Chapter 4: The economic rationale for explicit financial guarantees
Approaches to deal with financial failure
4.1 As discussed previously, the Financial System Inquiry (FSI) re-affirmed that prudential regulation is intended mainly to prevent disruptive failure; it is not designed to eliminate the consequences of risk-taking from the financial system.
4.2 This perspective on prudential regulation was, and is, broadly appropriate for the Australian financial system but is not necessarily inconsistent with the introduction of a limited explicit guarantee. Indeed, there is a delicate tension between protecting customers thought unable to assess counterparty risk, whilst requiring they bear their share of losses when the system fails to deliver protection.
4.3 Recent events suggest there may now be stronger arguments for explicitly protecting some individuals against losses on a narrow class of retail financial products. Specifically:
- The collapse of the HIH Group of Companies (HIH) together with previous episodes of government intervention in financial institution failures, led governments to respond to public concern by supporting some of the affected policyholders. This suggests that Australians expect and demand financial security on at least some financial products.
- International practice of formalising guarantee arrangements has developed, and Australia’s methods for protecting deposits with banks and other authorised deposit-taking institutions (ADIs), in particular, have become somewhat anomalous. An increasing number of countries are also providing greater protection for some non-deposit financial products, such as insurance and pension plans.
- Australian consumers’ engagement with the financial system continues to deepen as a result of explicit government policies, demographic trends and technological advances. Some common financial products are a prerequisite or ‘critical’ for participation in the modern economy.
4.4 The implications of these developments need to be weighed carefully. On the one hand, the creation of the policyholders’ support scheme for HIH and government intervention in other failures indicates that there is a perceived need to support those customers most exposed to financial institution failures. But, on the other hand, too comprehensive a system of support would constrain investors’ ability to take risks in pursuit of profit, and it would greatly undermine the efficiency of financial markets.
4.5 Australian governments historically have responded to infrequent financial institution failures by providing compensation to the most vulnerable customers funded from the general tax base.1 It can be argued that this risk is not recognised appropriately and therefore not appropriately priced in financial products. Moreover, the costs of failure are not necessarily being borne by the beneficiaries or by consumers of the types of products in question.
4.6 Financial institution failures will occur from time to time in any efficiently regulated financial system. When failures occur, particularly among prudentially regulated institutions, there is generally pressure on governments to underwrite at least some of the financial promises made by some types of failed institutions regardless of whether there was any prior commitment to do so.
4.7 In principle, there are several ways to deal with the risk and resulting problems of financial institution failures:
- Adopt a caveat emptor approach, denying responsibility for providing
any compensation for losses due to financial institution failure.2
Caveat emptor relies upon market discipline working effectively to
moderate the behaviour of riskier financial institutions even though there
may be information asymmetry problems inherent in the financial sector. Relying
on such a policy could lead to price and behavioural adjustments which might
deliver the most efficient financial resource allocation outcomes. The success
of such a policy stance would depend upon governments maintaining a consistent
- The history of government interventions in Australia, and convergence of international best practice on a different approach, suggests that sustaining a credible caveat emptor policy is problematic. Moreover, there may be legitimate system stability, efficiency, equity and broader socio-economic reasons for governments to choose to intervene to protect at least some classes of consumers.
- In addition, the legal duty of the Australian Prudential Regulation Authority (APRA) to depositors of ADIs as set out in the Banking Act 1959 may in itself create the expectation amongst depositors that they will be protected if an ADI fails. This would reduce the possibility that a caveat emptor policy could apply in Australia.
- Tighter control of the range of products on offer by selected institutions. The range of products that some prudentially regulated institutions offer could be restricted, or some financial products could be fully collateralised with risk-free securities — the so-called narrow banking model (Merton and Bodie, 1993). This would create a class of risk-free financial products for retail investors. Such an approach would be difficult in a modern cross-border financial system already occupied by financial conglomerates.
- More direct government provision of risk-free financial services and products. Governments have some experience in providing these products, as (mainly historical) examples of government-owned banking and insurance arrangements show. There continues to be some public underwriting of certain insurance products, to ensure financial safety and achieve other policy outcomes. It might be possible to extend this to other types of product. However, this would run counter to prevailing views on the appropriate role of government in the financial system and competitive neutrality.
