In June of this year, Richard Bove, a banking analyst at an American brokerage, Rochdale Securities, wrote an upbeat research report on Citigroup, the distressed giant banking conglomerate. Though he works for a small company, the white-bearded Bove has a big following. His opinionated stance and regular cable television appearances have made him one of Wall Street’s best-known commentators on the financial crisis.
Bove argued that Citigroup’s shares, then trading at just $3 each, could more than quadruple in coming years. This was, to say the least, bold. The bank has been one of the biggest casualties of the crisis, surviving only through massive state support. Earlier this year, its shares—worth $55 each in early 2007—fell below $1 on fears that it would be nationalised. It wasn’t, although the government has guaranteed a chunk of its trashiest assets and now holds a 34 per cent stake.
Bove’s tip was not based on the quality of the underlying business. “Citigroup is an unusually controversial company because its loan losses are at a level that suggests that the company would have difficulty surviving without government assistance,” he wrote. Indeed, Bove admitted that the bank’s loan book was possibly “one of the poorest ever written.” But there was a magic ingredient that would save the bank from the knacker’s yard: Uncle Sam. Since the government had deemed Citigroup “too big to fail,” he observed, it would be able to survive in private hands and possibly rebound down the line. “Some companies have developed such unique positions that their elimination would cause harm to others,” Bove said. “I believe Citigroup is such an organisation.” On this basis, he recommended its shares to investors.
Citigroup is a perfect illustration of the “too big to fail” problem. The global financial crisis, and subsequent action taken by governments, has shown that investors in banks of systemic importance—that is, those whose failure could bring down the financial system—enjoy a one-way bet. They get to keep the profits in the good times and are able to pass the losses to taxpayers in the bad.
Look more closely at Citigroup’s own case. Since 2007, it has written off more than $105bn—far more than the equity capital it had going into the crisis. Most of the bank’s capital came from depositors and other debt investors. None of these lenders has lost anything. Even equity investors, who should have been first in line to take a hit, have not been wiped out.
This safety net is not new. For decades, central banks have intervened to prop up institutions whose collapse might lead to a wider loss of confidence. What makes this crisis different is that it has required the authorities to intervene on a hitherto unimaginable scale. As well as providing insurance to depositers and liquidity to troubled banks on increasingly generous terms, they have also recapitalised failing institutions while in effect guaranteeing that anyone who has lent money to a bank will get it all back.
The result has been enormously costly. According to the Bank of England, governments and central banks in the US, Britain, and Europe have spent or committed more than $14 trillion—the equivalent to roughly 25 per cent of the world’s economic output—to prop up financial institutions. Combined with a global recession, this bailout has undermined the public finances of the developed world.
Why has this burden been placed on the productive economy? One main reason is the evolution, over the past 20 years or so, of global financial institutions of such size that the consequences of their demise could be catastrophic. When Lehman Brothers filed for bankruptcy last year, its failure caused the markets to seize up so badly that many of the world’s banks were threatened with insolvency. To counter this, the authorities had to make it clear that no large financial institution would be permitted to suffer the same fate. Since then, a string of financial institutions, including Citigroup and AIG in the US and Royal Bank of Scotland and Lloyds Banking Group in Britain, have received injections of capital from the state. Countless others have been propped up by central bank and government guarantees.
This state of affairs is not only unjust in that it loads private losses onto the taxpayer; it also sows the seeds of future crises. When analysts explicitly tip stocks on the basis of a government guarantee, as Bove did with Citigroup, “moral hazard” has risen sharply. The danger is that such rescues will merely encourage banks to take more risks, forcing the authorities to intervene on an even greater scale next time. Piergiorgio Alessandri and Andrew Haldane of the Bank of England have called this scenario a “doom loop.” Escaping it is one of the most pressing tasks facing policymakers today. Yet though there is broad agreement on the nature of the problem, there is little consensus on what to do about it.
Part of the problem is that the governments that host the world’s two largest financial centres—the US and Britain—are terrified that if they regulate the banks too tightly, they may kill off their competitive advantage. So they have been cautious about imposing new rules without first securing an international consensus—a slow and painstaking process. The apparent lack of action has triggered calls from across the political spectrum for more radical change. George Osborne, the shadow chancellor, and Bernard Sanders, the left-leaning US senator, have advocated tackling big banks head-on.
RISE OF THE SUPER-BANK
The government safety net for the banking system has not always existed. Until a few hundred years ago, banks had more to fear from defaulting sovereigns than vice versa. More recently, government has acted to protect the public from the effects of crises. In particular, to guard against damaging runs, governments have made deposit insurance more generous.
The reason that the safety net has expanded is that there are now more bigger banks. Between 1998 and 2008, the share of global banking assets accounted for by the world’s five largest banks—as measured by the Banker magazine—doubled from 8 per cent to 16 per cent. Since Citigroup was created through the 1998 merger of Citicorp and Travelers, its balance sheet has increased from $740bn to $1.9 trillion. And it is far from alone. Goldman Sachs had $217bn in assets in 1998, the year before it abandoned its status as a private partnership and floated on the stock market. It now has around $1 trillion. As recently as 1999, Royal Bank of Scotland had assets of just £89bn. By the time it was rescued last autumn, its balance sheet—boosted by the takeover of Dutch bank ABN Amro a year earlier—had swelled to almost £2.4 trillion.
