In this article we look at the implications of this battering on capital regulation. We look specifically at the Basel 2 framework in the course of implementation around the world; some have questioned whether the new approach is still relevant and whether the crunch has shown it to be misguided.
Before going there however, it is important to recognise that the credit crunch is not being driven by regulatory capital: when the definitive history is written it will need to start from the primary macro-economic drivers and go from this to the wide range of mechanisms through which the economic stresses were reflected into the banking system and beyond. At the macro economic level several commentators already cite as key drivers financial imbalances between the US and the rest of the world (particularly China), the very low US real interest rates over an extended period, the resulting drive to gearing/leverage, the consequent asset price inflation and tumbling savings rates in the West. As regards the mechanisms through which the crunch hit us, commentators quote, for example: poor management, poor valuation processes, loose bonus schemes and many other causes. Poor regulation and poor capital regulation in particular have also been attributed a role.
In the context of the macro-economics, if one looks at Basel 2 specifically, it would be interesting in our proposed history of the crunch to test the argument (admittedly to a degree a counsel of despair) that no system of regulation would have been able to withstand the wall of debt/money that built up in recent years: the financial incentives to leverage and bubbles across the global economy were too powerful to be controlled by the pages of a rulebook; rules can only do so much; and many institutions that were central to the bubble are in any case not constrained by traditional bank capital requirements. However, with the financial services business in its current situation, it is inevitable that the searchlight falls on capital regulation: if regulation does not stop all this what is it for?
As the Basel 2 framework is coming into place around the world, it is reasonable to ask if it caused the train wreck that is the banking system, was an innocent bystander, or what lessons it needs to learn from the crisis. In our view the train wreck was not a product of poor capital regulation: economic forces were far, far more important. Also, Basel 2 was not in place in the build up to the crisis so in a narrow sense one could say it was a bystander. However, while Basel 2 is a significant step forward on the old regime, it is not the final word and, as with all regulations, needs to adapt. It is a question of learning the lessons.
Basel 2 has three pillars: Pillar 1 is a formulaic approach to capital based on applying risk percentages to assets in a structured way, often using firms’ own risk models if the regulator has approved them; Pillar 2 requires management to assess the risks of the business more holistically and the capital needed to support those risks, and (critically) the regulators to review and approve that approach; Pillar 3 requires extended disclosure. The overall intention is for a better assessment of the risks a bank is taking on, better management of banks’ risks and more transparency about those risks. This philosophy is not yet fully in place, and has not had a chance to prove itself. However, we believe that the three pillars are right: a robust banking system involves more than the application of a few simple capital ratios.
The interaction of the three pillars is critical to assessing the potential impact of Basel 2 and perhaps the first learning point for the regulators is that a review is needed of how the three Pillars taken together could (if fully in place) have impacted the credit bubble and the following crunch: would they have enabled regulators to identify the impending crisis earlier by, for example, requiring better risk management, better management information, better stress testing and better disclosure? Alternatively would this intelligent approach have been compromised by lack of resource/experience at regulators, economic and personal incentives to greater risk-taking as the credit bubble grew, and the difficulty any manager, supervisor or analyst would have in stepping back and challenging the consensus view that saw not a growing bubble but a long benign Goldilocks economy (not too hot and not too cold)?
Looking at Pillar 1 in more detail, many commentators believe that the huge credit crunch losses show the risk based modelling that is integral to Basel 2’s risk based capital framework does not work. They first cite the inadequacies that have been shown in the credit ratings that Basel 2 uses in its standardised approach to credit risk; clearly the role of agencies before and during the credit crunch has been, and will doubtless continue, to be the subject of discussion. More work is needed here: it is difficult to find another simple indicator for credit risk without going back to the Basel 1 one-size-fits-all approach.
Many major banks use their own credit models for capital purposes, not ratings. The critics would reply that banks’ risk models (for any risk) are only as good as the limited data they use; that the integrity of risk models can be corrupted by the incentives on banks and their staff to take on more risk and that the regulators are insufficiently armed to fight back. George Soros addressed the use of Value at Risk models in regulatory capital calculations. He found this a shocking abdication by the regulators that allowed the credit bubble and consequent crunch:
“By relying on the risk calculations of the market participants, the regulators pulled up the anchor and unleashed a period of uncontrolled credit expansion” [George Soros: The New Paradigm for Financial Markets: the Credit Crisis of 2008 and What it Means].
The models he refers to have been used for market risk capital calculations since 1996, well before Basel 2 was a glimmer in the regulator’s eye, but they pointed the way to the wider use of models in measuring credit and operational risk that Basel 2 now permits. Arguably, if using VAR models in the trading book under Basel 1 allowed all this, what would credit modelling under Basel 2 have allowed?
Against this we have to bear in mind that the use of VAR in the calculations as described by Soros is focused around the trading book and hence short term market movements. It was not these that have caused problems for banks: that was the role (it seems) of longer term holdings of often illiquid and poorly secured assets.
Turning the argument again though, the critics could riposte that adoption of VAR in the trading book liberated banks to originate and distribute such assets by mutating originated long term assets into trading book securities, leading to the growth of balance sheets across the financial sector whose true liquidity was often questionable and whose value depended on opaque asset pools.
In addition to the static analysis of the adequacy or otherwise of capital requirements derived from models, the issue has to be looked at dynamically: many have questioned whether Basel 2 is setting up the seeds of a future problem by building procyclicality into the system. Absent some good risk management and Pillar 2 disciplines, the risk figures it generates can easily underestimate capital requirements in a boom, as default rates fall ever lower, and increase them as the defaults later hit, making a credit crunch bite ever deeper.