- Alter the relative position of stakeholders under the insolvency framework as it applies to regulated financial institutions. Under the current arrangements, the entitlements of retail consumers in insolvency are not always differentiated from those of other stakeholders, and these vary across types of financial product. Changing priority arrangements to enhance the entitlements of retail investors in failed firms does not provide for certain outcomes but it may reduce the loss suffered following the failure of their financial service provider. Other stakeholders, of course, would be made worse off by such reform, and could be expected to demand changes in contract terms to compensate.
- Commit to respond to financial institution failures on a ‘discretionary’ or case-by-case basis, tailoring assistance to suit the circumstances. This may preserve flexibility but provides a relatively low degree of certainty and may take some time to implement. Some of the uncertainty may be reduced by committing to some pre-determined criteria for providing assistance.
- An ‘explicit’ government provided or mandated guarantee. Such a guarantee defines limits on the losses which individuals could suffer on some financial products. The costs are then borne by taxpayers or the industry where the guarantee applies (and ultimately by the consumers and shareholders). The classic example is deposit insurance — a system in which bank deposits are protected up to a pre-specified limit. These arrangements could be extended to a limited number of other financial products.3
4.8 This Chapter reviews some of the general benefits and costs of explicit guarantee schemes, as required by the Study’s Terms of Reference. It is important to recognise that any assessment of explicit guarantees is complicated because some of the economic implications are ambiguous. The institutional and behavioural consequences of introducing explicit guarantees and the implications for the broader stability of the financial system are heavily dependent, in particular, on their scope and pricing.
4.9 Appropriately designed schemes may improve the efficiency and stability of the financial system in some respects. Conversely, badly designed schemes can embed inequities, create additional deadweight costs, and undermine market discipline and financial stability.
4.10 That said, there are some potential benefits of explicit guarantees that are less equivocal. They can, for example, relieve taxpayers of implicit liabilities, albeit by transferring these liabilities to shareholders and customers of financial service providers. They can also clearly delineate a set of relatively risk-free financial products for consumers who are not well-placed to assess, or who otherwise do not wish to accept, risk.
4.11 The context in which explicit guarantees are being assessed is also important. Much depends, for example, on judgements about community perceptions of government responsibility in the wake of a financial institution’s failure and expectations about the response of government to such perceptions.
An assessment of explicit guarantees
4.12 This Study assesses the general merits of explicit guarantees against the backdrop of the prudential regulatory framework. For example, it appears possible to consider, at least in a qualitative sense, the key implications of introducing a well-designed explicit guarantee, based upon:
- an analysis of the structure, incentives and expectations embedded within Australia’s existing regulatory framework;
- an assessment of known market failures, such as imperfect information, externalities and imperfect competition; and
- some assumptions about the likely behaviour of consumers, financial institutions, regulators, other creditors and governments.
4.13 It is also important to note that the benefits and costs of a guarantee depend on how it is designed, in particular the coverage, funding arrangements and extent to which efficient pricing can be achieved. Critically, the economic impact is also likely to turn on how an explicit guarantee might affect the behaviour of participants in the financial system.
4.14 One’s perspective on the impact of explicit guarantees will depend on the extent to which it is believed that some form of guarantee already exists within the financial system.
- For those who believe that no implicit guarantee exists, introducing a limited explicit guarantee may be viewed as introducing a distortion into the financial system, making it less efficient and potentially less stable.
- For those who believe that an implicit guarantee already exists, a well-designed explicit guarantee may be viewed as reducing the scope of an existing subsidy and improving the integrity of the financial system.
4.15 The international literature suggests that implicit guarantees are common, at least with respect to bank deposits.4 In Australia, the Reserve Bank of Australia (RBA) acknowledged their existence in relation to deposits in the following terms: ‘it is hard to believe that … democratically elected governments will (or should) stand by and watch a large number of citizens (and voters) lose money they thought was relatively safe’ (RBA 1997). Moreover, the World Bank classifies countries without explicit deposit insurance schemes as having implicit schemes (Demirguc-Kunt and Sobaci 2000).
Moral hazard and financial system integrity
4.16 The most common concern arising with any form of financial system guarantee is the potential it has to create moral hazard.
4.17 Moral hazard exists when people take risks because they know that someone else is protecting them against a financial loss. This increases the probability of loss, and it is unfair to the provider of the guarantee. Both explicit and implicit guarantees create moral hazard because they can affect the behaviour of owners, managers, customers and regulators of financial institutions, leading to increased risk-taking and risk of failure. (See Box 4.1).
