From Matt Taibbi:

Wall Street is buzzing about the annual report just put out by the Dallas Federal Reserve. In the paper, Harvey Rosenblum, the head of the Dallas Fed’s research department, bluntly calls for the breakup of Too-Big-To-Fail banks like Bank of America, Chase, and Citigroup.

The government’s bottomless sponsorship of these TBTF institutions, Rosenblum writes, has created a “residue of distrust for government, the banking system, the Fed and capitalism itself.”

I don’t know whether to laugh or cry.

First, this managerialism is nothing new for the Fed. The (ahem) “libertarian” Alan Greenspan once said: “If they’re too big to fail, they’re too big.”

Second, the Fed already had a number of fantastic opportunities to “break up” so-called TBTF institutions: right at the time when it was signing off on the $29 trillion of bailouts it has administered since 2008. If the political will existed at the Fed to forcibly end the phenomenon of TBTF, it could (and should) have done it when it had the banks over a barrel.

Third, capitalism (i.e. the market) seems to deal pretty well with the problem of TBTF: it destroys unmanageably large and badly run companies. Decisions have consequences; buying a truckload of derivatives from a soon-to-be-bust counter-party will destroy your balance sheet and render you illiquid. Who seems to blame? The Fed; for bailing out a load of shitty companies and a shitty system . Without the Fed’s misguided actions the problem of TBTF would be long gone. After a painful systemic breakdown, we could have created a new system without any of these residual overhanging problems. We wouldn’t be “taxing savers to pay for the recapitalization of banks whose dire problems led to the calamity.” There wouldn’t be “a two-tiered regulatory environment where the misdeeds of TBTF banks are routinely ignored and unpunished and a lower tier where small regional banks are increasingly forced to swim upstream against the law’s sheer length, breadth and complexity, leading to a “massive increase in compliance burdens.”

So the Fed is guilty of crystallising and perpetuating most of these problems with misguided interventionism. And what’s the Fed’s purported answer to these problems?

More interventionism: forcibly breaking up banks into chunks that are deemed not to be TBTF.

And what’s the problem with that?

Well for a start the entire concept of “too big to fail” is completely wrong. The bailout of AIG had nothing to do with AIG’s “size”. It was a result of systemic exposure to AIG’s failure. The problem is to do with interconnectivity. The truth is that AIG — and by extension, the entire system — was deemed too interconnected to fail. Many, many companies had AIG products on their balance sheets. If AIG had failed (and taken with it all of that paper, very generously known as “assets”) then all those companies would have had a hole blown in their balance sheets, and would have sustained losses which in turn may well have caused them to fail, bleeding out the entire system.

The value that seems to matter in determining systemic robustness is the amount of systemic interconnectivity, in other words the amount of assets on balance sheets that are subject to counter-party risk (i.e. which become worthless should their guarantor fail).

Derivatives are not the only such asset, but they make up by far the majority:

Global nominal exposure is growing again. And those derivatives sit on global balance sheets waiting for the next black swan to blow up a hyper-connected counter-party like AIG. And such a cascade of defaults will likely lead to another 2008-style systemic meltdown, probably ending in another goliath-sized bailout, and another few rounds of the QE slop-bucket.

The question the Fed must answer is this: what difference would it make in terms of systemic fragility if exposures are transferred from larger to companies to smaller ones?

Breaking up banks will make absolutely zero difference, because the problem is not the size but systemic interconnectivity. Losses sustained against a small counter-party can hurt just as much as losses sustained against a larger counter-party. In a hyper-connected system, it is possible for failed small players to quickly snowball into systemic catastrophe.

The Fed (as well as the ECB) would do well to remember that it is not size that matters, but how you use it.