Obama: End Proprietary Trading, Big Banks

Under the proposed rule, commercial banks would be prohibited from owning, investing in or advising hedge funds or private equity firms. Bank regulators would not be simply given the discretion to enforce such rules. They would be required to do so.
...
Administration officials said they also want to toughen an existing cap on bank market share. Since 1994, no bank can have more than 10% of the nation's insured deposits. The Obama administration wants that cap to include non-insured deposits and other assets.

From here.

I like the prohibition on commercial banks making equity investments on their own account, although the devil is in the details (for example, exceptions might be made for holdings that are tiny relative to bank assets and for assets acquired through the debt collection progress).

The biggest lesson that the financial press has largely failed to learn from the financial crisis is that FDIC regulation of excessive risk taking worked. Commercial banking has had an unprofitable few years, and bank failure rates have been above average, but on the whole, FDIC insured institutions and credit unions came out of the financial crisis relatively unscathed. Depositors haven't lost money. Bank runs in progress have been promptly stopped. Risky loans were mostly made by financial players outside the commercial banking industry. Bailout money loaned to commercial banks is well on its way to having been paid back. The amount that the FDIC has been forced to pay out to resolve failed institutions is small enough that the pre-existing FDIC premium charged on insured deposits will pay it back in a reasonable time without directly taxing anyone outside the industry.

In contrast, I'm deeply skeptical of the market share rule. I agree that excessive concentration isn't good from a systemic risk perspective, but the history of this part of anti-trust law in other industries is not very promising. It would be better to rules that take away the incentives that drive market concentration than to impose an arbitrary ten percent rule.

The argument that banks need to be global to serve the cash management needs of a global customer base doesn't cut it. This part of the business is a payment system and could be segregated from the banking fundamentals, just as the Mastercard/Visa system segregates the payment system element of credit cards from the credit granting part. Mastercard/Visa honestly is too big to fail, but because it takes on very little risk itself and is a producer cooperative owned by the banks (i.e. a mutual company), this isn't a big deal.

A more plausible argument is that large banks can make big loans to large enterprises with fewer transaction costs and better credit risk analysis than the alternatives, which are private equity funding by a small consortium of high asset investors, and public offerings of bonds by investment banks market to lower asset investors by a consortium of investment banks.

But, is that really true? I don't know. Certainly, lots of the lending that banks do: credit cards, car loans, mortgages and small business loans don't require a bank to be particularly large. But, perhaps big multi-national companies need bigger loans.

Still, even if that is true, maybe the extra transaction costs that an unavailability of bank lending imposes in the rarefied world of very large loans to very large businesses is worth it.

If a megaloan from a megabank goes bad, the whole megabank may collapse with systemic risk consequences (the old maxim is: if you owe the bank a million dollars, the bank owns you; if you owe the bank a billion dollars, you own the bank). But, if a big business defaults on its very large bond offering, the impact of the default is likely to be diversified and the losses are more likely to be experienced directly by wealthy investors without taking down financial institutions that serve other useful purposes in our economy with them.

Moreover, the bigger the loan, the weaker arguments about the importance of transaction costs get. You can afford to pay a lot of lawyers, accountants and investigators a lot of money to do due diligence on a deal worth tens or hundreds of billions of dollars. There are bond sale commissions, of course, but those aren't that much different from the point of view of a borrower than the surcharge that homeowners pay for jumbo mortgages.

We had very large business loans (in the form of bond offerings) long before we had very large banks, and somehow the economy managed anyway. In any case, if a very big deal will crater just because it has to pay a sales commission to investment bankers to sell bonds to finance the deal, maybe it isn't a good idea for society as a whole to be taking such a risky gamble. When big investments fail, innocent people are inevitably hurt in a way that the people at fault can't be held accountable for because the money is gone. Systemically insisting on slightly higher anticipated rates of return to allow investments like these to be made isn't necessarily a bad thing for our nation's macroeconomic health.

Even if it is much more efficient to have institutions that are in the business of making very large loans to very large businesses, it isn't at all obvious that this business has that many synergies with other elements of commercial banking. Small institutions specializing in large loans with money those institutions borrow themselves, basically a new generation of investment banks, should pose less risk to our economy than trying this business to commercial banking, particularly if they can't be financed with loans from commercial banks that are unsecured or are secured by financial assets.

The Obama administration's proposed tax on the assets of big banks, for example, set appropriately high, for example, might be a better way to discourage banking concentration than an outright arbitrary market share limitation, by reducing the competitive advantage that big banks might otherwise have over the bond market in this particular high systemic risk part of our economy.