The largest U.S. banks and their lobbyists have been trying hard to counteract the growing impression that they present an unacceptable threat to the economy. In a new series of papers, the Federal Reserve Bank of New York offers some evidence that they probably won't like.
Critics of the big banks assert that repeated government bailouts have created a perverse incentive: The bigger and more systemically important a bank becomes, the more certain its creditors can be that they will get rescued in an emergency. Such too-big-to-fail status, the logic goes, allows banks to borrow more cheaply than they otherwise would -- a taxpayer subsidy that encourages them to take the kind of risks that lead to disasters.
The contributions from the New York Fed economists support this narrative in two ways: They place a value on the big banks' funding advantage, and they suggest that banks with government support do tend to take bigger risks.
Looking at new bond issues from 1985 to 2009, New York Fed Vice President Joao Santos estimates that the largest banks paid 0.31 percentage point less when they issue A-rated debt than when their smaller peers did (after controlling for some idiosyncratic features of the bonds). He also found that the discount was significantly greater than what large companies enjoyed in other industries, undermining bank lobbyists' argument that the funding advantage stems from the general benefits of size rather than from their too-big-to-fail status.
The study has some important blind spots. Its bond data go only through 2009, so it can't show what, if anything, the 2010 Dodd-Frank financial-reform law may have done to reduce the taxpayer subsidy. Also, when grouping bonds by rating category, it ignores the rating uplift the largest banks receive thanks to the expectation of government support. This would have the effect of underestimating their funding advantage -- a problem common among studies of the too-big-to-fail subsidy.
Perhaps more interesting is a separate Capital Requirementsin which Santos and two co-authors find that banks make more bad loans when they have a reasonable expectation of getting bailed out in a crisis. Specifically, the authors look at a special credit rating, issued by Fitch, that assesses the likelihood of government support. A one-notch increase in the rating -- say, from A-minus to A -- is associated with an 8 percent increase in a bank's ratio of non-performing loans.
It wasn't all bad for the big banks. A group of three New York Fed researchers found one economically desirable advantage of size: Operating costs tend to fall as a percentage of assets when banks get bigger. This stands to reason. It shouldn't take twice as many people, computers and buildings to run a bank with $2 trillion in assets as it does to run a bank with $1 trillion in assets. What the study doesn't address are the difficulties of managing larger and more complex organizations -- difficulties that, as various market-rigging scandals and JPMorgan Chase & Co.'s London Whale fiasco have eloquently demonstrated, can outweigh economies of scale.
This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.
To contact the author on this story:
Mark Whitehouse at email@example.com
To contact the editor on this story:
Paula Dwyer at firstname.lastname@example.org