Afterward, I found myself with a lot of negative feelings.
The first act of the (short) tale can be found here. Here is one bit:
I asked Baily, as an expert on productivity statistics, whether he thought that any economist would claim to have a reliable measure of bank output. “Of course not,” he replied, nearly breaking into a laugh. I was glad to hear that response, because it reinforced my view that econometric estimates of scale economies in banking are not reliable. When you measure economies of scale, you are comparing the ratio of output to inputs at different-sized firms. It’s rather difficult to do that if you cannot measure the numerator.
The substantive upshot was this:
The panel was called to discuss a research paper commissioned for the Center and written by Martin Baily, Doug Elliot, and Phillip Swagel. The paper says that (a) we have little reason to worry about too-big-to-fail, because the FDIC is on its way to having enough authority to resolve big bank failures, (b) there are economies of scale in banking at the very highest levels, (c) there are transition costs to breaking up big banks, in that employees and customers would be left hanging waiting to see how the re-org falls out, and (d) breaking up big banks would not get rid of systemic risk, anyway.
I agree entirely with (d). I thought that (c) was a fair point, but there is such a thing in the corporate world as a spin-off, and it can be done. I disagreed with (a) and I was unpersuaded by (b).
Recommended, and here is more from Arnold on the same topic.