However one views this argument, we believe that risk based supervision is the right way to go. As banks and products become more complex regulators cannot dismiss under Pillar 1 the tools (such as credit modelling and VAR) that management uses to run the bank. It has to assess them, influence them and where possible, work with them, but above all maintain scepticism and challenge robustly. To do this, Pillar 2 should be top priority.
This is not just a conclusion for the major markets; in less developed markets and offshore centres the balance should be toward an effective Pillar 2. However, it will take resource. This will not be easy, particularly in smaller jurisdictions where expert resource is scarce,: supervisors may need to hold back on allowing the use of Basel’s more advanced approaches under Pillar 1 for this reason. However, all supervisors will need to raise their game to a position when they can really challenge the consensus, and Pillar 2 is the place to do that. Where Pillar 2 is already properly in place we have seen a noticeable up tick in the understanding supervisors have of banks’ business models, resulting risks and supporting capital needs.
Pillar 2 demonstrates the need for managements and supervisors to exercise experience and judgement. This emphasis on judgement should be a foundation stone going forward. In our view the crunch has (once again) demonstrated that mathematical models do not deal well with periods of market stress: stressed markets are far from normal and both maths and experience need to be brought to bear if a bank is to understand where its biggest risks lie. As part of this there has been, and should be, a growing regulatory focus on stress tests. Early (pre-crunch) Pillar 2 stress tests were often inadequate, and banks are increasing having to test their capital bases to destruction. They also have to be far more specific about the actions they would take if the supposed stresses materialised. We are seeing supervisors increasingly asking what it would take to break a bank, as well as asking management what they would do to stop it.
As part of the focus on experience and judgement regulators need to build their resource on economic issues: it is at least arguable that the divorce of banking supervision from central banking in many countries has (despite its many advantages) led regulators to granularise their analysis. They set up teams focusing on particular banks and carefully considering the risks of each bank separately on a bottom up basis. The old central bank breadth of vision across the economy and the risks within the financial sector have faded as regulators have focused on managing regulatory processes that tick off each bank one by one. Sector-wide issues have not received the attention they deserve if supervisors are to reach the informed views on banks that Basel 2 envisages. As an example, in the UK the Bank of England and FSA were criticised by the Parliamentary Select Committee and its Chairman, John McFall for not reflecting their top level economic analysis of risks into their bottom up supervision of banks:
“The Bank and FSA can no longer hedge their bets, throwing potential risks out into the ether and then washing their hands of the consequences. We must ensure that in the future such warnings are heeded and acted upon by those at the top of financial institutions.” [The Run on the Rock – January 2008].
It is easy for regulators, bankers and academics to focus on theory, not practice. In looking to refine Basel 2 we have the advantage (if it really is that) of a wealth of experience of crises and this needs reflection as policy is redefined. One huge area for review (admittedly on the fringes of Basel 2, but so significant it needs a mention) is the complex interaction of capital, liquidity, depositor confidence, deposit protection and lender of last resort.
The business model in banking typically involves funding income earning assets by building debt (from long term bond issues through to overnight wholesale and retail deposits) on a base of core capital. Historically the various elements of the pyramid and the support they get have been dealt with largely separately by regulators. For example, the capital definition and risk measures have been chewed over for years (and this is the Basel 2 link), but non-capital funding has been outside the model and dealt with in most countries by a separate liquidity regime of some kind (Basel 2 covers liquidity risk under Pillar 2, but only to a degree).
Similarly, deposit insurance (where it exists – not often for the offshore sector for example) has been treated as a compensation mechanism for depositors, not a support mechanism for the whole capital/funding pyramid, while the role of the lender of last resort has in many countries been kept deliberately opaque. Finally, the attribution of costs to the various support mechanisms from the authorities has been unclear, with a suspicion (or maybe certainty) of significant subsidy to the banking industry in many cases. In the Crunch, tumbling capital ratios and bank runs (of both retail and wholesale funding) have made clear the link between capital and liabilities.
They have also demonstrated the link between a bank’s capital and liabilities, and the explicit or implicit support the regulators extend. To deal with this, a new “general theory” is needed from the regulators.
As a starting point here we are seeing a growing focus on leverage ratios. These have long been a favourite of the FDIC in the US, but now other regulators (for example the Swiss) are re-examining what seemed to many, until 2007, to be a relic from another age. The essence is to put a restriction on the size of a bank’s tangible assets based on the bank’s core Tier 1 capital (in short, its permanent equity). Arguably many of the banks that required emergency capital injections had adequate capital under the Basel risk measures, but that did not stop the liability side destabilising in days, or hours, as the extent of potential write downs in the emerging recession was realised. Despite its inadequacies (notably not covering off balance sheet instruments well) a debate is needed to assess if the leverage ratio’s time has come again.
To sum up, the credit crunch has made very evident areas for further improvement in Basel 2, but many of these should have been identified far earlier. Basel 2’s ratings and models need to be used wisely, they are not a one-shot solution; Pillar 2 is critical and needs to be implemented effectively; regulators need sufficient resource to deal with complexity and courage to make judgements; and capital and liquidity regulation need to be properly integrated. Properly implemented and resourced Basel 2 remains an opportunity not a threat.
Nick Freeland is senior partner of PricewaterhouseCoopers in Grand Cayman; Patrick Fell heads the PricewaterhouseCoopers UK regulatory capital practice.