Drastic times

There is a growing consensus that large, global banking groups pose a threat to financial stability. Some have suggested large financial services groups be split up to focus on traditional markets and banking business - but is this the answer? By Duncan Wood

Thirty-five years ago, the US Department of Justice filed a lawsuit against what was then the only telephone company in the country, AT&T, setting in motion a long legal struggle that concluded eight years later with the monopoly being broken into pieces. It might seem far-fetched to suggest a similar fate could lie in store for the world's biggest banks, but one senior regulator in the UK believes that's exactly what some of his colleagues want to do.

He points to the G-30 report on reform of the financial system, which was published in January and chaired by the former head of the US Federal Reserve, Paul Volcker. "My feeling is that, lurking behind the report, is a wish by Paul Volcker that the authorities had the power to do to the really big banks what was done to AT&T - just split them up," he says.

That would be a radical move. But, as the crisis wears on and taxpayer anger builds, radical responses to the crisis are edging into mainstream discourse - many of them posited on the belief that banks such as Citi and UBS are too complex to be managed and too big for taxpayers to accept the systemic risks associated with their mistakes. Some suggest the parts of banks that perform a public utility should be separated from the parts that take big trading risks; others - including the G-30 report - suggest the further growth of banks should be limited.

Even six months ago, some of these ideas would have been dismissed without a second glance. One such solution has been proposed by Larry Kotlikoff, an economics professor at Boston University, who advocates a massive expansion of the originate-and-distribute model, so that all banking system risk would be held by the public in the form of funds rated by a government agency. In effect, banks would become originators that are stripped of all risk-taking responsibility (see box, A radical solution). Today, this kind of revolutionary talk is at least finding sympathetic listeners, says Kotlikoff.

"We have discovered taxpayers were exposed to huge amounts of risk without knowing it, and we can't let this kind of thing happen again. Now we have one shot at getting this right - it's vital. And what I'm proposing is very simple and it's doable. And, yes, I think there is the appetite and the political will for a radical solution," he says.

Others are expecting simpler packages of reform that address the specific problems highlighted during the crisis. But there's also an acceptance that public anger - stoked on both sides of the Atlantic by rows over bonuses paid by bailed-out institutions - could have a big bearing on what happens next. "My sense of the political situation at the moment is there's little appetite to undertake some of these more radical solutions because it is not obvious they don't come with their own set of risks," says Jim Wiener, a partner at Oliver Wyman in New York. "But there is a huge populist wave that is cresting at the moment. How it breaks I don't know, but maybe that could upset it."

The G-30 report pushes these issues into the spotlight, but stops short of taking up torch and pitchfork to storm the banking industry gates. It notes today's large, complex banks have shifted away from a model based on stable relationships towards one that is more volatile. "What is at issue is the extent to which these approaches can sensibly be combined in a single institution, and particularly in those highly protected banking institutions at the core of the financial system," the report states. It goes on to note that the crisis, with its shotgun weddings of failing and healthy institutions, will further concentrate influence, assets, deposits and risk "if permitted by law and regulation".

There are two somewhat distinct questions: whether some banks are so big and complex they can't be managed; and whether the financial system and wider society is willing to accept the systemic risk these banks represent. Both questions generate strong feelings.

Charles Goodhart, programme director in regulation and financial stability at the London School of Economics' (LSE) financial markets group research centre, says it is difficult not to conclude complex banking business models are part of the problem. "If you look at the banks that have got into trouble - Royal Bank of Scotland, UBS, Citi and Bank of America - most of these are really rather massive banks. If you are that huge and are doing lots of different things in lots of different countries, do you have the oversight, knowledge and expertise at the centre to run the thing properly?" he asks.

UBS itself conceded complexity was a problem when announcing a reorganisation last August. "Our review has clearly revealed the weaknesses associated with the integrated 'one firm' business model. Some of these weaknesses, such as blurring of the true risk/reward profile of individual businesses, are the source of substantial risk, as we have seen in the past few months. Others have led to the creation of excessively elaborate processes and unnecessary layers of complexity," the bank's chairman, Peter Kurer, said at the time.

