The Central Problem With Banks Is “Too Complex To Fail” Not “Too Big To Fail”

John Kay

Poor Bernie Sanders. How can you expect to become US president if you are not familiar with the relative spheres of competence of the Federal Reserve and Treasury department in the supervision of the nation’s banks? If you are not au fait with the different roles of the Securities and Exchange Commission, the Federal Deposit Insurance Corporation and the Office of the Comptroller of the Currency?

The senator from Vermont was part of the Congress that passed the Dodd-Frank Act extending financial regulation. Yet he has not even mastered the thousand pages or so of the act, far less the regulations and explanatory documents that have been published since.

Mr Sanders’ recent stumbles illustrate a misdirection in his attack on the banking establishment. The central problem is not so much “too big to fail” but “too complex to fail”: Lehman was a systemically important financial institution but not an important financial institution. Nor was it a big one; it had fewer employees than Citigroup today has compliance staff. Lehman’s collapse created major problems for the global financial system because of the extent of its interactions, with more than 1m outstanding contracts at the time of its bankruptcy. Similarly, Long Term Capital Management was insignificant in size when it failed but capable of massive impact by virtue of the exposure of other institutions to its activities.

There is some force in the claim that size in banking has actually been conducive to stability. Britain had no banking crisis in the Great Depression because the sector was highly concentrated. The US had many failures because the fragmentation imposed by restrictions on interstate banking meant that many banks lacked sufficient geographical or sectoral diversification to weather losses.

Ahead of the global financial crisis , it was argued that the growth of securitisation and other complex instruments similarly contributed to financial resilience. The reverse proved to be the case; trade between institutions represented concentration and multiplication of risks rather than diversification.

Complexity is the enemy of stability. Financial conglomerates have become too diverse and sprawling for their chief executives or boards to understand what they do. The same complexity creates endemic conflicts of interest and is associated with cross subsidy between activities. There are fundamental differences in the cultures required to trade derivatives, to give private financial advice to big corporations, to manage assets on behalf of savers and to provide an efficient retail banking service.

And these conflicts and interdependencies undermine resilience. Vertical chains of intermediation, which channel funds directly from savers to the uses of capital, can break without inflicting much collateral damage. When intermediation is predominantly horizontal, with intermediaries mostly trading with each other, any failure cascades through the system, as happened with Lehman. Today the assets of major financial institutions are predominantly the liabilities of other financial institutions; and vice versa.

These issues are compounded by the regulatory complexity that follows from attempts to monitor behaviour in impossible detail. As the size of the Dodd-Frank legislation shows, we have locked ourselves into a spiral in which regulatory complexity gives rise to further organisational complexity and the construction of yet more esoteric instruments. Even if legislators had better motives than the present corrupting structure that US campaign finance seems to allow, they cannot hope to have more than a basic knowledge of the rules they promulgate or the workings of the regulatory institutions they have created.

So should we break up banks? Bring it on, Bernie.

This column was first published by the Financial Times and on John Kay’s Blog.