Financial Stability Review, June
 G. Schinasi (2004), “Defining Financial Stability”, IMF Working Paper, No 04/187.
 see Leena Mörttinen, Paolo Poloni, Patrick Sandars and Jukka Vesala (2005), “ Analysing banking sector conditions - How to use macro-prudential indicators”, Occasional Paper Series No. 26, April; V. Sundararajan, Charles Enoch, Armida San José, Paul Hilbers, Russell Krueger, Marina Moretti, and Graham Slack (2002), “ Financial Soundness Indicators: Analytical Aspects and Country Practices”, Occasional Paper No 212, IMF;