The outbreak of the financial crisis in 2007/2008 brought liquidity risk measurement and management to the attention of practitioners, regulators and, to some degree, academicians.
Up until then, liquidity risk was not considered a serious problem and was almost disregarded by risk control systems within banks and by international and national regulations. Liquidity management and fundraising was seen as routine activity, simply a part of more complex banking activity, deserving little attention or effort.
Although the savvy approach would always be to forecast and devise scenarios under which extreme conditions occur, it was barely conceivable that such a difficult economic environment like the financial crisis of 2007 could ever occur. In the economic and financial environment in which banks used to run their business before 2007, liquidity risk simply did not exist. Moreover, it was never considered a problem that couold possibly extend beyond the limits of organizational issues and the development of basic monitoring tools. The design of procedures and systems were believed to cope with the small effects that banks suffered from liquidity risks.
As a consequence of these general considerations, the theory of liquidity risk was vague and restricted typically to market liquidity risk, which is the risk that assets cannot be sold swiftly in the market at a price close to the theoretical value. Although this is an important aspect of the broader liquidity risk notion, nonetheless it is just a small part of the full story, and in most cases not one strongly impacting banking activity. The only literature available on the aspects of liquidity risk concerning banking activity was mainly written by practitioners working in the industry and by a few academics. A notable example, among a few others, is the book edited by Matz and Peter, Liquidity Risk Measurement and Management: A Practitioner's Guide to Global Best Practices [87], which presents an excellent overview of the most relevant issues in liquidity risk management.
We felt there was a gap, though, between the need for improved practices after the events of 2007/2008, and what was proposed in the available research. The above-mentioned book by Matz and Peter was published in 2007 and, of course, could not deal with the increased requirements for risk liquidity practices.
This book tries to cover this gap: it should be seen as an attempt to introduce new tools and methods to liquidity risk measurement and management. We do not dwell on every facet of the subject; in particular, problems that are more related to organization and mechanisms that involve higher levels of management to cope with specific liquidity crises, are only briefly analysed for modifications that could be made to existing best practices in the current financial environment.
The book is organized in three parts. Part I is an overview of the crisis in 2007 and describes how it became globalized from the US economy to the rest of the world and how it altered its form during the subsequent years, up until 2013 (when this book was written). In Part I we also show how the banking business is changing (or will be forced to change) in response to the dramatic events that occurred. These triggered a regulatory overshoot the traits of which are extensively investigated towards the end of Part I (Chapters 4 and 5). One of the most challenging tasks was actually updating chapters in the face of (still) continuously evolving regulation, which represents one of the current main drivers of the liquidity framework. For this reason, this will most likely be the part of the book doomed to becoming outdated the quickest and no longer state of the art. The regulations mentioned and studied in this book are accurate at the time of writing in January 2013. Regulations that have been updated since January can be found at the book's website http://www.wiley.com/go/liquidityrisk
Moving from a macroeconomic point of view, we analyse the different types of liquidity risk and how they impact on a bank's business activity, in order to find how best to manage it from a microeconomic point of view, based on analysis of the actual structure of the balance sheet and of a comprehensive framework for pricing, monitoring and managing liquidity risk.
In Part II we start quantitative study of liquidity risk, first by introducing standard tools to monitor it: it is here we show how these tools can be enhanced and extended to cope with a substantially more volatile market context. The guiding principle is to draw approaches and models from the robust and thorough theory developed to evaluate financial contracts and to apply them, with a slight shift of perspective, to the measurement and management of liquidity risk. For this reason we stress the importance of concepts such as “cash flow at risk” and “liquidity at risk” that are not new, but have never really been widely adopted in the banking industry.
Starting with Part II, the reader will soon realize that topics are discussed as if there were a sort of pendulum, constantly swinging from fundamental concepts, hinging most of the time on balance sheet analysis and involving basic math (algebraic summation), to complex modelling with stochastic processes, grounded on rather heavy mathematical approaches. We would like to make it clear we have not really created new models to measure the liquidity risk, although in a few cases we actually do so. We only want to show how to apply already available theoretical frameworks and extend their use in the liquidity risk field. For example, we show how to adapt the option pricing theory approach to liquidity risk measurement and management.
