Too-big-to-fail is here to stay

By Felix Salmon
January 19, 2010
Andrew Ross Sorkin has a slightly odd column today, asking whether it's OK to have too-big-to-fail banks, and concluding -- well, not really coming to any conclusion at all, actually.

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

Andrew Ross Sorkin has a slightly odd column today, asking whether it’s OK to have too-big-to-fail banks, and concluding — well, not really coming to any conclusion at all, actually.

The oddest part of Sorkin’s column is that it has all the ingredients necessary to come to the sensible conclusion: that, as he quotes Alan Greenspan as saying, “if they’re too big to fail, they’re too big”.

But Sorkin follows up with this:

It was a surprising statement from Mr. Greenspan, given his free-market comment in 2005 that “private regulation generally has proved far better at constraining excessive risk-taking than has government regulation.”

There are two problems here. Firstly, Greenspan has since recanted that belief. And secondly, the problem with having too-big-to-fail banks is not that small-enough-to-fail banks are less likely to have excessive risk-taking. As Sorkin quotes Jamie Dimon as saying, banks of any size can run into problems. But the point is that if a small-enough-to-fail bank takes too many risks and fails, the systemic consequences are manageable. If a TBTF bank takes too many risks and fails, it can drag down the entire economy.

So why allow banks to get that big? Sorkin tries, valiantly, to come up with a remotely compelling answer to the question:

Shrinking the size of these companies may create other problems for the economy, particularly in this age of huge corporations.

If Pfizer, for example, needs to raise $20 billion for a takeover bid, or Verizon needs to raise billions to lay fiber optic cable for its FiOS service, they cannot efficiently go to 20 different community banks looking for the money.

And if corporations can’t readily obtain financing in the United States, they may simply seek out huge overseas lenders like HSBC and Deutsche Bank.

The fact is that if Pfizer or Verizon need to raise billions of dollars, there are broadly three ways they can do so. They can ask an investment bank to help them raise the money in the stock market; they can ask an investment bank to help them raise the money in the bond market; or they can ask a large commercial bank to help them raise the money in the syndicated loan market.

Pfizer and Verizon are well known in the bond markets, where thousands of institutional investors trade those credits every day. And when banks enter the syndicated loan business, they rarely make significant profits from the loans themselves. Instead, the lead managers get fees for putting the deal together, and selling off most of the loan to banks which have excess liquidity and/or want to develop their relationship with the company in question.

Ultimately, however, the economy would be better off if big companies raised more equity and less debt; and borrowed what they needed to borrow in the bond market rather than the loan market, leaving the banks to do what the bond market can’t, which is lend to individuals and small businesses.

And even if that doesn’t happen, I don’t think that giving HSBC and Deutsche Bank a little bit of a boost in the international syndicated-loan league tables is a particularly high price to pay for massively reducing the systemic risk inherent in the US financial system.

Sorkin then continues:

Think of the biggest dominoes: Bear Stearns, Lehman Brothers, Merrill Lynch and Morgan Stanley. They weren’t financial supermarkets.

Mr. Dimon was trying to make this point: Companies should be allowed to be as big as they want, so long as there is an orderly way to wind them down.

Actually, Bear, Lehman, Merrill, and Morgan Stanley were all too big to fail: you don’t need to be a financial supermarket to pose a systemic risk, although it helps. And as we saw over the course of the crisis, there is no orderly way to wind down a TBTF financial institution, short of bailing it out, like we did with Fannie Mae and Freddie Mac and AIG, with billions of dollars of public funds. The likes of Jamie Dimon love to talk about “resolution authority” not because it’s a good idea or even practicable, but just because they believe (with good reason) that if they continue to talk about this chimera as though it were real, regulators might refrain from cutting them down to a manageable size.

We’ve learned the hard way that financial institutions can’t continue to operate as a going concern when the markets don’t have confidence in them any more. A confidence crisis is a liquidity crisis, and resolution authority doesn’t address liquidity concerns. If a bank is forced to default on its unsecured debt, it will fail, and it’s simply not realistic to believe otherwise. Resolution authority might protect the taxpayer from having to step up with enormous bailouts, but it doesn’t reduce systemic risk.

If you want to truly address the problem of TBTF banks, there’s only one way to do so: make them smaller. But the fact is that Jamie Dimon et al don’t need to worry about that happening any time soon. None of the financial regulations currently being debated would force them to shrink noticeably. Which means that they will continue to pose a massive risk to the global economic system for the foreseeable future.

More From Felix Salmon
Post Felix
The Piketty pessimist
The most expensive lottery ticket in the world
The problems of HFT, Joe Stiglitz edition
Private equity math, Nuveen edition
Five explanations for Greece’s bond yield
Comments
9 comments so far

Exactly. You could make the case that big investment banks are needed to underwrite huge syndicated loans and then sell them on to a mixture of commercial banks, loan funds and other CLOs. You then need to explain why these huge underwriting commitments disappear at the first sign of trouble and borrowers have to go back to negotiating with a large club of commercial bankers. This inconvenience is also the price worth paying for a reduction in systemic risk.

Posted by gringcorp | Report as abusive

You shouldn’t be surprised by Sorkin’s column. I am not.

Sorkin wrote this book that everybody loves. It’s full of juicy gossips about the rich and powerful. He evidently couldn’t have written it without tremendous access. So, why should those bankers and high government officials grant him such access? What do they get in exchange? This kind of column is what they get, it’s the quid pro quo. Sorkin also wrote an article defending the “independence” of the fed.

It’s high time NYT starts this paywall. It would shield us from this kind of propaganda.

