Deutsche Bank, the blindfolded regulator, and the 26 billion euro vanishing trick

08.05.2013

The biggest banks in Europe have little to fear from the EU's new banking regulations. But the public does. Due to loopholes the new rules will not prevent costly bailouts in the future. The risks are made clear in a recent accounting sleight of hand by Deutsche Bank.

New EU banking rules were adopted in early April after more than four years of studies, debates, and a few tugs of war.  Banks will be obliged to possess bigger reserves – known as capital requirements – than in the past to make them more resilient to market downturns. These reserves act as a buffer of cash and safe assets that banks can always access in order to protect their creditors and secure the bank against insolvency; the ratio of capital to assets is a key measure of financial health. The promise is that the new rules raising capital requirement levels will help prevent or mitigate a fresh collapse in the financial system and the possible ensuing expensive bailouts of ailing banks. A brief glance would appear to show that the capital requirement increase is substantial; in some cases they appear to be doubled1.

So far, so good. However, there is a giant loophole that allows big banks such as Deutsche Bank to do the calculations of the minimum reserves required themselves. Allowing this kind of self-regulation threatens to undermine the new banking regulations. For example, recently, Deutsche Bank made an adjustment to its own internal calculation model for capital requirements, and this alone improved its books by 26 billion euros: quite a magic trick of creative accounting that puts the bank's slogan: “A Passion to Perform” in a new light.

This loophole was heavily criticised in the wake of the financial crisis, but despite a lengthy debate on banking reform, it remains in place – a sign that the banking lobby is still extremely powerful and effective at pursuing its own interests. As a consequence, the new rules do not seem to offer any guarantee that exorbitant bailouts of bad banks – which have so far cost taxpayers in the European Union 1,7 trillion euro since 2008 – can be avoided in the future. That is, not unless new pressure for more fundamental reforms emerge.

Capital requirements and loopholes

The new European Union rules are derived from the Basel Accords, international agreements on banking regulation negotiated by representatives of the biggest financial powers within the framework of the “central bank of central banks”, the Bank of International Settlements. The key feature of the Basel rules are capital requirements – that banks have to retain a certain amount of capital in reserve compared to their assets. The riskier and the bigger the assets (i.e. anything that has value, such as shares, bonds, financial instruments of all kinds etc.), the bigger the capital requirements.

While that sounds relatively straightforward, calculations to set the minimum rate of capital required for a bank to be considered financially secure have become very complex. The Basel rules from 2004 – called Basel II – introduced the option for banks to use their own advanced internal risk assessment models to make these calculations, known as the Advanced Internal Ratings Based Approach. This new approach was pushed heavily by two big global players in the financial lobby; the Institute of International Finance (IIF) that represents the biggest banks, and the International Swaps and Derivatives Association (ISDA), the largest financial trader association2. The financial lobby was keen to protect its own self interest in being able to take bigger risks with less capital. As banks are not interested in having their capital 'laying idle', and as they make their money from taking risks, they consider demands for big reserves as a nuisance. Therefore any loophole to scale down the reserves required is welcome.

The decision practically allowed banks to self-assess via their own complex models how risky their investments are, and thus determine by themselves the size of their capital requirements. And all big banks of some size seized the opportunity immediately, while smaller banks with fewer resources had to stick to a standardised approach used pre-2004, based on fixed, transparent, and not least externally produced assesments.

This has given larger banks an unfair advantage over small ones. At a medium sized bank, for instance the Danish bank Nykredit, out of 4.000 employees in total, 70 staff members are dedicated to risk modeling alone. This is a measure of just how economically valuable this type of internal modeling is to big banks. Smaller banks, on the other hand, do not have the resources of such a sizable staff, and will have to use a standardised approach used before Basel II. As a consequence, they will feel “the full weight of higher capital requirements”3. Crucially, the situation also gives big banks more space to take more risky investments backed up by less capital in the vaults, potentially paving the way for bankruptcies and financial crisis.

The financial crisis reopens debate

The main problem is the vast difference between what the banks report as their reserves (as a percentage of their assets) and what they actually have available according to a standard calculation. Before the financial crisis in 2008, many banks bragged in public that they were backed up by 10 percent capital, while in reality they were down to 1-3 percent, according to a study from the German Max Planck Institute4.

