A strong and competitive financial sector is essential to a productive modern economy. But when banks fail the costs are high. Following the failure of Lehman Brothers in September 2008, governments around the world acted decisively to protect retail depositors, maintain financial stability, and enable banks to continue lending. Without such action the consequences of the financial crisis would have been far worse – but the result was a significant burden on taxpayers.
To reduce the probability of a repetition of these events, G20 finance ministers and leaders have committed to implementing higher regulatory and supervisory standards to guard against excessive risks. The objective of these reforms is to ensure that the costs of failure of institutions are borne by shareholders and other creditors in an orderly way without triggering a systemic crisis. This should limit the circumstances in which any future government intervention is necessary. But in practice there may always be a risk that the potential costs to the wider economy of a systemic crisis will be sufficiently great that government intervention is appropriate. After all, systemic financial crises have been an intermittent but pervasive feature of financial systems throughout recorded economic history.
A key question is therefore how to ensure that any costs of government intervention are distributed more fairly across those in the financial sector who benefit, reflecting the risks and rewards associated with financial intermediation. This should mean ensuring that the costs of bank failure fall primarily to banks and bank investors, rather than taxpayers. A paper published by the UK government last week (HM Treasury 2009) considers in detail options for ensuring banks meet the immediate costs of interventions to prevent systemic failure. The provision of emergency liquidity facilities and of deposit insurance is already well established, so this article concentrates on the potential new proposals in the paper: contingent capital and systemic risk levies. It also discusses the case for additional taxation of the financial sector to ensure that the sector makes a fair contribution to society and broader social objectives.
Contingent capital and capital insurance
The Basel Committee is considering how to reform global capital standards and banks are likely to have to significantly strengthen the quantity and quality of capital that they hold. In the recent crisis existing subordinated debt and hybrid capital largely failed in its original objective of bearing losses, so the focus of regulatory capital requirements should be on capital that can absorb losses on a going concern basis, typically common equity. More equity will enhance the resilience of banks to shocks. However, if equity is insufficient to absorb losses, banks may have to try to raise more. And in a systemic crisis the cost of raising new capital could be prohibitively high, so that banks may find it difficult to raise new capital when they need it most.
Contingent capital or capital insurance held by the private sector could help address this potential issue and supplement common equity in times of crisis. There are a variety of proposals (e.g. Raviv 2004, Flannery 2009) under which banks would issue fixed income debt that would convert into capital according to a predetermined mechanism, either bank-specific (related to levels of regulatory capital) or a more general measure of crisis. Alternatively, under capital insurance, an insurer would receive a premium for agreeing to provide an amount of capital to the bank in case of systemic crisis.
From an economic perspective, the attraction of contingent capital or insurance is clear; creating such instruments, especially if they were traded, would improve both market discipline and market information. Unfortunately, however, as we have seen, it is precisely in a crisis that markets for such instruments – which will essentially be put options sold by investors to banks – can fail. Such instruments may ultimately not be appropriate for many fixed-income investors such as insurance funds which have in the past invested in subordinated debt and hybrid capital instruments; and the systemic stability consequences would need to be explored.
Alternatively, the Government could offer a capital insurance scheme. For example, Caballero and Kurlat (2009) propose that the central bank should issue tradable insurance credits, which would allow holders to attach a central bank guarantee to their assets during a systemic crisis; or the government could set out an explicit arrangement to deliver capital and liquidity to banks in times of a systemic crisis in return for an up-front fee, as proposed by Perotti and Suarez (2009). Trigger events, fees and the equity conversion price would all be set in advance.
Such a scheme would provide greater certainty for market participants about the circumstances and limits of government intervention, while also ensuring government receives an up-front fee for implicit support. However, there would be potential for moral hazard resulting from reduced loss-sharing across subordinated debt and hybrid capital, unless such capital converted to equity ahead of any government injection. In general government-provided capital insurance would seem to be less useful and less flexible than a wider systemic levy or resolution fund, an option outlined below, although capital insurance could be an element of any wider solution.
Systemic risk levies and resolution funds
Since it is impossible to eliminate entirely the risk that some of the costs of intervention will have to provided by government, there is a case for ensuring that in future the financial sector itself meets these residual costs through a systemic risk levy or resolution fund. A “systemic risk scheme” would have a wider scope than existing deposit-guarantee schemes in that contributions would be sought from all systemically important firms rather than just retail deposit-takers; and wider coverage in that it would fund the costs of restoring financial stability more broadly rather than just the costs of compensating retail depositors. Funds could be used towards the recapitalisation of failing or failed banks, to cover the cost of other guarantees to creditors, or potentially other costs from resolution.
