One question that a lot of market watchers have is why bank failures seem to cascade. Why do we keep discovering that so many banks apparently follow very similar investment strategies? Don't the banks employ lots of very well paid people to make decisions about which loans to make, which assets to buy, and where to invest? How is it they all keep deciding the same thing?
The fact that so many banks buy the same kinds of assets is what makes for a crisis. If it were just one or two banks that unwisely over-indulged in mortgages or sovereign debt, the financial system could easily handle their failure. This is why, for instance, the failure of MF Global didn't provoke a crisis: most other US financial institutions weren't following Jon Corzine's strategy of loading up on the debt of Italy, Spain and Ireland.
It turns out that there are very good reasons banking is so homogenized. In the first place, the capital regulations around the world incentivize banks to over-invest in certain favorable asset classes. The rules for capital cushions encouraged banks to buy mortgages and sovereign debt. In Europe, banks weren't required to have any capital cushion at all for sovereign debt, which made government bonds very attractive.
Another factor driving homogenized banking is the potential for bailouts. Banks that buy only what other banks are buying know that they won't fail unless other banks are also failing, in which case central bankers and governments will intervene to prevent a systemic crisis. Banks that strike out on their own are allowed to fail if their bets are wrong.
Imagine, for example, if a dollar-eating microbe destroyed all the dollars in a single bank's vault. The depositors would be insured and wouldn't lose any money. But the bank itself would be out the money, unless it had private insurance for dollar-eating microbes. The Fed probably wouldn't intervene to replace the lost dollars.
Now imagine that the same microbe destroyed the dollars in every bank vault in the country. In that case, the Fed would intervene to replace the lost currency. When a problem becomes systemic, regulators step in.
This pattern is built right into the law. Dodd-Frank prohibits the Fed from launching emergency facilities for single institutions. If this law was in place in 2008, the Fed probably couldn't have provided AIG with the emergency loan that kept the company out of bankruptcy. The Fed is only allowed to open rescue funds if they are generally available to all members of the banking system.
This means that "going it alone" is riskier for a bank’s management, as well as for its creditors and shareholders. Everyone involved rationally rewards banks that follow the pack with higher share prices and more attractive loans than are available to individualist banks.
Ironically, it's this very pattern that creates systemic risk. When the strategy everyone is following goes wrong, the system begins to collapse, and the entire banking herd runs off the cliff together.
Is there a way out of this trap? Surprisingly, regulators may have stumbled upon it in 2008. Allowing Lehman Brothers to fail even though it wasn't really much worse off than, say, Merrill Lynch, created the possibility that bond holders could lose money even with banks that are part of the banking herd. This could have diminished the benefits that members of the banking herd enjoyed.
The lesson was all the more powerful because of the seeming arbitrariness of letting Lehman go while rescuing Merrill or Morgan Stanley. If the point is to take away the benefits of being in the herd—and therefore discourage herd banking—you have to cull almost at random.
Unfortunately, the experience of letting Lehman fail is so widely deemed as a disaster that many are confident regulators will never again take that route. So the herd mentality of banking is once again firmly in place.
Questions? Comments? Email us atNetNet@cnbc.com
Follow John on Twitter @ twitter.com/Carney
Follow NetNet on Twitter @ twitter.com/CNBCnetnet
Facebook us @ www.facebook.com/NetNetCNBC