But not everyone buys that explanation. "It's not remotely the fault of the business model. It was just that people were being stupid. That was much more important: gross incompetence and greed," says an equity analyst at one large bank in the UK. He points out the crisis has been pretty indiscriminate in its choice of targets - many big, complex banks have been left more or less unscathed, while Northern Rock, a relatively simple wholesale-funded mortgage lender, was one of the first financial institutions to be hit.

Bad management

The common factor linking all these victims has been bad management rather than a bad business model, argues Don van Deventer, chairman and chief executive at risk software vendor Kamakura in Honolulu: "I think what is clearly a very dangerous combination is to be large, complex and of below average intelligence - and I think American International Group is demonstrating that they fall into that category" (see box, Failures in risk management).

Still, there is consensus that staying on top of the risks of a large, complex organisation is a difficult thing to achieve and that banks will suffer big losses from time to time. What separates the two camps is the question of whether, despite these difficulties, good management teams can be expected to avoid the catastrophic blunders that necessitate a bail-out. David Clark, a former senior adviser to the Financial Services Authority, who now sits as a non-executive director on the boards of Westpac Europe and Tullett Prebon, and is chairman of the Wholesale Market Brokers' Association, sums it up: "It is very hard to manage these large, universal banks, but the evidence shows it can be done. Look at HSBC as an example - a vast organisation that has had its problems and mistakes, but which has been able to resolve them without needing public support. But it is difficult. It requires strong governance and a lot of investment in risk management."

The second question is arguably the bigger one: if it is so difficult to steer big, complex banks through modern financial markets, does society want to bear the risk of their failure? "If we want money to be cheap for the economy most of the time, do we have the appetite to support these banks when they blow up now and again?" asks the UK bank's equity analyst. "If the answer is no, then you don't allow banks to be levered and they will charge 20 percentage points over base rates for a mortgage, because that is the unleveraged return that you would require."

No society will accept that kind of cost, he argues, so it becomes a question of how much leverage banks will be allowed to take and how they are regulated. In other words - find ways to make the old system work better.

Oliver Wyman's Wiener agrees: "There are ways to reduce the systemic risk of these large institutions without outlawing them." He suggests blowing away the webs of opaque counterparty exposures that connect the large banks to each other, and points to the push towards a central counterparty and clearing system. This is a step Clark also wants to see, but he warns it needs to be handled carefully. "The political and media attention being given to this issue portrays a battle going on between the OTC and exchange-traded markets - this is very unhelpful. OTC markets are crucial for liquidity and it would be very counterproductive to undermine this. The regulatory policy issues coming out of this are huge, and it is essential they get them right."

There are also more unusual ideas, although none can count on universal support. For example, Ingo Walter, the Seymour Milstein professor of finance, corporate governance and ethics at New York University's Stern School of Business, wants to see a dedicated regulator created for large, complex financial institutions. But Klaas Knot, director of supervisory policy at De Nederlandsche Bank (DNB) and head of the Basel Committee's risk monitoring and management group, would prefer to avoid splintering regulators. "If you have too much dispersion of supervision, you have all kinds of co-ordination and responsibility problems," he argues.

In the UK, there have been murmurings that it might make sense to break up big banks so the stable units can conduct traditional banking business supported by large retail deposit bases, while the riskier trading arms would be free to use volatile wholesale funds. The equity analyst scoffs at the idea: "There is interest in this from the UK Treasury Select Committee, but it's basically the same model the US had - you take the nasty wholesale things and ring-fence them away from the good, wholesome banking businesses. How do you think that has worked? Has that been a) good or b) a complete train wreck? Copying that would be insanity."

Ultimately, any suggestion big banks should be broken up also has to make the case that smaller, simpler banks are going to expose the system to less risk - that's a problem. "You mean simple banks like Northern Rock?" asks DNB's Knot, rhetorically. The collapse of a single smallish bank would not result in a system-wide problem, but history is littered with episodes where supposedly easy-to-manage banks have run into trouble en masse, requiring taxpayer-funded bail-outs: the problems of Germany's Landesbanks and Japan's regional banks; or, perhaps most notably, the US savings and loan crisis. "These all help illustrate the central problem is one of management. Being complicated and of below average intelligence is fatal, but being small and simple and of below average intelligence is also fatal," says van Deventer.