Why did the banks expand so fast? Apologists have advanced two possible explanations. One is that the globalisation of capital flows has created the need for large financial institutions that can operate around the world. The second is that very large banks are more efficient.
The first is difficult to accept. If you spool back ten years, Goldman Sachs was considered to be a strong global bank. Its balance sheet is now almost five times bigger. It is simply hard to believe that the ideal size for a global bank should have risen so sharply in a single decade.
The efficiency case is also weak. Much of the academic evidence contradicts the idea that size itself makes a bank more efficient. A 2002 report for the US Federal Reserve looked at banking mergers in the US, Europe and Japan and concluded that scale provided advantages only up to a low level of total balance sheet assets (about $50bn). Beyond that there were disadvantages in greater size arising from the complexity of running such large organisations—especially multinationals.
The report did note that some very large banks generated increased shareholder returns. But this was not because they were inherently more profitable; it was because they were permitted to hold less capital against the assets on their balance sheets. They were, in effect, allowed to borrow more than smaller institutions. This was supported by policymakers because they accepted the received wisdom that large banks were safer than small ones, as their risks were more diversified. They also felt big banks could afford more sophisticated risk management systems. This view has been exploded by the crisis. One of its lessons is that the interconnectedness of today’s markets makes the benefits of risk diversification smaller than had been thought.
Do we really need big banks at all? Many bankers would emphatically say yes. “The idea that we could run modern economies with mid-sized savings banks is totally misguided,” Josef Ackermann, the chief executive of Deutsche Bank, one of the world’s largest commercial and investment banks, said in November. But their big corporate customers are unconvinced. Companies would prefer to have more, smaller banks so that they can spread their business to ensure they get the best deal. Most large commercial or investment banking transactions, after all, involve a syndicate or group of banks to divide up the risk. The same applies to foreign-exchange deals. The customer searches for offers and takes the best one.
For derivatives (and other instruments such as shares and bonds) the real requirement is for deep markets rather than big banks. And the liquidity needed for such deep markets depends on having many participants. Banking concentration inhibits this. The one area where scale is clearly an advantage is in taking principal positions. A bank with a large balance sheet (containing lots of interest-earning loans and other assets like bonds and derivatives) can use this balance sheet to extend a risky bridge loan to a company seeking to make a takeover—and thus earn hefty transaction fees. It can also use its balance sheet to trade directly on the markets. But these are both risky activities.
All that size really does is to allow banks to take on additional risks such as those mentioned above—and the state encourages them to do so by giving those banks a “too-big-to-fail” guarantee. As Mervyn King, the governor of the Bank of England, has observed, this is back to front: “Anyone who proposed giving government guarantees to retail depositors and other creditors and then suggested that such funding could be used to finance highly risky and speculative activities would be thought rather unworldly. But that is where we now are.” King’s words carry extra clout because the Conservative party has pledged to restore responsibility for overseeing banks to the Bank of England. Similarly, his public criticism of reform efforts to date have highlighted tensions with the Labour government.
But while bigger banks add little for the customer or shareholder, they have proved lucrative for top executives. Individual bankers have had an overwhelming incentive to pursue size for its own sake. And the crisis has vindicated those who have. While hundreds of small banks have failed over the past two years, Lehman’s is still the only big institution that has been allowed to collapse. Indeed some big banks, like Bank of America and JPMorgan Chase in the US and Lloyds Banking Group in Britain, have emerged from the crisis larger than before it.
If this process continues, the biggest banks may not only be too big to fail; but too big to rescue. Iceland learned this lesson last year when the collapse of its banking system—which had far outgrown its economy—triggered a fiscal crisis. Swiss officials have acknowledged that Switzerland cannot afford to take on all the liabilities of UBS and Credit Suisse, its largest banks. And Britain, which has a banking sector with assets worth five times its annual national economic output, is also vulnerable. This is why the doom loop is such a threat.
To escape its effects, it will be necessary not only to curb the urge to ever greater scale but to roll it back. This means finding a way to shrink the very large banks, and reverse the forces that encouraged them to become so large in the first place. This will not be easy. Despite their recent humbling, large banks retain formidable lobbying power. Yet the alternative is to assume that regulators will become much smarter in the way they that regulate them. And that is simply Pollyanna-ish.
Until now, the advocates of shrinkage have focused on structural measures to cut bank size, such as crude caps on balance sheets. The trouble is that these require you to set a somewhat arbitrary limit, and open up the possibility of all sorts of harmful arbitrage. Another approach that has been advocated is to force banks to separate their retail banking from their investment banking operations—splitting the “utility” from the “casino.” This recalls Glass-Steagall, the depression-era US law that barred commercial banks from underwriting or trading securities on Wall Street.
