How to Break Up the Banks - Solving the "too big to fail" problem in the future structure of the global financial system

How to Break Up the Banks. By Adrian Blundell-Wignall
Solving the "too big to fail" problem in the future structure of the global financial system.
The Wall Street Journal, page A13

The financial crisis that sparked the worst recession in decades is in abeyance, but not yet over. Nonperforming loans and other assets of doubtful quality still weigh on many banks. Financial reform to date has focused on improving capital rules and processes. What has not yet been addressed is the future structure of the global financial system.

Contagion risk and counterparty failure have been the main hallmarks of the crisis. While some large diversified banks that focused mainly on commercial banking survived very well, other smaller and less diversified banks, particularly those focused on mortgages, and financial conglomerates that built on investment banking, the structuring of complex derivatives and proprietary trading as the main drivers of growth, suffered crippling losses. In principle, sound corporate governance and a strong risk-management culture should enable banks to avoid excess leverage and risk taking. But human nature being what it is, there are likely always to be some players eager to push complex products and trading beyond the sensible needs of industry and long-term investors in order to drive profits. Indeed, right now such activity is driving the rapid profit growth of some banks, with little having been learned from the past.

As the system will always be hostage to the "gung-ho" few, the question is whether there is a better way to structure large conglomerates in order to isolate commercial banking functions from such high-risk activities. In discussions at the OECD, we have been reviewing possible options. One proposal, which we now submit for consideration, is that banking and financial service groups could be structured under a variant of non-operating holding companies (NOHCs), in all countries.

Under such a structure, the parent would be non-operating, raising capital on the stock exchange and investing it transparently and without any double-gearing in its operating subsidiaries—say a bank and a securities firm that would be separate legal entities with their own governance. The subsidiaries would pay dividends through the parent to shareholders out of profits. The nonoperating parent would have no legal basis to shift capital between affiliates in a crisis, and it would not be able to request "special dividends" in order to do so.

These structures allow separation insofar as prudential risk and the use of capital is concerned without the full divestment required under Glass-Steagall or in response to the recently-expressed concerns of Paul Volcker and Mervyn King—such extreme solutions should remain the proper focus of competition authorities. With an NOHC structure, technology platforms and back office functions would still be shared, permitting synergies and economies of scale and scope. Such a transparent structure would make it easier for regulators and market players to see potential weaknesses. Mark-to-market and fair value accounting would affect those affiliates most associated with securities businesses, while longer-term cost amortization would dominate for commercial banking. It would create a tougher, non-subsidized environment for securities firms, but a safer one for investors.

If a securities firm under this structure had access to limited "siloed" capital and could not share with other subsidiaries, and this were clear to the market, this would be priced into the cost of capital and reflected in margins for derivative transactions. The result would likely be smaller securities firms that are more careful in risk-taking than has been the case under the "double gearing" scenarios seen in mixed or universal bank groups.

Finally, if a securities affiliate were to fail under such a structure, the regulator could shut it down without affecting its commercial banking sister firm in a critical way—obviating the need for "living wills." Resolution mechanisms for smaller, legally separate entities would be more credible than those needed in the recent past for large mixed conglomerates—helping to deal with the "too big to fail" issue. To protect consumers, deposit insurance and other guarantees could apply to the bank without being extended to the legally separate securities firm.

The world is still waiting for a full reassessment of what banks do and how they compete. Until now, the implementation of regulatory standards and accounting rules has been eased. Fiscal policy has supported the economy and interest rates are being kept low to support the underlying earnings of banks and their ability to issue new equity in rising markets. This strategy may work in the short term. But it can't go on forever. Sooner or later we will have to exit from the extraordinary measures that have used trillions of taxpayer dollars to save the institutions that took the world economy to the brink of another Great Depression.

The structure of organizations and how they compete will be critical to future stability. Going forward, the aim must be to keep the "credit culture" and the "equity culture" separate so that government implicit and explicit insurance does not extend to cross-subsidizing high-risk market activity, and so that contagion and counterparty risk can be reduced. The right balance must also be struck between sufficient size conducive to diversification and strong competition to meet consumer needs at reasonable costs.

The capital and derivative markets are inherently interconnected globally, so counterparty risk looms large. Under present structures, if one participant fails, everyone is in trouble. We can't let the world go through that turmoil again.

Mr. Blundell-Wignall is deputy director of financial and enterprise affairs at the OECD.