In a discussion paper published today, the Bank contributes to emerging ideas on how macroprudential instruments might be designed and deployed to help to restrain the build-up of risks within the financial system. The aim of doing so would be to make the financial system more resilient and the real economy more stable.
Reducing risks to the financial system as a whole - systemic risk - has emerged as a public policy priority. The international debate is addressing three key, related elements: the structure of the financial system, the regulatory framework and the resolution framework. Working with UK Tripartite and international colleagues, the Bank is contributing to each dimension of this debate*.
In relation to the regulatory framework, a key challenge is to achieve a re-orientation towards systemic risk. This is where macroprudential policy would fit in. Macroprudential policy is a missing ingredient from the current policy framework. That implies too great a gap between macroeconomic policy and the regulation of individual financial institutions.
Financial stability is fundamentally concerned with maintaining a stable provision of financial services to the wider economy - payments services, credit supply, and insurance against risk. This is the starting point for any macroprudential policy instruments. The practical question is whether such instruments can be made operational. It is possible to conceive of more ambitious objectives, such as forestalling asset price bubbles. By moderating exuberant increases in the supply of credit, macroprudential policy might sometimes help to contain asset bubbles. But it would be unrealistic to make the prevention of asset bubbles a specific objective of the regulation of the banking system.
The Bank's discussion paper identifies two principal sources of systemic risk that macroprudential policy would ideally aim to address. First, the tendency for the banking system to become overly exposed to risk in the upswing of a credit cycle and to become overly risk-averse in a downswing - 'aggregate risk'; and second, the tendency for individual firms to take insufficient account of the spillover effects of their actions on risk in the rest of the financial network - 'network risk'.
On the first, the paper examines whether it would be practical to dampen cyclical over-exuberance through a regime of capital surcharges on top of existing microprudential capital ratios. These surcharges could be applied to headline capital requirements or at a more disaggregated level to different classes of lending and exposure. Increasing capital requirements in a credit boom would generate greater self-insurance for the system as a whole and, at the margin, act as a restraint on overly exuberant lending. These ideas could usefully be debated alongside existing international initiatives to dampen procyclicality in regulatory minimum requirements, and to encourage earlier provisioning against expected losses in loan portfolios.
On the second, as the FSA and others have discussed, capital surcharges could be set across firms so as broadly to reflect their individual contribution to systemic risk, based on factors such as size, connectivity and complexity. This would lower the probability of those institutions failing and so provide some extra systemic insurance. It would also provide incentives for those firms to alter their balance sheet structure to lower the systemic impact of their failure. This agenda too is being debated internationally.
Against that high-level background, the paper discusses some challenges in introducing such a macroprudential regime in practice. It identifies some possible indicators, quantitative and qualitative, that could help judgments on how capital surcharges could be set. They would largely be about the macroeconomy and the financial system as a whole.
The paper suggests that it is unlikely that macroprudential instruments could be set solely according to fixed rules. Judgement might be needed to make robust policy choices. That would call for assessments of the resilience of the system, credit conditions, sectoral indebtedness and systemic spillovers - all of which vary over time and according to circumstances.
At the same time, it would be important for constraints to be placed on any macroprudential regime to ensure transparency, accountability and some predictability - a regime of 'constrained discretion'. That would call for clarity around the objectives of policy, the framework for decision-making, and policy decisions. It also suggests the need for robust accountability mechanisms.
Another important issue would be the degree of international co-operation. To be wholly effective, a macroprudential regime might require international co-ordination. But, even in its absence, appropriate macroprudential instruments might still be able to strengthen the resilience of the domestic banking sector.
The discussion paper does not reach definitive answers, nor does it advocate a particular operational regime. Rather it suggests some possible directions for the international debate in the period ahead on how the authorities might deploy policy instruments to lower the incidence and cost of future systemic crises.
The Bank invites comments on, and criticisms of, the ideas expressed in the paper.
* A series of recent speeches have highlighted the importance of re-evaluating the structure of the financial system, improving the framework for financial crisis management and resolution, and revisiting the regulatory framework - in particular, the potential role for macroprudential instruments.
Notes to Editors
1. Recent speeches from the Bank have covered various aspects of the debate on structure, regulation and resolution. These are available on the Bank's website:
2. A speech by Paul Tucker, Deputy Governor for Financial Stability, on macroprudential instruments was published on 22 October 2009
The Debate on Financial System Resilience: Macroprudential Instruments (108k)
3. The Bank's June 2009 Financial Stability Report included discussion of the need for macroprudential policy instruments.