Box 4.1: Examples of potential moral hazard in the financial system
Guarantees can encourage the shareholders of a financial service provider to accept greater risk-taking by the institution in the hope of a higher reward. They know that retail investors will not object and demand compensation for increased risk because they are protected by the guarantee.
A guarantee can also encourage retail investors to target products with the highest promised return, irrespective of the inherent risks. Service providers competing for funds can only satisfy this preference for maximum nominal return by undertaking more risky activities.
Moral hazard also applies when the trade-off between risk and return is not quite so apparent. For example, customers may seek the cheapest general insurance cover available, without considering the risk of the provider.5 Insurers then face competitive pressure to lower the price of insurance and to invest the premium income in a more risky portfolio. Either way, a moral hazard exists if there is a guarantee attached to the policy.
If the original value of a market-linked investment is guaranteed, there is a strong incentive for the provider or investor to invest in the riskiest assets in the hope of maximising the potential upside, knowing that their losses will be covered if the strategy fails.
Managers of a financial institution whose remuneration is linked to growth and profitability may be more inclined to undertake risky strategies if not subject to the restraint imposed by customers demanding compensation for increased risk.
Regulators who are not subject to appropriate incentives and accountability arrangements may be more inclined to indulge in forbearance towards troubled institutions in the knowledge that (some) consumers are protected if the institution is unable to recover.
4.18 Moral hazard concerns potentially exist for both implicit and explicit guarantees. However, the design features of an explicit arrangement can substantially ameliorate the problems. In this sense, the moral hazard might be successfully contained to parts of the system where moral hazard may matter less — for example, across a limited range of low risk products and only for relatively unsophisticated customers. Explicit guarantees could also provide a more sustainable basis upon which governments could establish a caveat emptor policy in regard to consumers of non-guaranteed products or liabilities; a strategy to minimise moral hazard.
4.19 In the Australian context, one argument in favour of explicit guarantees over financial products may be that the better-designed schemes are more efficient than any prevailing implicit guarantees. Guarantees of any kind can impair the efficiency of the financial system. But a well-designed explicit guarantee can make the problems less likely and less serious. In a system where implicit guarantees currently exist, introduction of industry-funded explicit guarantees returns the burden of risk to the financial system. This may not only improve the financial position of taxpayers; it could also enhance the efficiency of the financial system.
4.20 Implicit guarantees are considered to be especially strong for depository instruments because the banks which offer them are often large, highly leveraged, and crucial to both the payments system and the wider economy; that is, the ‘Too Big to Fail’ doctrine (Garcia 1996). Although this internationally popular theory emphasising systemic concerns remains plausible, the guarantees associated with HIH, a general insurer, suggests that the size of the firm, the size of consumer losses and/or the fact that an institution is prudentially supervised may be more defining factors.
4.21 Explicit guarantees do not automatically eliminate moral hazard existing under a system of implicit guarantees. If they strengthen and clarify protection, they may discourage investor awareness and monitoring of risk. If they protect well-informed stakeholders who are capable of assessing and monitoring institutional risks, they can intensify moral hazard and weaken market discipline on financial institutions.
4.22 Indeed, a potential benefit in not having an explicit guarantee is that the coverage and extent of compensation is uncertain and stakeholders may be more cautious about where they invest their money. Potentially this will reward better-managed and more prudent financial institutions. An explicit guarantee could reduce this reward for quality.
4.23 Moral hazard can be exacerbated when there is an explicit guarantee for products which are inherently risky or when the price of the guarantee does not reflect the risk of the institution.
4.24 Bohn and Hall (1997) explore the possible moral hazard implications associated with insurance guarantee funds in the United States (US). Because of the time lag between collecting premiums and paying out on policies, they argue that insurers are effectively borrowing money from policyholders. Therefore, the existence of guarantee funds might allow riskier insurers to write policies for (borrow money from) policyholders at rates that do not sufficiently reflect their default risk. That is, they might compete vigorously on the basis of price, cover and service rather than the quality of their promise.
4.25 In some schemes, the possibility of moral hazard raises important issues of fairness. If the explicit guarantee is funded by industry and it is improperly priced, then it could lead to well-managed firms paying unfairly for the risk borne by other service providers. This is an impediment to competitive neutrality and may create incentives to undertake excessive risk in the pursuit of return.