The original objective of the Basel II rules was actually to raise capital requirements. However, thanks to the introduction of the internal risk models, this was derailed. Despite higher requirements, as a consequence of the leeway provided by Basel II, capitalisation of big banks dropped considerably5. This drop in capital ratio due to this "self-regulation" approach to banking was cited as one reason why the sector showed little resilience during the financial crisis in 2007 and 2008. “It is stating the obvious to say that the faith placed in models has been dented by the events of the recent crisis,” said Wayne Byrnes, Secretary General of the the negotiating body of the Basel Accords, the Basel Committee, recently6.

The financial crisis naturally led immediately to a review of international banking regulation. Negotiations started in 2009 and were concluded in the summer of 2010 with the agreement called Basel III. This process was again characterized by fierce lobbying by the biggest banks, spearheaded by Joseph Ackermann, CEO of Deutsche Bank, acting on behalf of the Institute of International Finance (IIF). The IIF succeeded in convincing the negotiators not to raise capital requirements too much, arguing this would be detrimental to growth7.

And as a bonus, in the aftermath of the crisis the banks managed to maintain the loophole created by internal models. Though some procedures were tightened, in fact, the internal risk models were never in the danger zone during the negotiations over the next set of global guidelines, Basel III.

Fixing numbers, hiding losses

As for the capital requirements that were adopted with Basel III, the banks have been in an excellent position to defend themselves from any real impact from higher capital requirements in the aftermath of the crisis with this loophole of self-calculation models assured. In fact, it seems banks have been preparing for years to deal with a potential increase. In anticipation of higher capital requirements, they seem to have been gaming the system by producing low estimates of how risky their assets are, with the aim of bringing down the capital they would in the end be required to have available. This can be seen from their assessments of risk: as the required capital is set according to numbers measuring the 'riskiness' of banks' assets; between 2007 and June 2012, the banks' own assessments of the risk was very optimistic. In a number, their 'riskiness' (Risk Weighted Assets, RWA) fell from 65 percent to 35 percent8.

Judging by recent discoveries, some banks went very far in their efforts to stave off an increase in capital requirements. The scandals are now popping up one by one, and the sums contested are not in the petty cash category. In January 2012, Wall Street Bank JP Morgan Chase changed its risk assessment model and in the act hid losses on derivatives deals  that turned out to amount to no less than 4,8 billion euro (6,2 billion dollars)9. This incident lead to an investigation for fraud. Later, in March this year, the Bank of England accused UK-based banks of overstating their capital by 59 billion euro and hiding losses of 36 billion10.

Over the past few years, it seems, the fact that higher capital requirements seemed inevitable with the onset of Basel III, has inspired the banking community to become very creative in response. In their advanced computer models they’ve juggled with numbers so as to ensure the bottom line would not produce uncomfortably high capital requirements.

Deutsche Bank secretly on the brink

Deutsche Bank is no exception. Neither at present,  nor when the financial crisis hit.

At the height of the financial crisis in 2008, Deutsche Bank was not regarded as a problematic bank, and its top management did everything in its power to retain an image as a rock solid giant. The books showed a fairly good position in that the bank seemed to meet the capital requirement criteria and the CEO of the bank, Joseph Ackermann, famously announced: “I would be ashamed if we were to accept government money in this crisis.” “At no point”, he claimed, had Deutsche been in danger11.

But in June 2011 the first testimonies emerged that indicated Deutsche Bank had not been in as good a state as it had claimed. Three whistleblowers from inside the bank itself have come forward and delivered evidence that Deutsche Bank hid a loss of about 13 billion euro in a complex derivatives deal with a nominal value of 130 billion euro. So while Ackermann smiled confidently and convincingly, during the peak of the financial crisis in 2008 when most banks were suffering distress, behind the scenes his top risk managers were struggling to cover up losses.

The reason why this episode is particularly embarrassing to Ackermann and Deutsche Bank is because the German Government was operating with a capital threshold of 8 percent, typical at the time. Banks below this level of capitalisation would be invited to join the government's support scheme. Officially, Deutsche Bank stood at 10 percent, but if the information on the losses had been properly reported, the bank's level would have been below 8 percent – bringing it within the threshold for government support and severely undermining its reputation as a solid bank. This would have exerted a severe blow, had Deutsche been compared to banks that were bailed out: its level was close to Commerzbank that did receive government support after sustaining a loss of 285 million euro – a number considerably less, in fact, than the losses covered up by Deutsche Bank12.

