New Mega-Bank Capital Requirements Could Create Systemic Risk

While we're supportive of efforts to force the banking behemoths to carry additional capital, there are reasons to be cautious. Capital requirements are not a panacea.

The global regulators in Basel agreed to a sliding scale surcharge on "systemically important financial institutions" (or Sifis) that will range from 1 percent to 2.5 percent, with the ability to rise to 3.5 percent if the Sifi keeps growing. The best part of the surcharge is not that it is a guarantee against failure (it isn't) or against bailouts (it isn't that either). It's that it will reduce the advantages Sifis enjoy by virtue of being too big to fail.

This could actually reduce the size of the behemoths over time because one of the principal drivers of giantism in banking is not economies of scale or economies of scope—these simply vanish beyond a far smaller and more reasonable size. The driver is not purely economic at all. It's political: The biggest banks have the most political influence and the political machinery will not allow them to fail in a disorderly way that damages their creditors. Take away the funding advantage by making their assets more expensive for them and you've gone a long way to reducing the cause of banking giantism. (Or should I call it Sifism?)

The danger is that increasing capital requirements also increases the incentives to game the requirements. It makes finding arbitrage opportunities more rewarding.

This is not a small danger. One of the principal reasons the bursting of the housing bubble triggered a financial crisis was that bank balance sheets held too many mortgage-related assets. The reason for this was simply that bank capital regulation rewarded this.

The pre-crisis bank capital rules required just $2 in capital for every $100 in highly-rated mortgage-backed securities. Commercial loans required $10 of capital for every $100. Non-securitized mortgages required $5. It was obviously far more profitable to hold highly rated mortgage-backed securities than any other bonds with similar yields.

Why did the capital rules work this way? In part, it was because regulators shared the ideology of many bankers according to which a national housing downturn was all but impossible. Highly-rated mortgage bonds had lower capital requirements because they were considered safer than other investments.

Risk-weighting assets, as this system of requiring different amounts of capital for different bonds is called, became a source of systemic risk. It encouraged a buildup of mortgage risk on bank balance sheets. It fed the housing bubble by creating a limitless demand for mortgage-backed securities—a hidden form of credit expansion. It drove down yields for traditional mortgages, sending banks scrambling for higher yields in non-traditional mortgages such as subprime.

Banks were not required to adopt the regulatory view of housing risk.

But they were penalized for rejecting it. As a result, many of the banks did adopt the regulatory view—and when that turned out to be mistaken, many failed.

Higher capital requirements will increase the cost of rejecting the regulatory view of risk. This means that risk-weighting becomes an even higher-stakes game, and the regulatory contribution to systemic risk even more important. To put it differently, higher capital requirements make banks even more vulnerable to regulatory error.

This is depressing. We need higher capital requirements to restore a market balance to the funding advantages of Sifis. But the cost of them could be more systemic risk from homogenization.

_________________________________________

Questions? Comments? Email us [e-mail]

Follow John on Twitter @ twitter.com/Carney

Follow NetNet on Twitter @ twitter.com/CNBCnetnet

Facebook us @ www.facebook.com/NetNetCNBC