4.26 In extreme cases, moral hazard can increase aggregate risks in the financial system. In particular, it can encourage providers of capital to finance risky projects which would not otherwise be eligible for credit on the same terms.
4.27 An increasing number of countries tackle these problems by pricing the explicit guarantee according to the riskiness of the service provider. In theory, this should solve the moral hazard problem and stop inequitable transfers between firms. But in practice, risk-sensitive pricing of guarantees remains difficult, and the inevitable pricing errors may be unreasonably expensive for some institutions and cause moral hazard in others.
4.28 Several other features of prudential regulation arrangements can work to mitigate moral hazard. Requirements that ownership of financial institutions is well-diversified limit the ability of any group of owners to induce greater risk-taking by the institution in response to guarantee schemes. Also relevant is the existence of significant minimum capital requirements which ensure that owners incur a substantial loss if increased risk-taking leads to adverse outcomes. Prudential oversight of governance arrangements and risk-taking can also serve to constrain any managerial incentives towards excessive risk-taking. Similarly, regulatory sanctions on managers who have previously been responsible for failure (and the role of ‘fit and proper’ tests) are important in this regard.
Consumer protection/monitoring costs
4.29 Explicit guarantees for retail consumers of financial products might generate a more even level of protection.
4.30 The potential costs involved in continuously monitoring the health of any institution are very high for retail customers, relative to the extent of their exposures. This information asymmetry problem is one of the reasons why governments choose to prudentially regulate certain financial institutions. However, since prudential regulation is not intended to prevent all failures, a limited explicit guarantee could enhance welfare by removing, or substantially reducing, the need for protected consumers to incur their own monitoring costs. It would ensure a limited supply of risk-free or lower-risk financial assets.6
4.31 At the same time, the health of the system depends on sophisticated investors utilising their superior capacity to assess and price risk. A limited explicit guarantee may encourage non-guaranteed stakeholders to undertake more rigorous monitoring and risk assessment than if broader implicit guarantees are thought to exist.
4.32 Explicit guarantees would lead to greater consistency and certainty in the degree of protection which eligible consumers would receive from financial losses across the different financial sectors. As noted in Chapter 3, the depositor preference provisions of the Banking Act require that the assets of a failed ADI must be applied first to meeting deposit holder liabilities. Provisions of the Life Insurance Act 1995 together with those of the Corporations Act 2001 also give a degree of preference to holders of life policies over the assets of the relevant statutory fund. General insurance policyholders generally are treated equally with unsecured creditors in any wind-up, although the specific ranking of policyholders can be affected by the terms of reinsurance arrangements and the application of State and Territory laws.
4.33 Under current legislation the extent of losses faced by consumers also varies according to the degree of insolvency of an institution; that is, the extent to which the value of liabilities exceeds that of assets. An explicit guarantee gives certain compensation irrespective of the shortfall in assets relative to liabilities.
4.34 Among ADIs, for example, credit unions and building societies have less exposure to wholesale borrowing markets than do banks. An important implication of this is that a higher proportion of the liabilities of smaller institutions is covered by the depositor preference provisions of the Banking Act, making these provisions relatively less effective in shifting losses to other stakeholders in the event of a failure (Table 4.1).
Table 4.1: The funding structure of ADIs
ADIs (Jun 03)
Total assets: total liabilities
Tier 1 (risk weighted) capital ratio
Total (risk weighted) capital ratio
Australian assets: Australian deposit liabilities (excl. certificates of deposit (CD))
Deposit liabilities (excl. CDs): liabilities
Non-deposit liabilities: liabilities
Source: Australian Prudential Regulation Authority.
4.35 This does not mean that building societies and credit unions are necessarily more risky — these typically hold larger capital buffers than banks and undertake a different mix of activities — but their depositors would be more vulnerable should their equity capital ever be exhausted.
4.36 More generally, guarantees arguably can also serve to level the playing field in terms of risks borne by large, sophisticated consumers/investors and smaller, retail customers. While both groups benefit from prudential regulation, large investors also have better ability to assess risks, access superior information and have access to protection mechanisms that are not available at reasonable cost to small investors. For example, wholesale investors can more easily reduce their exposure to risk by diversifying their portfolios. Individual investors are less able to divide their wealth, and therefore risks are more concentrated with individual providers. In addition, individual investors are less able to negotiate pricing and terms in order to ensure they are fairly rewarded for risk.