Posted by EmilianoZ | Report as abusive

“A confidence crisis is a liquidity crisis, and resolution authority doesn’t address liquidity concerns.”

…and capping the liabilities of any one financial institution, either explicitly or via a tax, somehow addresses these confidence/liquidity problems? Howso?

“Resolution authority might protect the taxpayer from having to step up with enormous bailouts, but it doesn’t reduce systemic risk.”

Agreed that providing a mechanism for winding-up financial institutions would not unambiguously reduce systemic risk. But what would? It’s not plain to me how a liability cap would mitigate systemic risk. Would capping bank size reduce the likelihood of a systemic crisis, the severity thereof, or both? Howso?

Posted by Sandrew | Report as abusive

“A confidence crisis is a liquidity crisis . . . .” Only if the lack of confidence is mistaken. If it’s accurate, the institution is insolvent; that makes the crisis an insolvency crisis.

“. . . make them smaller.” How much smaller? There’s no bright line here.

Posted by Philon | Report as abusive

What about the example of Canada? If I’m not mistaken, the Canadian financial system is dominated by a very few very large banks (large in relation to Canada’s GDP) which are tightly regulated and therefore didn’t encounter any serious trouble over the past couple of years.

I could very well be mistaken about this, btw, I am not an expert on Canada’s banking system, I just sort of remember having heard this somewhere.

Posted by expatsp | Report as abusive

We’ve learned nothing.

“Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius — and a lot of courage — to move in the opposite direction.”

“We can’t solve problems by using the same kind of thinking we used when we created them.”

“The definition of insanity is doing the same thing over and over again and expecting different results”.

Posted by csodak | Report as abusive

Yes csodak, and ‘God does not play die with the Universe’.

Felix, it is ‘Too Big, Has Failed’. TBHF. It has another name: LIBOR.

FISW.

TGIF.

Posted by Ghandiolfini | Report as abusive

What about the example of Canada? If I’m not mistaken, the Canadian financial system is dominated by a very few very large banks (large in relation to Canada’s GDP) which are tightly regulated and therefore didn’t encounter any serious trouble over the past couple of years.

I could very well be mistaken about this, btw, I am not an expert on Canada’s banking system, I just sort of remember having heard this somewhere.
======================================== ===============
The Canadian banking system is ranked #1 in the world mostly due to it’s well thought out and conservative approach to investment and lending. Ironically the areas in which they were hurt most were in their numerous American bank aquisitions. It boils down to better and more stringent regulation and a more common sensical approach. The US has allowed a more unrestrained and irresponsible approach to lending and teh fabrication of fiscal products that have no basis in good business.

Posted by Phantus | Report as abusive

Phantus: Canada is different from the US in enough ways that it probably isn’t intellectually sound to say that their less disastrous banking system over the last few years indicates that their banking model is correct and would work in the US.

Most importantly, there aren’t that many people there and the population is more homogenous and stable than the diverse/dynamic US population. Secondly, they have a huge amount of natural resources relative to their population (oil, NG, timber, gold, etc.). Thirdly, they have single-payer healthcare, while medical costs are a huge cause of bankruptcy in the US (look at Elizabeth Warren’s research on this). Finally, most parts of Canada simply aren’t very attractive places to live (cold, undeveloped, far away from everything), so you don’t get the same speculation on sunny vacation homes that fueled bubbles in FL, CA, NV, AZ, etc.

If you compare the Canadian banking system to the banking systems of Minnesota and Vermont, I think you’d see fewer differences. Being up in the north in less densely-populated areas they simply had fewer opportunities to speculate on disastrous condo developments, subprime mortgages to speculators, and huge CDO deals (although they did manage to get in a little bit of activity before the crash – if you look back at filings the major Canadian banks weren’t totally free of subprime and structured finance problems).

To sum up, I don’t think Canadian banks are that much smarter than BAC or C – I think they were just lucky not to be in the same proximity to the most awful bubbles. Given the opportunity to make huge money speculating in (for instance) California real estate, most people (especially bankers) take it.

Felix: Retrospectively, the way we should have managed large institutions like Bear, Merrill, etc. was to offer them a choice: either hold more capital or accept and acknowledge that if you misprice risk you will fail and your stockholders, debtholders, and counterparties will all suffer.

If those banks had publicly made the choice to risk up and markets had a clear sense that government would not bail them out, there is no way that they could have borrowed enough money or done enough unsecured transactions to get as big as they did or pose the problems they did.

On the other hand, if large banks chose to hold more capital and to proactively remedy capital problems when trouble materialized (everyone dragged their feet even after problems became obvious in 2007), again they couldn’t get as big because equity investors would be less interested in the lower rates of return possible while holding higher capital.

If everyone is holding high enough capital, when one institution really blows it and runs into trouble, better-capitalized institutions supply capital on punitive terms and profit from their prudence. This is how markets work and it’s a great thing. If management prefers to hold out for a bailout or irrationally cheap funding, let them blow up the company – it won’t take too many iterations of this mistake before most reasonable people learn the lesson. You can’t regulate all unreasonable people out of existence and it’s foolish to try. Their failures are great lessons for the marginally reasonable.

The only regulation you need is fraud prevention/response, minimization of information asymmetries between banks and investors through disclosure rules and education, a plan for resolution of failed banks, and a requirement to hold enough capital that speculators can’t do too much damage with other people’s money. This is quite a bit of regulation and it’s hard enough – we don’t need federal regulators who couldn’t even manage these tasks to take on a bunch of new ill-defined missions involving figuring out the right credit card terms and the magical size of the ideal bank.

Posted by najdorf | Report as abusive
Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/