Too Big to Fail

The call to break up the large banks has enjoyed considerable popularity in the United States.^[253] A salient ongoing issue is the extent to which a too big to fail problem exists and, if it does, whether the combination of Basel III and the post-Great Recession reforms have fixed it.

The principal response of regulatory reform to TBTF has been to impose additional capital requirements on the largest banks (an extra 2.5 percent of risk-weighted assets in required equity and an additional 9 percent in non­equity TLAC for G-SIBs). In the United States, the Dodd-Frank Act sought to end TBTF by providing for Orderly Liquidation Authority to ensure that large banks can be smoothly shut down. The Federal Deposit Insurance Corporation (FDIC) subsequently set forth the corresponding imple­menting mechanism, the Single Point of Entry resolution of the holding company while keeping its viable subsidiaries operating.

As discussed in chapter 5, much of the analytical framework on this issue has focused on an implicit TBTF subsidy to risk taking, which some empirical studies demonstrate. However, empirical evidence seems to be emerging that since the crisis the implicit subsidy has disappeared or even turned negative, as investors consider the additional regulatory burdens on the largest banks and (perhaps) ponder the new orderly resolution arrange­ments (which in the case of the European banking union include manda­tory bail-ins of creditors). Moreover, data for the United States show that the largest banks did not experience proportionately larger losses than me­dium-sized banks in the Great Recession, which they would have done if they had systematically taken on excessive risk induced by a TBTF subsidy.

Even if the largest banks did not, or no longer, gamble with anticipated taxpayer money, dominance of the sector by a few large banks would involve an inherent potential external diseconomy in the form of the systemic spill­over from the collapse of one or more of them.

The results of research on banking economies of scale remain ambig­uous. Some recent analyses show decisively important returns to scale. One study purporting to show no returns to scale when using social pricing is contradicted by another study showing the persistence of economies of scale even pricing at smaller-bank funding costs. Methodological challenges include the problem of biased estimation if a single parametric output function (such as the translog) is applied across the full size range and the problem of ambiguity in measuring what is the “output” and what are the “inputs” in the production function. Nor is there a readily accessible litera­ture setting forth, in case studies or otherwise, analysis of key functions that can only be accomplished by large banks.

At the same time, it is taken for granted that in many other sectors (au­tomobiles, aircraft, pharmaceuticals), large size is essential to efficiency. It

is by no means evident that the prior assumption should be that in banking there are no important economies of scale. Nor is it clear that the prior as­sumption should be that there are no economies of scale in today's economy that warrant the relatively much larger sizes of the biggest banks in compar­ison to their sizes two decades ago.

In the United States, observers who worry about large bank size per se might take some comfort from the fact that in recent years the size of the largest US banks has shrunk from the peak level reached in 2007 and that their assets are considerably smaller relative to GDP than in Europe, Canada, Australia, and Japan (see chapter 5). The assets of the 10 largest US banks fell from 72 percent of GDP in 2007 to 62 percent in 2015, and the assets of the top five US banks account for 50 percent of GDP in the United States versus about 200 percent in Canada and Australia and about 125 percent in Japan and Germany (chapter 5).

The most direct implication of this study for this issue, however, is that the possible TBTF problems nevertheless underscore the importance of en­suring that bank capital is adequate to minimize the risk of financial crises. Achieving the right balance between economies of scale and negative exter­nalities arising from large scale is best ensured by making certain that the likelihood of crises is reduced through stronger capitalization. The impor­tance of adequate capitalization is further reinforced if one is skeptical about the plausibility of “immaculate bankruptcy” under the Orderly Liquidation Authority. It seems optimistic to presume that business as usual would con­tinue at the major subsidiaries of a failed large bank holding company in resolution—the central premise of Single Point of Entry.