4.37 Wholesale investors can also use certain credit risk transfer products, such as credit default swaps, to buy protection against the risk of failure. At the moment, the markets are not heavily used for this purpose, but the technology is developing and is certainly available. Retail investors, by contrast, do not have access to these products, either because markets do not exist for small exposures or because they do not exist for credit exposures to small institutions. Nor do they have the financial sophistication necessary to effectively use such products.
4.38 Other possible explanations for considering the addition of explicit guarantees as a consumer protection mechanism include:
- Consistency with the prudential framework. Governments may be held liable for failures because these failures may seem to suggest inadequate prudential regulation. 7 This reasoning ignores the plausible co-existence of prudential regulation and failure in competitive markets. However, it is probably so widely held that governments are not able to formally abrogate responsibility for socialising losses (Llewellyn 1999). Misunderstanding the role of the prudential framework and the existence of depositor preference, for example, may artificially inflate community expectations of financial safety.
- Precedent. In the case of deposits, the FSI observed that Australians have rarely been exposed to substantial losses. They may therefore believe that they are well secured, with this perception supported by the response of governments to the failures of HIH and other institutions.
- Community views on fairness. People may regard the incidence of losses on retail financial products as simple bad luck. The case for compensation on these grounds no doubt seems strongest in cases where the losses cause extreme financial hardship. An example of community acceptance of the need for compensation in unfortunate circumstances is Compulsory Third Party (CTP) motor vehicle insurance.
Financial system and macroeconomic stability
4.39 Explicit guarantees on retail financial products may help support the stability of financial systems. This is more so for deposit-taking institutions, and is one explanation for why explicit guarantee schemes are more common internationally for deposits than for other products. They also provide a mechanism for reducing the impact of financial shocks on the economy, in particular, by preserving the ability of consumers to maintain spending and productive endeavours.
4.40 Explicit guarantees can help stabilise financial systems because they can reduce the chance of bank runs and contagion. Diamond and Dybvig (1983) showed analytically that runs can happen because banks engage in maturity transformation, using high leverage and sequentially callable liabilities (that is, demand deposits which can be withdrawn on a first-come, first-served basis). Taken together, these attributes of banks give depositors strong incentives to withdraw their deposits quickly, whenever they fear either that their bank may be insolvent, or that there may be a run on the bank.
4.41 An important insight of the Diamond-Dybvig theory is that runs can destroy even solvent institutions, because they are driven by self-fulfilling panic. International experience shows that in extreme cases, this panic can become quite general. The run can spread from one institution to another, disrupting the payments system and creating disorderly conditions in financial markets and the wider economy.
4.42 Credible guarantee schemes reduce the chance of this happening. If depositors are confident that their funds will be accessible regardless of the condition of the ADI, then they have little incentive to withdraw their money in response to bad news or rumour.
4.43 In the Australian context, any explicit deposit guarantee may be unlikely to extend beyond a subset of individuals and therefore might cover only a fraction of the liabilities of systemically important institutions. Household deposits, for example, represent only 22 per cent of the Australian liabilities of the four major banks. In these circumstances, it might be argued that such a limited guarantee would do very little to assist systemic stability since it would do nothing to reduce the incentives of non-guaranteed, wholesale customers to participate in a run.
4.44 It is important to recognise that ensuring system stability requires regulators to focus both on avoiding disruptive failures of systemically important institutions per se and, if a failure of any consequence does occur, avoiding any broader, unwarranted loss of confidence in the creditworthiness of similar institutions.
4.45 Therefore, by preventing such spillover effects, even a quite limited explicit deposit guarantee could provide a useful complement to the regulatory framework for forestalling financial system instability. Similar reasoning suggests that limited insurance policyholder protection schemes could also be important to sustaining confidence in other sectors of the financial system in the wake of institutional failures. The preservation of financial stability is not, however, a primary motivation for limited financial guarantees; instead it may be a welcome consequence.
4.46 A guarantee scheme can have a (more indirect) impact on financial stability via its effect on moral hazard. To the extent that the protection offered by a guarantee affects the behaviour of stakeholders in the financial system, there is the potential for increased risk-taking and risk of failure. In turn, this could increase the chance of a major systemic crisis which impacts on the stability of the financial system as a whole. As discussed previously, the design of any guarantee scheme is therefore critical to limiting moral hazard, not only to maintain the efficiency of the financial system but also to minimise any potential impact on financial stability.
