I understand the arguments for not breaking up or nationalizing the largest banks, both in general and in the middle of a crisis. Those arguments are predicated on it being a bad idea, either too difficult or legally impossible. One talking point that I have not heard is that large banks are good for the economy. This is lobbyist talk, but let’s hear it anyway. From Simon Johnson:
Ms. FARRELL: We have created them [our biggest banks], and we’re sort of past that point, and I think that in some sense, the genie’s out of the bottle and what we need to do is to manage them and to oversee them, as opposed to hark back to a time that we’re unlikely to ever come back to or want to come back to.”
That’s not an argument. Under what circumstances would our biggest banks be the best arrangement of the banking sector? It’s important to realize how large the largest banks are. Here is a list of the Top 50 Bank Holding Companies from the Fed. I’m going to give you two charts. One is a histogram of those top 50:
Since this is in thousands, e^9 represents trillions above. Notice how fast the dropoff takes place. The other is the top 20, with an additional column for asset size as a multiple of the size of the 20th bank:
One potential argument is that something about the financial sector requires natural monopolies – clear leaders who can overcome informational and trust problems inherit in financial transactions, who can attract the top talent and thus allow the biggest players an extra level of certainty that they are dealing with the best. Natural monopolies is common thought for sectors with large fixed costs and low-to-zero marginal costs and increasing returns to scale, something not true of banking.
But take a look at that ratio column. We could take the top 4 banks, break them in half, and the halfs would clock in at….1st, 2nd, 3rd and 6th. See how far out the top players are in the histogram? Cutting the size of the largest banks would still likely leave them in the front of the pack, perhaps even by a distance. And it’s why I’m skeptical of marginal changes (higher capital requirements, for instance) as drivers of shrinking the size of the largest banks, since they would have so far to shrink down.
Another argument is that the scale at which the financial sector moves at these days requires large financial leaders. James Kwak grabs an obvious example, a large J&J bond issuance of $1.6bn, and shows that it splits up fine among several banks. The managerial responsibility of bond issuers is executed with no problem at mid-sized banks; I don’t hear complaints otherwise.
Another argument is that there is some sort of risk management scaling to size. The larger a bank gets, the better it is at diversifying, attracting risk-management talent, keeping losses down. Here’s a graph of expected loss, a measure of how poorly a bank has done, by asset size, taken from the Stress Test (so reported by the bank’s own models):
Notice the lack of variance among the big players there – they’ve all taken a big hit in this crisis. So I don’t think there’s a benefit to stability of the institutions themselves.
(Does anyone know how to get access to Bank Holding Documents in large data files? I want to aggregate and test this at a larger scale. Email or comment if you do.)
How about consumers? I’ve seen no evidence that larger banks help consumers, and indeed I’ve seen evidence that as bank consolidation happens, large banks ‘respond asymmetrically’ to interest rate changes, so they raise rates with the Fed faster, and lower rates when the Fed lowers them slowers, milking consumers. That’s exactly what you’d expect someone with market power to do.
In the next week I’ll see if I can get some historical data in here, but comparing the size, both absolute and relative, of the largest bank-holding companies now against the period from 1980-1999, it doesn’t strike me at all that the mergers and consolidation that has taken place has lead to better growth for the country.