Deutsche Bank and the 26 billion euro magic trick

Deutsche Bank has also delivered an astonishing recent example of how much an adjustment to a banks' risk assessment model can mean in terms of its capital ratio. In 2012 Deutsche Bank had a major problem with the capital requirements in place. Its accounts showed it was about to miss the mark, so the bank did three things to avoid the problem: it sold off risky assets, it bought derivatives to insure against losses on other assets, and last but not least it dramatically adjusted its internal models on risk assessments. In total, the bank managed to adjust its balance sheets to the tune of 55 billion euro. Of this, a full 26 billion euro was the result of fiddling with its risk assessment models alone13.

What does this mean? It means the bank reviewed its own computer models that had been set up to assess the value – or rather the riskiness – of the banks’ assets. This procedure allowed it to present a much more optimistic view of its investments than seemed warranted. Since there is ample space to define the models internally, the move has not been disputed by regulators so far.

The advantage to the bank was to bring its books closer in line with the capital requirements in force, putting it on similar footing to other big banks albeit “at the lower end”14. But in terms of society’s interest in resilient banks, the change to the model did nothing but obscure the bank's real financial health.

The affair has been closely followed by the business press, and by financial institutions. One of them, Portuguese investment bank Espirito Santo, has a research department according to which the use of a standardised method, such as the one used by all before Basel II, would lead to a 52 percent increase in Deutsche Banks “risk weighted assets”, which means the bank is very far off the mark in its stated estimation of the capital it needs to have available in order to comply with the standards15.

The 26 billion dollar conjuring trick, however, worked magic for Deutsche Bank as its shares went up.

Belated concern

Once again, there seems to be evidence enough to show that banks are using well-designed models to drive down capital requirements to a level where the rules on their reserves become almost meaningless. On occasion this is even recognized by the banks themselves, as when analyst Mike Harrison at Barclays stated in February: “The Basel rules stand or fall by the RWA (risk assessment) calculations. If there are questions on how banks calculate their RWAs, the right amount of capital is almost a moot point if you cannot trust the denominator.”16

Scandals such as those outlined above have caused a stir in the regulatory community, since they are far more than isolated anecdotes: they involve a significant number of major players in the world of banking. As a result, the Bank of International Settlements is currently looking into these internal models17. Both internationally as well as in the European Union, the question being raised is whether a standardised approach to risk assessment with much less space for creativity would be a way forward.

In Europe too, the problem is systemic. At the moment, the European Banking Authority (EBA) is conducting investigations into the risk models of the banks, and in an initial report, the EBA asks what the reason is for the big variation in banks’ assessments of their assets. The Authority come to the conclusion that half of the variation owes not to different regulatory approaches or different structures in the balance sheets, but to variations in the banks' models. “The dispersion calls for further investigations and possibly policy solutions”, the EBA’s chairperson Andrea Enria said in February18.

Models not challenged in the EU

But looking at the legislation adopted, a policy solution seems far away. The new European Union laws are an adaptation of its own version of the Basel rules – a weaker version by several accounts19. As far as the internal risk modeling is concerned, it was never seriously contested and hardly even formed part of the debate. While steps are being taken to include internal modeling in EU law – whereas before this was left to member states – this is not necessarily with a view to adopting strong rules on the matter.

Continuing to allow banks to use their own internal models in the new EU legislation was actually a key demand of Deutsche Bank and presumably of other big banks. In Deutsche Bank's communication with the Commission, consolidating internal risk assessments in the European Union was on its wish list. They preferred to see a more common approach across banks to internal risk modeling. In particular they wanted to see quicker approval by regulators and a set of rules that would allow ample access to the use of internal models20. To get this far, a separate chapter on international ratings would be required – one that would leave space for common standards. This wish was fulfilled by the Commission, introduced in the draft regulation, and subsequently endorsed by the Council and the European Parliament with some loose ends to be decided at a future stage.