4.47 Insolvency processes for financial institutions are likely to be lengthy, complex and expensive. The liquidation process for ADIs and insurance companies can take many years. Compared to a discretionary approach, an explicit guarantee provides some additional certainty to protected consumers that a failure may be resolved more quickly than would occur through the insolvency process, with greater certainty for consumers as to product coverage and greater certainty also about the possible scale of compensation.
4.48 In the event of failure, a guarantee scheme provides a mechanism where selected liabilities to a group of consumers are transferred to the scheme. In essence, the guarantee scheme would assume the group’s place in the insolvency queue. An explicit scheme can also ensure that governments and regulators have a well-defined approach to deal with financial institution failure when it occurs.
4.49 Explicit guarantees can create additional deadweight costs. In particular, depending on how schemes are designed and funded, it is possible that administrative and compliance costs associated with explicit guarantees will be higher than those involved in discretionary responses to institutional failure. That said, explicit schemes may better redistribute the costs of actual compensation away from taxpayers generally towards participating institutions and/or their stakeholders.
4.50 Further discussion of the costs of explicit guarantees is provided in Chapter 7.
Competition and competitive neutrality
4.51 By establishing a credible pre-commitment about how failures would be managed, a well-designed explicit guarantee may remove any advantages of larger institutions perceived to be ‘too big to fail’.
4.52 Implicit guarantees are a form of subsidy and, like all subsidies, they distort economic outcomes. They can cause a transfer of funds to the financial institutions where the guarantee is thought (rightly or wrongly) to be strong, thereby distorting competition within the financial system. Achieving competitive neutrality requires that the beneficiaries of a guarantee are charged an appropriate premium.
4.53 Therefore, implicit guarantees distort the competitiveness of institutions in the financial system. Institutions generally perceived to be implicitly guaranteed are advantaged, with the greatest benefit accruing to the most risky. An explicit guarantee which is correctly priced to reflect relative institutional risks of insolvency could address this distortion.
4.54 Explicit guarantees may distort the spectrum of risk, by increasing the range of financial assets that are deemed risk-free. They might also cause similar products to be priced differently because they lie on either side of the boundary dividing guaranteed from non-guaranteed products. For example, cash management accounts offered by ADIs involve counterparty risk and might be guaranteed. However, there may be no guarantee for (non-prudentially regulated) cash management trusts operated by ADIs’ funds management subsidiaries which involve market and agent risk.
4.55 A further concern with explicit guarantees is that, depending on how they are funded and priced, they may constitute an (additional) barrier to market entrants. A feature of some pre-funded schemes, for example, is that new entrants are required to contribute premiums for some minimum period (or amount) even though a scheme may be fully funded and premiums suspended for existing market participants. Similarly, under risk-based pricing, a new entrant commencing with the minimum required regulatory capital might be charged a higher premium than many incumbent players.
4.56 Potentially offsetting these price barriers is that explicit guarantees provide new entrants with instant ‘charter value’. In other words, a guarantee obviates the need for new entrants to have demonstrated a history of prudence in order to gain the trust and support of consumers of the guaranteed products.
4.57 Implicit guarantees are a contingent liability of government, culminating as a liability to taxpayers. They could also inflict substantial capital losses on holders of government bonds, if government debt issues (which would put upward pressure on interest rates) were needed to finance the payments resulting from implicit guarantees.
4.58 Appropriately designed explicit guarantees can help to protect taxpayers from the costs of future financial institution failures. This is done through limiting coverage and establishing appropriate funding and pricing mechanisms.
4.59 Although this may serve to reduce the impact of financial institution failures on public finances, it obviously involves a transfer of risk back to the financial system.
4.60 It is important to note that major crises may occur which affect the stability of the financial system and which no guarantee scheme would be capable of dealing with, no matter how well-designed or capitalised. In such circumstances, the role played by a guarantee scheme would, at best, be in support of other government actions designed to safeguard the workings of the financial system and economy. An explicit guarantee scheme can deal best with an individual or limited number of failures which do not involve potential costs to the scheme which are very large relative to the scale of the financial system. Large scale, systemic problems must be dealt with in other ways.
4.61 Explicit guarantees may adversely affect the incentives of regulators. If retail customers are protected against risk, regulators may be more accommodating of troubled institutions. This can make supervised institutions less efficient or even lead to a greater likelihood of their failure.
