We've discussed the issue of bank regulation on this blog from time to time.  With a handful of banks holding the lions share of the assets in the financial system, the problem of too big to fail remains in place.  Two broad solutions exist.  One is to downsize the large financial institutions so that they are not too big to fail.  The approach is complex.  No one really knows what the optimal size of a bank should be.  Moreover, there is no particular agreement on the method.

The other is to impose additional prudential requirements designed to reduce the risk of failure.  The banks will still be too big to fail but will be less likely to fail.  The Volker Rule is an example of regulation in this category.

Daniel Tarullo, a government of the Federal Reserve Board, has weighed in on this issue.  He has called for limits on size (the speech is here).  Specifically, he has called for a limit on total non-deposit liabilities.  As he proposed:

The idea along these lines that seems to have the most promise would limit the non-deposit liabilities of financial firms to a specified percentage of U.S. gross domestic product, as calculated on a lagged, averaged basis. In addition to the virtue of simplicity, this approach has the advantage of tying the limitation on growth of financial firms to the growth of the national economy and its capacity to absorb losses, as well as to the extent of a firm's dependence on funding from sources other than the stable base of deposits.

In other words, banks would incur a limitation on their non-deposit sources of funding, effectively limiting their size.  According to the WSJ, Tarullo "is the highest ranking regulatory official to call for limiting the size of banks."

The approach raises plenty of questions, as Tarullo himself noted.

Of course, the difficult question would be the applicable percentage of GDP. The answer would depend on a judgment as to how much of an impact the economy could absorb. It would also entail a judgment as to how large and complex a firm needs to be in order to achieve significant economies of scale and scope that carry social benefit. Depending on the answers to these questions, there may be a need to balance the relevant costs and benefits. There would also be important secondary questions such as whether to exclude from a firm's calculated liabilities only insured deposits and which asset base to use in calculating non-deposit liabilities.

Whether these sorts of reforms will ever occur remains to be seen.  But it reflects a growing sentiment that the problem of too big to fail was not solved by the reforms in Dodd-Frank.