Generally, though, the models barely even featured in the political debate. To the extent that they were,  the position of the big banks was defended, not challenged. Among the hundreds of proposals discussed in the European Parliament, for instance, only one was directly relevant to internal models. This proposed amendment, tabled by the German MEP Werner Langen, asked the Commission to respect the “proportionality principle”, should the Commission decide to adopt “technical standards” for the models21. This amendment, finally included in the adopted legislation, sent a message to the Commission not to be too anxious to regulate the internal models, very much in line with Deutsche Bank's position22.

At the moment, the Commission is awaiting the results from the EBA’s investigations before it considers whether the “remaining degrees of freedom are still too large”23 and whether it should take new initiatives. But most likely this will merely boil down to technical standards to be proposed by the EBA end of this year24. Furthermore, what is worrying about the procedure is that it is basically a matter left to the Commission – in a close dialogue with 'The Banking Stakeholders Group' – an advisory group established under the umbrella of the European Banking Authority, dominated by the financial sector, including banks such as BNP Paribas, Rabobank, Santander and Deutsche Bank25.

Banking regulation in crisis

With this in mind, and the Commission’s position so far, it would be a big surprise if the big banks will see their internal models come under any significant regulatory control due to proposals adopted via this process. And the lack of political attention given to this matter – partly due to the fact that the banks easily dominate this kind of complex topic, partly due to strong banking lobbies – means there are few indications of any significant developments.

Indeed, looking at the scale of the disaster created by this type of self-regulation, it would be a more healthy response to discard the models altogether. Specialists at the OECD would agree. In a contribution to a hearing in the German Bundestag recently, two such specialists, Adrian Blundell-Wignal and Paul Atkinson, testified that: “The core problem is the Basel risk weighting system, designed to introduce an illusory 'risk sensitivity' that relates minimum capital requirements to 'risk-weighted assets (RWA)', instead of actual balance sheets. This has evolved into a system of extreme complexity that invites regulatory arbitrage to reduce RWA relative to Total Assets (TA), defeating the entire purpose of capital adequacy rules.”26

In other words, the attempt to make banks more resilient by increasing capital requirements – which is fundamentally what Basel III and the new EU laws are about – will be in vain, if the banks are allowed to persist in using their own internal models.

The argument goes even further than that. The questions posed by quite a few specialists at the OECD27, the Bank of England28, and by regulators in the US29 is actually more fundamental: it’s whether capital requirements are in the end the best tools to avert the collapse of banks in the first place. They stress that the present rules have become overly complicated and make it difficult if not impossible for regulators and supervisors to keep track of the banks. Basically they’re advocating a type of banking regulation with little space for the kind of self-regulation that the current risk assessment models embody.

A new approach to banks

In summary, the situation gives a strong impression of banking regulation in crisis. At the very moment when the European Union has adopted its own diluted version of the Basel rules, the key basis of Basel itself is being questioned. One of the reasons Basel ended up in this mess, was successful lobbying against standardised regulation by the banks. Both at the international level as well as in the European Union, big banks have had significant influence over the policies supposedly being developed to regulate them. This includes their self interested demands for the space and flexibility offered by the use of internal risk modeling, which in the end could turn out to be the most significant outcome of the Basel negotiations and the EU implementation.

At the moment, Deutsche Bank is under scrutiny for a number of business strategies, including tax fraud, the cover up of losses of the Italian bank Monte Passchi, as well as for the creative way it hid its own losses in 2008. These cases might even end up with convictions. In the case of the 26 billion euro trick, however, it seems likely that the rules were followed and everything was done by the book.

But doing everything by the book is no help to society if the rules in the book are made by and for the banks. Creative accounting is incredibly dangerous as it has the potential to paint a confident façade over a bank that might be in big trouble, and that could end up costing millions of people dear. And most importantly, considering they were a key factor in creating the financial crash of 2008, lax banking regulations need to be reformed so that we are not doomed to repeat this crisis.

The question is what kind of reforms would be able to safeguard the public from expensive bailouts and the austerity measures that tend to accompany them? A way forward could be to end the era of self-regulation, create a diverse banking sector with no banks that are too big to fail, and with public banks as an increasingly important component. Big private banks have had their chance.

This article continues after the banner

Subscribe to our newsletter