Hopefully, our intent will be clear when the chapters devoted to the modelling of market risk factors and behavioural models are read. In these chapters we used a number of instruments, available in the theoretical toolkit prepared for the valuation of derivative contracts, to solve specific problems related to liquidity risk. We left aside our initial fear and opted, like pioneers in unexplored territory, to take routes that eventually may prove not to be optimal or even wrong, but our aim was to show a different mindset when approaching the liquidity risk problem rather than to provide the best solutions.
We must acknowledge that others have tried to adopt a similar approach; namely, Robert Fiedler [89] and, more recently, Christian Schmaltz [109]. Continuing in their footsteps, we applied a bottom-up method by modelling the main items of a bank's balance sheet. In theory, the bank is then able to simulate the entire balance sheet on a very granular basis, allowing for a rich set of information that can be extracted for liquidity risk purposes.
The theoretical apparatus developed for derivative contract evaluation is even more fruitful because modern liquidity risk does not only refer, as typically in the past, to the quantitative dimension of cash flow imbalances. In fact, a new and sometimes even more important dimension is the cost of liquidity that financial institutions can raise in the market. The dramatic increase in the levels and volatility of funding spreads paid over the risk-free rate is a factor that definitely cannot be disregarded in the pricing of contracts dealt with clients and the more general planning of banking activity. This is why we devoted the third and final part of the book to the analysis of this topic.
We start Part III with definitions of funding costs and counterparty risk and the interrelations between them, which demonstrate that banks are ultimately forced to consider the cost to raise liquidity in the market as a business-related factor that cannot be hedged. We present a new framework to model funding costs keeping in mind the multiplicity of sources and the dynamicity of the activity. We introduce a novel measure of risk implicit in the rollover of maturing liabilities and we show how corresponding economic capital should be allocated to cope with it and how corresponding costs should be included in the pricing of products a bank offers to its clients.
The inclusion of funding costs is much more subtle when dealing with derivative contracts; this is why we dwell in the final chapters of the book on possible approaches to dealing with them and point out how the classical results of option pricing theory are modified when these additional factors are taken into account in the evaluation process.
In conclusion, we would like to thank everyone who helped us in elaborating the ideas presented in the book. The list would be so long and would involve so many colleagues, who have analysed and discussed these topics with us over recent years, that not only would many pages be required to name them all, but we would also run the serious risk of forgetting someone.
However, Francesco would like to thank his employer (Banca IMI) and bosses for creating a conducive and stimulating environment, in which many topics treated in this book have found continuous reference regarding analysis and applicability; many friends and colleagues, within the counterparty risk management desk and financial engineering desk, the Finance & Investments Department and the Capital Market Department of Banca IMI, the risk management desk and treasury desk of Intesa Sanpaolo, who have always been ready to exchange ideas and interesting opinions and contributions about these topics; last but not least, to all colleagues of the market treasury desk, for their friendship, support and constructive example set during their daily activity. Obviously, the book expresses only the views of the authors and does not represent the opinions or models of Francesco's employers (Banca IMI and Iason).
Finally, we want to acknowledge a number of people to whom we owe a special debt of gratitude. In particular, Raffaele Giura, of Banca IMI, who constantly discussed with us many of the issues covered and always gave us insightful and fruitful perspectives to examine in depth. Antonio discussed many of the ideas related to the liquidity of derivative contracts with Fabio Mercurio. Caterina Covacev and Luca Visinelli, of Iason, contributed massively to refining some of the models presented, to writing the code to test them and to preparing the examples. Finally, Francesco Manenti, of Iason, helped us to estimate and test behavioural models for non-maturing liabilities.
Although the book is a joint work, during its writing we split the tasks so that Francesco mainly dealt with the topics in Part I, whereas Antonio focussed on Parts II and III. We hope the reader will find the text conducive to a better understanding of liquidity risk and perhaps go on to develop and improve the ideas outlined.
Antonio Castagna
Francesco Fede
Milan, January 2013