We now know that sufficiently large financial institutions enjoy implicit insurance from the taxpayers, because they’re “too big to fail”: i.e., so big that letting them fail risks cascading failure, as letting Lehman fail turned out to do.
The value of that insurance depends on the size of the institution and the risks it takes.
So charging all large financial institutions, whether called “banks” or not, an “insurance premium” based on size and risk on a seems both equitable and efficient. (Efficient, because offering that “insurance” for free gives institutions an incentive to grow too large and take too many risks.) Neither size nor risk can be measured without error, but I don’t doubt that some of the folks at the Treasury are clever enough to work out something adequately accurate.
In addition to equity and efficiency, of course, such a tax would raise revenue, which the Federal government could certainly use more of.
Oh, yeah: and proposing such a big-risky-bank tax would be overwhelmingly popular. I’m disappointed that the Administration still seems disinclined to tax transactions as well, and if the new tax is designed to yield only $120 billion total I’ll be disappointed again, but this looks like a good start on helping the Administration regain the political initiative, especially because Republicans will be forced to oppose it unanimously.