Study of Financial System Guarantees Australian Government, Department of the Treasury,

Chapter 2: Australia’s experience with failure and international experience with guarantees

Overview

  • The incidence of significant financial institution failure in Australia has been relatively low by international standards.
  • Government responses have rarely involved attempts to prevent insolvent institutions from failing or to shield creditors or shareholders from loss. Rather, they have generally sought to mitigate the impact of failure for certain customers.
  • A number of lessons are evident from experience:
    • community expectations of government support appear widespread, and may be stronger for particular financial institutions and products;
    • the nature and associated consequences of a failure are likely to differ substantially across the sectors and in each case;
    • the probability, extent and associated consequences of a failure can be difficult to predict in advance; and
    • the time between failure and resolution can be significant.
  • Internationally, deposit insurance and insurance policyholder protection schemes are becoming widespread. Australia is one of only two Organisation for Economic Cooperation and Development (OECD) countries without some form of explicit deposit insurance.
  • Explicit guarantee schemes for other products offered by prudentially regulated financial institutions are growing in number throughout the world.

Financial institution failure in Australia

2.1 Throughout the last two centuries and across all sectors of the Australian financial system, there have been only infrequent failures.1 Australia’s experience with restructuring within the prudentially regulated sector has been marked more by relatively uneventful mergers and well-managed exits than by spectacular failures and their associated consequences. Australian prudential regulators have played an important role in the process of managing the smooth exit of troubled institutions.2

2.2 In recent decades there have been some large and notable failures such as the collapse of the State Bank of South Australia and the State Bank of Victoria, Pyramid Building Society (Pyramid), and the case of fraud and subsequent collapse of life insurers Occidental Life and Regal Life. Most recently, there has been the failure of general insurers, the HIH Group of Companies (HIH), and medical insurance providers, United Medical Protection/Australasian Medical Insurance Limited (UMP/AMIL), were placed into provisional liquidation. These examples, however, need to be considered in context. Current generations of Australians have experienced relatively few instances of financial institution failure.

Early history

2.3 Australia’s early experience with failure is punctuated by a number of important episodes. The banking sector throughout the nineteenth century experienced considerable turbulence and numerous bank failures. Key events included the 1826 liquidity crisis and the depressions of the 1840s and 1890s.3 The twentieth century was less volatile with just three Australian banks suspending payment. The last bank failure in which Australian depositors lost money (and then only a minimal amount) was that of a trading bank, the Primary Producers Bank of Australia, in 1931 (Fitz-Gibbon and Gizycki 2001). Since the early 1930s, banking sector problems have been resolved without losses to depositors.

2.4 In the case of life insurance companies, between 1901 and 2003, 11 life insurance companies entered liquidation, with the majority of these occurring during the late 1920s and early 1930s, and one in 1954. Since the introduction of the Insurance Act 1973 in the early 1970s, 30 general insurers have entered liquidation (Australian Prudential Regulation Authority (APRA) 2003), with several cases of failure.

Recent history

2.5 Volatility in the banking sector re-occurred in the late 1980s and early 1990s. Deregulation of bank lending and removal of restrictions on foreign bank entry led to increased competition and a desire by some institutions to grow rapidly. This took place in an environment in which asset prices, particularly commercial property prices, were increasing quickly, and credit assessment procedures in many financial institutions had not adjusted to the newly liberalised environment.

2.6 The result was extremely strong credit growth secured against increasingly overvalued commercial property. While resulting losses were sustained throughout the banking system, they had greatest impact on banks owned by State governments and foreign banks.4 The problems were sufficient to lead to the failure of the State Banks of Victoria and South Australia. In the face of large losses, public confidence in the banking system weakened during 1990 and 1991.

2.7 By and large, the losses incurred by the State banks were paid for by the taxpayers of the States concerned. The State governments (as owners) had unconditionally guaranteed all liabilities (not just deposits) of these banks. Hence, a relatively broad range of stakeholders were protected.

2.8 The fragility of public confidence arising from failures in the banking sector in the early 1990s extended to non-bank deposit-taking institutions. The failure of Pyramid marked the most significant failure in this sector at that time. Pyramid was Victoria’s largest building society and Australia’s second largest. The Victorian Government ultimately ‘bailed out’ depositors at a cost to taxpayers of over $900 million, leaving other creditors and investors, including holders of redeemable preference shares to bear losses.