Some have taken the idea of such a split even further: the economist John Kay has argued that the way forward is to create regulated “narrow banks,” which would take deposits and invest them only in safe government bonds. Yet even if a new Glass-Steagall-style separation could be achieved, it is hard to see that it would make the system safer. The problems of the crisis were not limited to investment banking: Northern Rock was about as far removed from a complex investment bank as it was possible to get. Moreover, there is no guarantee that the “casino” side of the industry could be credibly left free of state support. A decade ago, the failure of Lehman Brothers might not have caused too much trouble (a well-known City investment bank, Barings, collapsed in 1995 with barely a ripple). By last year, it had become too entwined in the financial system, acting as a large trading counterparty to banks, companies and investors.
Indeed, Kay’s plan might make the financial system less, not more, stable. If narrow banks can only lend to the government then all other lending, including mortgages, credit cards and small business loans, would have to be funded in the wholesale financial markets—precisely the area that has seized up over the past two years. Moreover, these arguments for separating retail from investment banking do not explicitly deal with the question of size. After all, Glass-Steagall was conceived when the US had thousands of small banks, not a handful of big ones. Even if it were possible to engineer a similar split on a global basis, it would not prevent the existence of a large investment bank like Goldman Sachs. Yet it is precisely size that has proved to be so dangerous.
WHICH REFORMS WILL WORK?
There is no “silver bullet,” as Adair Turner, chairman of the Financial Services Authority puts it, to deal with the size problem. What is needed is a series of interlocking policies to tackle pointless bigness. These would neither set arbitrary limits to balance-sheet growth nor deter expansion that genuinely added value. But they would stop the sort of reckless dash for scale that we have seen.
The main thrust of banking reform over the past two years has been to make banks safer by expanding their buffers of capital and liquidity—thus reducing the likelihood of future failures. Regulators are already shifting away from the approach that allowed large banks to hold less capital against their assets. Turner has proposed forcing large banks to hold greater capital buffers than smaller ones. This would act to balance out any benefits banks enjoy from being big.
A second approach worth adopting would be to deal with the cross-subsidy that state-guaranteed deposits supply to risky trading activities. Restoring Glass-Steagall may be a step too far, but banks should be forced to separate risky activities and fund them without recourse to the guaranteed parts of the institution. These activities could be defined. So, for example, investors lending to a bank’s in-house hedge fund would be exposed to that unit, and not lumped in with the depositors. Citigroup’s former chairman, John Reed, likens it to the compartmentalisation of a ship: “If you have a leak, the leak doesn’t spread and sink the whole vessel.”
Lastly, to deal with the risk of global banks becoming too big to rescue, such institutions should be organised into a confederation of national subsidiaries, each with its own reserves of capital. In a crisis, governments would then be responsible for bailing out their local subsidiary. This would prevent the full costs of rescuing a large bank from falling onto the taxpayers of the bank’s home country. The idea has won support from large banks like HSBC and Santander, the latter of which already operates along these lines. António Horta-Osório, chief executive of Abbey, Santander’s subsidiary, told a London audience in November that the structure creates a “firewall” that can prevent contagion. Investment banks, which tend to operate on a global basis, are less keen.
If these measures were fully implemented, the result should be that banks would become smaller again. For if the costs of being big outweighed the benefits, then bank executives would be under pressure from markets to shrink. Those banks that insisted on operating on a global scale would have to prove to their investors that it was worthwhile.
Simply by making it a regulatory objective to limit unnecessary size, the powers that be would be sending a message. And if it doesn’t work? Then the only option would be to disprove Bove’s assessment—that some organisations cannot be eliminated—in a cruder fashion. Governments would have to insist that banks could not grow bigger than a certain size. That might be unsatisfactory, but it would be necessary to protect taxpayers from the blunders of pointless big banks.
A BUFFER’S GUIDE TO BANKING
Banks are so central to the modern economy that it is impossible to imagine life without them. But, as the past two years have shown, they are also dangerous. The vulnerability of banks rests on an innovation, dreamed up almost 400 years ago, called fractional reserve banking. Before then, banks acted mainly as places where people could keep coins and precious metals. Banks issued notes to their customers promising to return their coins on demand. Over time, customers realised that it was simpler to exchange the notes rather than retrieve their coins: paper money was born.
An even bigger shift came when banks realised that only a small number of customers were likely to ask for their coins back at the same time. Therefore the bank needed to keep only a fraction of the coins it had on deposit; the rest could be lent out in return for interest. This process—taking in short-term deposits and lending them out for longer periods—is known as “maturity transformation.” In fact, banks now lend much more than they have on deposit—in effect creating money.
This all has great benefits, but it also makes banks vulnerable to a loss of confidence. That is why governments regulate banks more than other enterprises and often step in with liquidity help when banks do not have enough short-term cash to satisfy depositors. (It is also why most countries have schemes to insure depositors against losing their money.)
Banks also need capital of their own (equity capital and other forms) to cover bad loans. But during the boom the proportion of capital to loans (loans which generate interest are a banks “assets”) fell as low as 3 per cent. This is called an increase in leverage. Until the crisis, regulators had been happy to allow banks, especially big ones, to hold less capital because it was assumed that their risks were well spread. Understandably, regulators now want banks to increase their capital reserves.