The perils of "too big to fail"

 Simon Johnson - former chief economist for the International Monetary Fund -  at NYTs "Economix" on the implications and perils of "too big to fail."  (One question as food for thought - if, as Johnson notes, banks are financed mostly by debt rather than equity, why are these institutions so beholden to stockholders, who aren't putting up much stake in the project relative to their ability to profit and the unprecedented "security" of their limited investment because of "too big to fail" ?):

In an interview with The New York Times in July, Sheila Bair, the departing chairwoman of the Federal Deposit Insurance Corporation, said of her experience over the last few years: “They would say, ‘You have to do this, or the system will go down.’ If I heard that once, I heard it a thousand times.”

No responsible official wants the entire financial system to crash; this would be incredibly disruptive to all Americans and potentially lead to a worldwide depression. Knowing this, many people who want bailouts on generous terms use “contagion fear” as part of their sales pitch.

How are we to know if a particular event, like deciding not to bail out a big bank, will lead to contagion that spreads to other financial markets? Contagion is the key issue.


In Europe, the failure of just one large bank can do macroeconomic damage; governments there have let individual banks build balance sheets that are bigger than the country’s gross domestic product. In the United States, we too should fear megabanks; the big six bank-holding companies have become much bigger in recent years and now have combined assets worth more than 65 percent of G.D.P.

But even the biggest (currently JPMorgan Chase) has a balance sheet of just under $2.3 trillion, while our G.D.P. is around $14 trillion. (Each of the top 10 global banks is around the $2 trillion mark, according to the latest available data from The Banker, but they are based in countries of very different sizes. Be careful in using these numbers as accounting conventions for banks may vary – in particular, derivative exposures are arguably understated in United States banks’ published accounts relative to European banks.)

If JPMorgan Chase were in trouble – a completely hypothetical scenario – and its creditors faced, for example, 40 percent losses, this would be very bad news (40 percent is an arbitrary number often used as a baseline by creditors thinking about potentially distressed situations; it looks as though creditors to Lehman Brothers will, on average, end up losing a good deal more of their exposure).

JPMorgan Chase is mostly financed with debt because, alone among modern executives, bankers do not believe in much equity financing for their businesses. So the financial losses to debt holders could be on the order of $1 trillion.

That would be enough to bring on a serious recession. But it might not, by itself, bring the world economy to its knees.

The case for arguing that JPMorgan Chase is too big to fail rests on the notion that its failure or the prospect of serious losses at that one bank would create fear across a broad range of other assets. Asset price declines could then push other banks into insolvency; we saw some of this in the fall of 2008. There is a potential downward spiral.

One response to this is: Let ’em fail. That’s tempting, particularly if you think that one set of traumatic failures would teach everyone that, next time, there will be no bailouts.

Experience from the 19th century in America is not particularly encouraging in this regard. There were no bailout mechanisms – the Federal Reserve was not invented until 1913 and the federal government did not have the resources to save the financial system, even if it wanted to.

Nevertheless, the United States experienced serious banking panics at regular intervals throughout the century, sometimes due to disruptions beyond the control of bankers but often due to the overextension of credit. Banks in those days had to hold “specie” (gold and silver coin), which was generally payable on demand when people wanted to redeem their bank notes (which were issued by private banks, not the government, before the Civil War).

“Suspensions” of the convertibility for bank notes into specie were disappointingly frequent. During the period 1815-30, according to Davis Rich Dewey’s “Financial History of the United States” (see page 155), “Failures of banks were common and bill holders and depositors suffered much.” There was a major panic in 1819, and another in 1837, perhaps brought on by the policies of President Andrew Jackson. There was a panic in 1857 in the run-up to the Civil War, and also in 1860 and 1861...

Much of Charles Kindleberger’s classic financial history, “Manias, Panics and Crashes” – the title tells you everything – covers the period before central banks operated in their modern form. Various schemes at the state level – and later at the federal level – attempted to control or constrain risk-taking by individual banks, but the unfortunate reality is that the financial lifeblood of American commerce has always been prone to lending too much and incurring heavy losses.

People see one bank failure and fear the consequences – hence, panics and runs on the bank. Anyone who thinks that we solved this problem with forward-thinking central banks or fiscal interventions has not been paying close attention – either to the United States or to Europe – over the last three years.

We really need transparency on the exact exposures of banks. To whom have they lent and on what basis? Just telling supervisors does not help us at all, because sharing information only behind closed doors is just another potential mechanism for regulatory capture.

For large financial institutions – those with more than $100 billion in total assets – everything should be out in the open. If you want to operate in relative secrecy, stay small. The costs of today’s opaqueness are huge, creating the basis for the plea, “Save us or you’ll lose the world.”

And if the published information is too complex for outsiders to understand, big banks must be forced to simplify their operations until they become sufficiently transparent. Potential contagion risks must be measurable and apparent to the marketplace, including to independent analysts who have access only to public information.

Needless to say, if any institution poses major contagion risks, as measured on this basis, it should be forced to become safer.

Richard W. Fisher, president of the Federal Reserve Bank of Dallas, spoke recently about what he called the “pernicious problem” of too-big-to-fail financial institutions, “in a culture held hostage by concerns for ‘contagion,’ ‘systemic risk’ and ‘unique solutions’”:

“Invariably, these behemoth institutions use their size, scale and complexity to cow politicians and regulators into believing the world will be placed in peril should they attempt to discipline them. They argue that disciplining them will be a trip wire for financial contagion, market disruption and economic disorder. Yet failing to discipline them only delays the inevitable – a bursting of a bubble and a financial panic that places the economy in peril.”

You can only discipline a big troubled bank if you understand and can measure the consequences of its failure.