Speaking at the Institute of Regulation & Risk, North Asia (IRRNA), in Hong Kong today, Andrew Haldane - Executive Director for Financial Stability - considers the costs and benefits of structural reform to the banking system and particularly whether regulatory policy should consider the prohibition as well as the taxation of activities.
He begins by considering the costs of the financial crisis that might be attributed to banks. Direct fiscal costs are likely to underestimate the impact on the wider economy of the financial crisis whereas the value of lost output, some of which might be permanent, might significantly overstate the costs attributable to banks. If authorities want to levy a tax on banks to reflect the costs of the crisis, then ". a more precise measure may be needed of banks' distinctive contribution to systemic risk", Andrew Haldane says. He attempts to provide one such measure based on ratings and bond yields to estimate the reduction in banks' funding costs that arises from perceived government support to safeguard stability. The estimates are no more than illustrative but suggest the costs are both large and largely accounted for by institutions that are perceived as 'too-big-to-fail'.
In considering how banks should shoulder their share of systemic costs, Andrew Haldane notes that the focus to date has been largely on the role of prudential regulation in the form of higher capital and liquidity buffers - a taxation solution to systemic risk. A second approach seeks a prohibition solution, involving the separation of bank activities across business lines or geographies.
He discusses the benefits of prohibition for financial system resilience under three headings: modularity, robustness and incentives. On modularity, he notes that a decentralised, modular structure has strengthened resilience in many types of network, and that such structures have often required policy intervention to bring them about. The banking system in contrast is characterised by relatively few large banks conducting diversified but similar activities, resulting in a fragile system lacking overall diversity. On robustness, he suggests that a regulatory regime needs simple rather than complex rules and the focus of policy needs to be the avoidance of extreme outcomes, as is the case in other industries that manage systemic risk. He thinks robustness favours acting on the organisational form of the system rather than through the participants operating within it. He says: "Using these robustness criteria, it is possible to assess whether restrictions might be preferable to taxation.". And he contrasts the experience of the relatively simple Glass-Steagall Act to the complexities of the Basel II regulations. On incentives, Andrew Haldane explains that there are natural incentives within the financial system for banks to explore tail risks and avoid regulatory control. He says: "This dynamic means it is hazardous to believe there is a magic number for regulatory ratios sufficient to insure against tail risk in all states of the world", which has led some economists to propose radical structural redesign such as narrow banking or mutual fund banks.
Turning to the costs of prohibition, Andrew Haldane considers the economies of scale and scope that might be lost as a result of restricting bank functions. He finds little evidence to suggest that larger scale or scope is better in banking. He says, ".limits on the optimal size and scope of firms may be as much neurological as technological. this crisis has provided many examples of failures rooted in an exaggerated sense of knowledge and control. Risks and counterparty relationships outstripped banks' ability to manage them."
In conclusion, Andrew Haldane notes that a large part of the effort of the international community over the past few years has been directed at resolving large banks - for example, through improved resolution regimes and living wills. Such approaches he says, ".could help achieve the modularity, robustness and better aligned incentives which restrictions otherwise deliver. But if this effort is unsuccessful, past evidence and present experience pose a big question about existing banking structures". He stresses that, ".it is possible that no amount of capital or liquidity may ever be quite enough. Profit incentives may place risk one step beyond regulation. That means banking reform may need to look beyond regulation to the underlying structure of finance...".