However, the debate is almost certainly not going away, and nor should it, says the LSE's Goodhart: "At the very minimum, there has got to be an argument for trying to ensure financial intermediaries - not just banks, but all financial intermediaries - do not get to the kind of size where their failures threaten the entire economy."

A RADICAL SOLUTION

As responses to the crisis go, they don't get much more radical than the one being championed by Boston University economics professor Larry Kotlikoff. He argues because society at large bears the risk in the event of a banking catastrophe, this risk-taking should be formalised and made as transparent as possible by having a government agency step in to play the quality-assurance role traditionally played by rating agencies.

"We've had a complete meltdown in the financial markets and no-one is going to trust the banks again - with good reason. We need disclosure and we need independent ratings, and the government is the only entity that can provide that," he says.

Kotlikoff's solution - which he calls limited-purpose banking - would allow banks to continue making consumer and corporate loans, but all the assets would then be transferred to a new government agency, prospectively dubbed the Federal Financial Authority, where they would be pooled and packaged into funds with varying levels of risk. Banks would be left with no credit exposure whatsoever. It would all be held by the public - which Kotlikoff argues is just an acknowledgement of the fact that society already acts as the ultimate backstop for the financial system. Of course, as things stand, wider society only has that exposure in the event of a financial catastrophe, while bank earnings and capital mop up anything short of that. Would the general public be willing to take on all this risk?

"People would need to understand that life is risky," he argues. "But in this system, at least they would know what risk they were taking on and could have confidence about the amount of risk involved."

Limited-purpose banking isn't being talked about in the corridors of power at this point, but Kotlikoff says other economists with whom he has discussed the idea think it has merit. More importantly, he insists the time is right for a complete redrawing of the map. "The banking system has already imploded. What we need to do is build a new system that actually works, that everyone will see and understand, so this never repeats itself," he says.

 

FAILURES IN RISK MANAGEMENT

To a casual observer, the huge losses many large, globally active banks have suffered during the crisis might have come as a surprise. For Don van Deventer, president and chief executive at software vendor Kamakura in Honolulu, the only surprise was that it took so long to happen. "They have been behaving this way for over 20 years. They didn't suddenly go from being bright to being below average," he says.

The first time alarm bells started ringing was in 1984, when van Deventer was appointed treasurer at First Interstate, then the seventh biggest bank in the US. He arranged a meeting with his counterpart at a large, globally active bank to get some advice. "So I went to see him and I asked how often he reviewed the bank's consolidated risk position and he told me he didn't have those numbers. I said, 'Does that not concern you?'"

Apparently, it didn't. Van Deventer's counterpart argued the bank was so huge and so diverse there was no way for the sum of its risks to threaten the organisation's stability. "I walked out of that meeting shaking my head and thinking the biggest bank in the US was flying completely blind," he says.

Later, in 1997, van Deventer - now at Kamakura in Japan - was sitting in his office when he received a phone call from an Indian software developer who was working on an upgrade of the bank's asset/liability and interest rate risk system and looking for advice. Van Deventer asked for an outline of what was in place at that point, and says he could barely believe the reply. "It was absolutely the most primitive description of a risk management system that I had heard in 20 years. They had the same infrastructure I had when I started in banking at Security Pacific in 1977, and nothing better than that. It just confirmed for me (the bank) was on a course that was basically dedicated to the lowest common denominator of risk management."

More recently, a former colleague of van Deventer landed a job working on the trading floor of one of the bank's subsidiaries. The two had a chat about what the bank's risk infrastructure looked like from the inside. "He told me he'd never seen anything like it - the entire organisation was being run on spreadsheets," he says.

Van Deventer adds the bank has made an effort in recent years to upgrade its risk management systems - but argues the stories he heard from inside the organisation were evidence the bank was simply not committed to managing risk: "For 25 years, they had been really well below average in risk management technology, but it wasn't very well known to a lot of people. It should have been very well known to the US bank regulators," he says.