4.62 Explicit guarantees may also complicate closure rules for ailing institutions, and the foreclosure rules that best support explicit guarantees can often encourage weak institutions to absorb too much risk. This may be the effect where the regulator needs to be aggressive in foreclosing on failing institutions — especially those with low intrinsic worth or charter value — in order to prevent or limit the moral hazard of guarantees (Acharya 1996). Yet this strategy sometimes works perversely. For instance, it may force troubled institutions with low charter value to assume increased risk (in the hope of higher returns) at precisely the time that regulators would prefer that they be more cautious (Marshall and Prescott 2000).
4.63 On the other hand, explicit guarantees may have beneficial effects on the incentives of regulators to act earlier and decisively to minimise losses when confronted with impending failures. In particular, in circumstances where the regulator is confident that the most vulnerable consumers will be protected from the impact of foreclosing on an institution. The critical issue in this regard is the appropriate design of incentive and accountability arrangements for regulatory authorities rather than the existence or non-existence of a guarantee scheme per se. The interrelationships between a guarantee scheme and other regulatory arrangements are considered in Chapter 9.
4.64 It is evident that there is a range of cogent arguments both for and against the adoption of explicit financial sector guarantees. Where the balance lies obviously depends on the weightings that are attached to them and the appropriate design of any scheme.
4.65 In the Australian context, the case for adopting explicit guarantees rests partly on the presumption that there already exists a strong implicit guarantee of retail customer claims on financial institutions, and hence, many of the distortions associated with guarantees already exist. The crucial question then is whether moving from an implicit to an explicit guarantee offers a better public policy outcome.
4.66 With respect to design, it is clear that to the extent a case can be made for explicit guarantees, these should be limited to a small range of retail products offered by APRA-regulated institutions. Any sensible guarantee scheme requires monitoring of guaranteed institutions but duplication of APRA supervision would be inefficient.
4.67 Informed customers and large investors provide a crucial source of market discipline and can assess and take steps to protect themselves against the risk of institutional failures. Scheme design issues necessary to maintain such discipline are further discussed in Chapter 6.
4.68 To realise the full potential benefits of explicit guarantees they must be appropriately priced, ideally reflecting the individual insolvency risks of participating institutions. Although public policy and practical considerations could inhibit pure risk-based pricing this may not be a major problem in a well-supervised financial system. Issues concerning funding and pricing are discussed in Chapter 8.
1 A history of financial institution failure and government responses in Australia is contained in Chapter 2 and Appendices 4.2 and 4.3.
2 For logical consistency, such a policy stance might also require a winding back of the scope of prudential regulation.
3 A possible variant might be to encourage voluntary, private provision of such guarantees. Such a market does exist in relation to deposit insurance in North America, as an adjunct to the government-run schemes in Canada and the United States. It is not clear, however, that in the absence of such base government cover, private insurance would be a viable proposition. At the very least, a prerequisite would be a credible government policy stance of non-intervention in financial institution failures which, as noted above, is likely to be problematic in the Australian context. The ultimate solvency of a private provider is also an issue for the success of such schemes.
4 See, for example, Miller 1996; Santomero 1997; Benston and Kaufman 1995; Goodhart 1991; Llewellyn 1999. Gropp and Vesala (2001) interpret a reduction in risk-taking by European banks following the introduction of explicit deposit insurance schemes as indicative of a removal of broader implicit insurance.
5 Fifty-six per cent of respondents to a recent ANZ Bank survey (ANZ 2003) indicated that they consider either price or the convenience of their existing provider when renewing insurance.
6 Gorton and Pennachi (1990) provide a theoretical justification for deposit insurance as a mechanism which under some assumptions enhances social welfare by creating a supply of risk-free liquid assets to protect uninformed participants in the financial system.
7 One of the findings of the recent ANZ Bank survey (ANZ 2003) was that 3 per cent of respondents felt that all financial products were guaranteed by the regulators. However, this was in response to a question asked about financial sector regulation generally, not just those products or institutions that are prudentially regulated. In contrast, a survey conducted in the United Kingdom by the Financial Services Authority (FSA 2003) tested respondents’ appreciation of whether (prudentially) regulated firms would be allowed to fail (and that they could lose money as a result). Thirty-three per cent of respondents correctly acknowledged that all regulated firms are potentially allowed to fail. Another 33 per cent thought that only some types of prudentially regulated firms would be allowed to fail, while a further 19 per cent thought that no regulated firms were allowed to fail. Twelve per cent responded that they did not know.