Managing Capital in Financial Institutions
Capital is key to any financial institution. Companies in other industries need capital to buy property and production equipment. For financial institutions, the primary function of capital is to cover unexpected credit and market risks losses, because risk of such losses inevitably accompanies a bank’s core business of lending money and making markets. Thus, it is crucial for financial institutions to build an advanced economic capital framework and how that plays into current initiatives to implement the Basel II Capital Accord.
Capital matters to most corporations, but in different ways. Non-financial companies need capital mainly to buy property and to build or acquire production facilities and equipment to pursue new areas of business. While this is also true for financial institutions, their main focus is different.
Banks evaluate and take risks daily as part of their core business processes. For example, commercial lending involves weighing the credit risk of new loans and their associated mitigates. This means analysing the credit quality of the underlying obligor, the effectiveness of guarantees, collateral, cross-default and other forms of credit protection. But today, best practice does not stop there. It also is necessary to evaluate the impact of portfolio diversification (e.g. in terms of geographical or industry concentration of exposures) and the degree of correlation among exposures on the bank’s balance sheet.
Another example is trading activity whereby a bank benefits from high trading volumes (by earning the bid/ask spread) and hopes to gain from proprietary net positions, but must bear some degree of market risk in the process.
Given the central role of market and credit risk in its core business, a financial institution’s success requires that it be able to identify, assess, monitor and manage these risks in a sound and sophisticated way. The growing emphasis by banking supervisors world-wide on best practice risk methodology shows how important such processes have become.
In order to assess and manage risks, a bank must have effective ways to determine the capital that is necessary to absorb unexpected losses – expected losses being addressed through loan pricing and/or provisions. Also, profits need to be evaluated relative to the capital necessary to cover the associated risks – so-called risk adjusted performance measurement, or RAPM. These two major links to capital – risks as a basis to determine capital and the measurement of profitability against risk-based capital allocations – explain the critical role of capital as a key component in the management of bank portfolios.
So, if you can measure risks, shouldn’t it be straightforward to determine risk-based capital (commonly referred to as economic capital or risk capital)? And why are the two terms – regulatory capital and economic capital – so sharply distinguished?
The challenge of determining economic capital lies in the fact that all the various risks that banks are facing (market risks, credit risks, operational risks) have a very different nature, are measured by specific metrics (if they can be quantified at all) and are difficult to capture by one common metric or method – even value-at-risk approach. Given perfect circumstances, such a common risk metric would not only be able to capture and quantify all relevant risk exposures of the bank, it would also have to be able to account for all the correlations between different risk categories and exposures. In reality, economic capital frameworks within financial institutions are often very fragmented. On one hand they feature sophisticated capital and risk measurement methods for specific parts of the overall risk books. On the other hand, these frameworks often suffer serious deficiencies with respect to calculating economic capital from an integrated risk point of view.
The reason for the gap between regulatory capital and economic capital is twofold: One is that banks are free to choose how to calculate economic capital in a way that they believe accurately reflects the riskiness of a specific position or portfolio. This leads to various levels of sophistication regarding economic capital frameworks, making it difficult to compare the capital management practice of different financial institutions.
In contrast, methods to calculate regulatory capital will be prescribed to a far greater extend by regulators. This is designed to assure greater comparability among banks, provided the rules do not distort risk of one type of activity relative to others (a continuing subject of dispute among banks and regulators.) Also, regulatory capital rules naturally tend to have a time lag in terms of incorporating new risk measurement methods and models because they are driven by a consensus seeking process both among regulators and banks.
Both elements, the freedom to apply the most appropriate economic capital methods (as opposed to having to follow prescribed regulatory approaches) and the time lag in the evolution of regulatory capital versus economic capital methods, result in a gap between the two approaches. The question is not whether this gap is likely to disappear but whether the gap is too big. If so, “capital arbitrage” happens whereby banks try to restructure those risk positions which they believe generate unrealistically high levels of regulatory capital and the risk of such positions is transferred to the capital markets through securitization. This results in regulatory capital being released and used for other businesses or risk positions that require less regulatory capital relative to its associated economic capital levels – the latter reflecting the bank’s best estimate of the true risk of the respective exposures or portfolios).
As figure 1 shows, building an advanced economic capital framework assumes that banks find conceptual solutions to measure risks, define shareholder value metrics and determine capital based on risk measures and correlations. From an implementation point of view, economic capital frameworks mean key challenges in building comprehensive data structures and the supporting technology as well as mastering a significant cultural change. In practice, the cultural challenge and its implications for staff incentives and compensation is often one of the main causes for the failure of capital management projects.
