Where Bank Herstatt (1974) and Continental Illinois (1984) have led, Northern Rock has followed. Like those before it, it got into trouble because it lacked liquidity. Its fate pushes into the spotlight the risk that nobody dare speak about, and that regulators have paid little heed to.
Since 1988, regulators have defined the international standards for banks in the areas of credit, market and operational risk. But they have ignored liquidity risk. They do so at their (and our) peril.
The text book – and Northern Rock’s Annual Report & Accounts’ – definition of liquidity risk is that of “insufficient funding to ensure that the bank is able to meet its obligations as they fall due”. While it is rarely mentioned, it is the most common cause of bank failure. If liquidity risk is why most banks fail, then it would be logical to assume that it attracts a high level of disclosure. This is not the case.
A Liquidatum study of risk commentaries published within bank annual reports found that the greatest attention is paid to credit and market risk – both risk dimensions with agreed international standards in place – followed by operational risk, the latest risk to receive Basel attention; liquidity risk limps behind in last place.
Liquidity risk is a ‘binary’ risk: you either have enough, or you do not. Furthermore, it is all about confidence. When this is lost, banks will fail. Banks do not like talking about liquidity risk because it kills confidence and, historically, they have not been rewarded for honesty on this particular topic.
The scant regulatory attention that liquidity has attracted has been qualitative. Both the Bank of International Settlements and the Institute of International Finance have published guidelines for managing liquidity risk. Northern Rock would, arguably, have passed both tests. The focus will now be quantitative.
Is there an agreement as to a common quantitative approach? Some are arguing for a capital-based approach, others for concepts such as LaR – liquidity at risk. Some throw their hands up at the data challenge – describing the behaviour of so-called non-maturing assets and liabilities (NomAL) as beyond the ability of regulators to quantify. Whatever the difficulties, deriving common quantitative approaches is worth pursuing.
There are problems; one of them is data availability. Market and credit risk default data can be collected daily. Liquidity risk events occur only very rarely and do not succumb to stochastic analysis due to their binary nature. But a bank can take sensible first steps – such as measuring its ability to withstand a number of different liquidity scenarios.
Matt Ridley, chairman of Northern Rock, is quoted as saying: “We were hit by an unexpected and unpredictable concatenation of events.”
This is not exactly accurate and is an explanation often used by failed institutions.
Thinking the unthinkable
What is true is that Northern Rock suffered a liquidity squeeze of a magnitude and duration that was not thought possible. Even without the benefit of hindsight it was clearly possible – it was just unlikely. This is the trap that a stochastic approach creates. Banks need to focus on the unthinkable events that would expose them most: then they must demonstrate an ability to continue meeting their obligations under those conditions.
Many liquidity risk managers have standard scenarios under which they model the liquidity resources of the bank. Most of these focus on bank-specific events, such as a double-notch downgrade by the ratings agencies. Alternatively, risk managers looking for unlikely scenarios for ‘catastrophe’-type models frequently look to the past: the Asian Crisis, the fall of Long-Term Capital Management (LTCM) and 9/11. This is insufficient for liquidity risk scenario analysis.
Instead banks need to analyse their balance sheet structure to identify where they are most at risk and then envisage scenarios that could cause them to be tested. These might include:
- A collapse of secured funding due to settlement failure or the repo markets being closed for 10 days.
- A run on retail deposits for non-name-specific reasons (such as weather related events: a flooding of the river Main in Frankfurt, a storm surge overcoming the dykes in the Netherlands).
- An inability to use eligible assets (such as a strike in Paris – resulting in Banque de France staff being barricaded out of their building).
- A collapse in the demand for certificates of deposit and commercial paper (such as a withdrawal of securities lending reinvestment funds from the CP markets for fiscal issues).
Most importantly, warning lights need to be created to highlight the impending realisation of such unimaginable events. This combats inertia and encourages proactive behaviour to protect the balance sheet from the “unexpected and unpredictable”. The table above shows just such an analysis for five banks. We have used Liquidatum’s proprietary liquidity strength measure (the ELIA ratio – Enduring Liabilities: Illiquid Assets ratio) which measures the ‘stickiness’ of an institution’s assets and liabilities – both on and off balance sheet. A ratio in excess of 1.0 indicates that the bank will satisfactorily fund itself despite the circumstances of the particular scenario.
We have calculated the ELIA ratio of each bank under three different scenarios: A 14-day closure/cessation of the global repo markets, a one-month short global liquidity squeeze and, finally, a scenario mimicking the circumstances of the markets in 2007.
All five banks have very different ELIA behaviour under these conditions. Four of the five have ELIA ratios in excess of 1.0 under all conditions. Most banks positions vary considerably in the different scenarios. One bank, Bank C, is very liquid under the one-month liquidity squeeze scenario but has insufficient liquidity under the Sub-prime Crisis scenario. Another bank, Bank E, is equally solid in all three scenarios.
Role of central banks
An important issue that has arisen as a result of the crisis in 2007 is the role of ‘lender of last resort’. This role was created in the 1800s to protect the system in the event that all other sources of funding to a bank failed. Then, borrowing money from the lender of last resort was understandably associated with severe financial difficulties.
The role has evolved but, problematically, the perception of the general public has not.
Today, liquidity trades by the minute rather than by the day. Banks are required to store liquidity in eligible assets (collateral) that can be pledged in order to release payments during the day. If anticipated payments are not received, the collateral is not returned and the bank is required to borrow money at a penal rate. The problem arises in the subsequent press sensationalising of ‘normal’ bank borrowing from the central bank.
Misunderstandings are compounded by the lack of a level playing field across the major financial centres.
Regulators will spend the next few years standardising the approach to liquidity risk oversight. At the same time, central banks must establish their own code of common practice. This will make borrowings from the central bank a financial statistic, not front page headlines with attendant falls in share prices.
The trend towards globalisation in financial services has run strongly. Significant extension of financing to the property sector in one part of the world is funded by savings in another. These interdependencies, assisted by new techniques to transfer risk, are on the increase and the effects are becoming more significant, exacerbated by the lack of a pan-regulatory approach to liquidity.
In addition, institutions other than banks have entered the lending businesses. This includes not only the now ‘traditional’ non-bank financial institutions but also structured investment vehicles (SIV), insurance companies and some hedge funds and private equity houses. Many of these engage in the time honoured process of ‘lending long and borrowing short’ – taking liquidity risk. They are, however, not governed by the same liquidity regimes as banks, nor do they have access to central banks as a lender of last resort.
This broader range of players, many of whom have a substantial impact on the global markets for liquidity, must also understand the fundamentals of liquidity risk, and adhere to best practice. If not, pricing and behaviour will be driven by the lowest common denominator and then all the central banks and regulators in the world, backed by global tax receipts, will not be able to ensure the integrity of the global financial system.
Liquidity risk premium
Basel I and II have taught that it is only when you ration a resource that it begins to be priced properly. For the moment, very few corporate bankers are charged the liquidity risk premium that the market demands.
In the marketplace, long-term money is, ceteris paribus, scarcer than short-term money. This is not normally reflected in the internal allocation of liquidity in banks. This is despite the fact that most major banks now have sophisticated pricing and transfer pricing mechanisms that permit departments to trade in balance sheet and capital resources. It is only when ‘Basel III’ is fully implemented do we expect that this will also be true of liquidity resources.
The risk that dare not speak its name? It is not until liquidity risk is transparent and accurately priced will we be able to talk about it freely.