Such a levy might be charged on either a pre- or post-funded basis. It could be weighted towards financial firms which, because of their size or the nature of their business, were a potential source of significant risk to the stability of the financial system. The levy would not fund insurance for any individual firm but rather would support intervention, if needed, to stabilise the system as a whole, hence avoiding some of the moral hazard problems typical of insurance schemes. Given that financial crises requiring direct government intervention are relatively rare in developed economies, any fund would have to be built up over an extended period. When financial crises do occur, however, they can be exceptionally costly to the public finances, and therefore the fund may ultimately need to be quite large.
A key consideration in designing any systemic levy or general resolution fund would be how to assess systemic significance and/or a firm’s contribution to systemic risk, and how to reflect that in any levy. In principle, a well-designed levy could achieve two objectives. It would both raise funds to cover the costs of restoring financial stability and, to the extent that the levy accurately reflected both systemic risk and institution-specific risk and charged for individual institutions’ contribution to those risks, it would embed incentives that would reduce the probability that such costs would ever materialise. For these reasons, a prefunded, rather than a “survivor pays” approach appears more attractive in principle.
Ensuring the financial sector makes a fair contribution
Going beyond the measures described above to deal with systemic risk, there might be other reasons why the financial sector should make a greater fiscal contribution:
- to defray the wider fiscal, economic and social costs of the recent crisis;
- to help correct what might be considered excessively risky or destabilising activities that may have negative externalities;
- if elements of the financial services industry were shown to be generating supernormal returns to executives or shareholders – economic rents – because of the existence of market failures, then there may be a case for increasing taxation on these returns;
- the global nature of the financial services industry and the mobility of its activities might suggest that a more internationally coordinated approach would help ensure the sector makes a fair contribution through tax, irrespective of where firms are located or where the activity takes place.
A financial transaction tax has been suggested as a potential method of ensuring that the global financial services sector makes a fair contribution. It has been argued that some financial transactions have little or even negative social value (see e.g. Krugman (2009), Turner (2009)); that even if such a tax reduces liquidity in some markets, there are likely to be diminishing marginal returns to liquidity, and so any negative impact would be minimal; and that the potential revenues could be large. Full analysis of the potential economic implications of introducing a transaction tax will help determine the desirability of such a tax, the level at which it should be set if it were introduced and the likely consequences.
As well as considering the economic impact of a financial transaction tax, there are some significant issues to explore regarding its design and implementation. Key points include:
- identifying the tax base: to protect against avoidance a financial transactions tax would ideally have as broad a base as possible, including over-the counter transactions;
- establishing a means of tracking transactions in order to implement the tax, given financial transactions are currently recorded through a range of exchanges and other systems;
- setting a rate, or rates, to ensure the introduction of a financial transaction tax does not have a negative economic impact, given the different margins on particular types of transactions;
- determining a means for allocating revenue raised, given the international nature of many financial transactions; and
- defining a method for monitoring and ensuring compliance, and determining action that should be taken in the event of avoidance or evasion.
Common principles and next steps
The IMF will report in April to the G20 on these issues. Any proposals should respect the following principles:
- Global – some options could realistically only be implemented at a global level, while others would require international agreement and coordination on key principles to be effective;
- Non-distortionary – avoiding measures that would damage liquidity, drive inefficient allocation of capital or lead to widespread avoidance;
- Stability enhancing – actions must support and not undermine the regulatory action already being taken. This is likely to mean any option would take several years to implement; and
- Fair and measured – financial services must be able to continue to contribute to economic growth and any additional costs should be distributed fairly across the sector. A thorough impact assessment must be conducted prior to implementation.
Caballero, Ricardo J. and Pablo Kurlat (2009) The “Surprising” Origin and Nature of Financial Crises: A Macroeconomic Policy Proposal, Federal Reserve Symposium at Jackson Hole, August
Flannery, Mark (2009) Contingent Tools Can Fill Capital Gaps, American Banker, Vol. 174 Issue 117.
HM Treasury (2009) Risk, Reward and Responsibility: The Financial Sector and Society, December
Krugman, Paul (2009) Taxing the Speculators, New York Times 27 November
Perotti, Enrico and Javier Suarez (2009) Liquidity Insurance for Systemic Crises, VoxEU.org, 11 February
Raviv, Alon (2004) Bank Stability and Market Discipline: Debt-for-Equity Swap versus Subordinated Notes. Unpublished working paper
Turner, Adair (2009) Responding to the financial crisis: challenging assumptions, speech to the British Embassy, Paris 30 November