Regulatory capital arbitrage activities and its tendency to undermine formal capital adequacy regulations has been one of the key drivers behind the BIS initiative (led by the Basel Committee on Banking Supervision) popularly known as Basel II. Starting in 1999, this effort has produced three increasingly comprehensive consultative papers and a wide debate on risk management and capital adequacy approaches. One of the main goals has been to align regulatory capital more closely with economic capital, thus reflecting a higher risk-sensitivity of regulatory capital. This should ultimately lead to less capital arbitrage and less perverse incentives by decreasing the differences between regulatory capital and economic capital for specific positions or portfolios.
What is a practical way to build and implement an integrated capital framework which not only meets Basel II capital rules but can also add significant value to a financial institution by allowing it to effectively and efficiently allocate economic capital to its risky portfolios and measure profitability against those allocations?
From a regulatory point of view, implementing a Basel II compliant regulatory capital framework means being able to perform the capital calculations and generate reports in a flexible, transparent, auditable way to meet the requirements arising from Basel’s three pillars:
In order to calculate regulatory capital, compare and benchmark against economic capital and perform credit risk analysis, financial institutions need sophisticated modeling and analysis capabilities. Also, they need to be able to handle stress testing (including multi-dimensional ‘what if’ analysis) and reporting for market disclosure whilst being able to model different rules and rule sets according to different national jurisdictions.
Following Pillar 1 requirements, banks will probably use a flexible calculation engine, working across multiple dimensions. Flexibility means particularly that the system’s front-end facilitates the application of multiple variants of the same basic formulae in a context-sensitive manner to handle different regulations in local jurisdictions. Also, users should be able to modify formulae components (e.g. to model proposed changes in regulation), yet keep a ‘locked-down’ version for external reporting.
A further requirement of a system for the calculation of regulatory capital is the ability to perform stress testing based on Pillar 2 requirements relative to the IRB Approach. The system should support not only the simulation of a variety of credit-risk sensitive conditions over time (e.g. simulating PD or LGD scenarios) but it should also facilitate the integration of credit portfolio models (Credit VaR model) and other internal economic capital models. A strong competence will be the ability to compare and contrast regulatory and economic capital in all relevant dimensions (at the enterprise-wide level and broken down into portfolios, borrower groups, single names and transactions) in a consistent and reconcilable way on a single platform. Pillar 2 requires that the user must be able to demonstrate that credit risk calculations are being used as an integral part of the credit risk process and to manage the portfolio in a responsible, ‘risk-aware’ manner; this includes the use of stress testing (regulatory and ad-hoc what-if analysis). The system should facilitate drill down functionalities to document fully the origin and contents of credit risk calculations.
Pillar 3 requirements mandate the regular publication of detailed disclosures covering all relevant portfolios within the bank, broken down in multiple ways and including qualitative information and quantitative data, (e.g. Pillar 1 and Pillar 2 output). An appropriate system will have to provide a full audit trail for disclosure publication.
In addition to providing regulatory capital calculations, the system should also provide the functionality to support internal economic capital approaches and/or integrate standard Credit VaR models. This functionality includes the generation of Monte Carlo simulations based on a variety of distributions (Poisson, Beta, Gamma, Weibull etc.), the calculation of ‘Extreme Loss’ estimates with any user-defined confidence interval, the ability, through ‘extenders’, to access external models and calculations. Ideally in-house PD/LGD/EAD models can also be integrated with and built on the same technology and platform.
Essentially, we believe that the final form of the Capital Accord is not the core implementation issue but rather the ability to have an institution’s data well organized and centrally accessible to perform and document the necessary calculations. We also believe that addressing the data issue properly will allow banks to leverage their Basel II efforts to improve their fundamental risk management processes and not just pour money into regulatory compliance alone.
It is key for financial institutions to take an integrated view on risk measurement and capital determination given that the various elements – risk-adjusted pricing of products, assessing the risk exposures on different aggregation levels (enterprise-wide level, business unit level, product and/or transaction level) and the measurement of profitability against risk or the allocation of economic capital – are different sides of the same coin and should not be viewed separately.
What this means in practice can be explained by a quote of the former CRO of Canadian Bank CIBC, Dr. Robert Mark: “What was my most important task as CRO of CIBC? The answer is risk transparency: when front managers wanted to do a transaction that made no sense from a risk point of view then my task would be to show them what this meant in terms of economic capital. They could then make the transaction if they wanted to but they had to bear the economic capital on their books and measure profitability against it.”
Solving the integrated risk and capital management puzzle inevitably will be linked to ability of a bank to manage its risk portfolios more effectively than competitors. Thus, developing an advanced risk management practice and an economic capital framework is key to gaining competitive advantage from Basel II compliance efforts.
About the co-authors:
David Rowe is Executive Vice President for Risk Management at SunGard Trading & Risk Systems, based in Palo Alto. Dean Jovic is Group Managing Director for Risk Management/Basel II at SunGard Trading & Risk Systems, based in Zurich. Richard Reeves is Managing Director of BancWare Whitelight at SunGard Trading & Risk Systems in London.