Apologies for the month-long silence.
The US and UK financial regulators responsible for dealing with the next Lehman Brothers disaster have been making joint plans to do so. A couple of days ago they released a draft. The next Lehman is likely to be one of the other so-called Globally-active Systemically Important Finanicial Institutions (G-SIFIs). The list of 28 includes the likes of Barclays, Goldman Sachs, HSBC, RBS, Santander.
Normally when a company (banks are companies) goes bankrupt it is dealt with by bankruptcy law. This lays out a well-established process that’s been working for decades, for companies large and small, from airlines to car manufacturers. The people who lent money to the company (the creditors) get together with the company’s owners and try to negotiate a deal. If the creditors agree to reschedule payments or reduce the amount to be paid back to a manageable amount (in exchange for e.g. reorganising the business so that payback is more likely) then the company can stay in business. If not then company is closed down and the corporate equivalent of a yard sale is held. Either way, this takes time.
However, G-SIFIs are not like ordinary companies. That’s true in many ways but let’s focus on one. In the time it would take for bankruptcy proceedings to complete, the rest of the financial system would have probably gone into meltdown. Too many other financial companies (including the other G-SIFIs) are owed too much money by the G-SIFI that is failing. In short, our financial system is fragile by design.
So what is the solution that our regulators are proposing? Stephen Lubben at Credit Slips provoked me to look at the draft. Let us being at paragraph 10: “Following the crisis, an overhaul of the framework for dealing with large and complex financial institution failures was required. While it may be useful to strengthen the current bankruptcy code or administration rules to improve the handling of financial failures, systemic considerations warrant having an alternative resolution strategy.”
Notice the implicit assumption here: we are going to have to live with large and complex financial institution failures; therefore we should prepare to deal with such failures.
Er, anybody else spot the ELEPHANT IN THE ROOM! Why do we have to live with large and complex financial institutions that are prone to failure? How about instead we eliminate them e.g. by breaking them up into smaller and more locally-orientated organisations. Indeed the US regulators do indicate than when they get a chance to use their resolution powers, they anticipate using them to break the banks up: “The FDIC would also likely require the restructuring of the firm—potentially into one or more smaller, non-systemic firms that could be resolved under bankruptcy” (paragraph 25). (All we get from the UK regulators is “the firm would be restructured to address the causes of its failure” in paragraph 35.)
To be fair, the drafters of this document were bureaucrats who couldn’t overstep their remit to start talking about breaking up the banks BEFORE they fail. I however am (currently) under no such constraints.
Breaking up the G-SIFIs is a credible option. The Independent Commission on Banking explored it, and though it opted for a ring fence in its final recommendations, it seriously considered and saw many merits in full structural separation of retail banking and riskier activities. Had full separation been pursued, it would have led to the break-up of the G-SIFI. Breaking up companies for competition reasons (requiring the disposal of subsidiaries or certain operations that threaten competition) is not unusual so why not for G-SIFIs that threaten society with financial calamity and not just weak competition?
What is lacking here is a pressure from people like you and me on and support of politicians to break up G-SIFIs. Without us there is little counterbalance the enormous pressure exerted by the army of lobbyists hired by the G-SIFIs.
G-SIFIs will be with us for some time, even if popular pressure to break them up were to arise, so it’s right and good that our regulators are working hard on resolution plans for them. But we shouldn’t let that distract us from breaking up the G-SIFIs into… what shall we call them then? How about L-HUFIs (Locally-orientated Harmless but Useful Financial Institutions)? (Alternatives welcome in the comments section below!)
Reframe Finance 
In theory, I agree with your sentiments, however as Neil Sedaka once (well, twice actually) pined, “breaking up is hard to do”. Remember, when a company goes into administration it is a last ditch effort to save a failed business. On a risk/reward calculus there is very little risk (since the business is already bust) and everything to gain. In a break-up scenario however, banks (e.g. the people who run them) would be required to manage the deconstruction of incredibly complex organisations while continuing to run their day to day operations – no small task! As I am sure you know, most mergers and acquisitions go rather badly. Reverse engineering businesses the size Barclays or RBS would be extremely difficult to do and if done poorly could result in the very calamities they are intended to avoid. The next biggest obstacle of course is cost – not merely the administrative costs associated with executing the break-ups (which would certainly run into billions of dollars) but the ongoing operating costs of running the smaller entities. Most of the G-SIFIs have large (and largely effective) shared service facilities that provide non-core services (e.g. HR, Legal, IT, etc.) to their respective business units. The “L-HUFIs” (as you cheekily call them) would almost certainly have to bear those costs directly at a higher rate than they currently enjoy in a shared system. These aren’t insurmountable problems, but they add significantly to the risk side of the risk/reward calculus.