2.9 Pyramid’s problems had flow-on effects for other non-bank financial institutions in Victoria, with the highest profile case being the OST Friendly Society. Like Pyramid, OST was heavily exposed to the property market, and its problems were eventually resolved by a merger with IOOF (the largest friendly society). Pyramid’s difficulties may also have contributed to short-term deposit runs on the Bank of Melbourne and Metway Bank (former building societies). The runs stopped shortly after the Reserve Bank of Australia (RBA) issued press releases stating that the banks continued to meet prudential standards and were soundly managed. The RBA did not provide emergency liquidity support in any of these cases (Gizycki and Lowe 2000).

2.10 The start of the 1990s was also a difficult time for credit unions. There had been several notable failures in the late 1980s, where members’ funds were protected by State-based support schemes, other credit unions and/or governments.5 Consumer uncertainty arising from these events and fallout from the Pyramid failure created temporary liquidity problems for a number of credit unions.

2.11 Life insurance failure in Australia has been rare in the post World War II period. The most recent cases of failure are that of Occidental Life and Regal Life which were unable to meet their obligations due to the improper use of $65 million from statutory funds. Payments by the Bank of Melbourne to remedy the problems which occurred in the settlement process during the aborted sale substantially eliminated any shortfall in assets. The insurance companies were subsequently taken over by Mercantile Mutual Life Insurance Company Limited. The worst affected policyholders lost less than 10 per cent of their policy value and up to one year’s uncredited interest on their savings.

2.12 General insurance failures have been more common although relatively minor and infrequent with VIP Insurances, Palmdale, Bishopgate and New Cap Re being the most recent cases prior to the failure of HIH. From the late 1990s into the early part of the current decade, the general insurance sector experienced major problems internationally. The collapse of HIH, then Australia’s second largest insurer, marked a significant failure of a major general insurer. Considerable and widespread consequences surrounded the collapse and the Government announced a Royal Commission to examine the circumstances of the failure. The Government also announced the HIH Claims Support Scheme (HCSS) for some affected policyholders.

2.13 Many of the broader and contagious consequences of the HIH failure were a result of the company’s significant market share, particularly in some of the statutory classes.6 This raised general concerns about the impact of the failure of a general insurer on the financial system and the economy more broadly. This period also witnessed UMP/AMIL, a large medical insurance provider, being placed in provisional liquidation. The Government later announced a package of reform measures.

2.14 There have been some instances in Australia where members of superannuation funds have lost money as a result of mismanagement or fraud or theft by a trustee. The best known example is Commercial Nominees of Australia (CNAL), which was trustee for a number of superannuation funds. CNAL’s actions in 2000 resulted in an estimated 25,000 investors losing a proportion of their superannuation savings, which amounted to a total loss of over $24 million, or around 8.5 per cent of CNAL’s funds under management.7

2.15 Appendix 4.1 provides a timeline of failure and an overview of subsequent policy reform measures. Appendix 4.2 provides some case studies of Australian financial failures.

The lessons from history

2.16 Failures throughout Australia’s history demonstrate that, while some significant losses occurred, the prevailing regulatory arrangements were broadly successful in dealing with the pressures that emerged during those periods.

2.17 Despite the rarity of failure in Australia and the infrequency of government financial assistance, community expectations of government intervention appear strong. There may be a number of possible explanations for this. One possible misconception is that because the Government is responsible for the regulatory framework, it therefore guarantees those systems against failure.

2.18 The nature of failure will differ according to the type of institution involved and its relative business mix. For example, banking failures typically involve erosion in asset values linked to general economic conditions or imprudent lending practices. Alternatively, they may be precipitated by ‘runs’ by depositors that reduce the bank’s liquidity and ultimately threaten its solvency. Insurance failures generally are associated with random catastrophic events, prolonged investment market downturns and/or long-term risks in matching current premium revenues with future liabilities. Each instance of failure will be different and affect stakeholders in different ways.

2.19 Experience also demonstrates that the final cost of a given failure can be difficult to predict and can vary widely in each case. Differences in financial structures and legal arrangements across sectors make comparisons difficult. The accounting value of an institution’s assets and liabilities prior to failure will not necessarily be an accurate indicator of the likely shortfall or cost of resolution. For instance, general insurance liabilities, especially ‘long-tail’8 liabilities, can be difficult to determine and may be serially underestimated. Such ‘discrepancies’ may not materialise until after the failure of the institution. In addition, the value of remaining assets of a failed institution may deteriorate as resolution is taking place. This makes determining the level of potential exposure to failure problematic.

2.20 A feature of past financial disturbances and institutional failures in Australia is that they have often served as a catalyst to significant subsequent policy and regulatory reforms. Appendix 4.3 provides an overview of some of the recent policy responses to failure.

2.21 There can be no absolute certainty, however, that even with best practice regulation and supervision, failures of significant financial institutions will not occur in the future. Governments and regulators cannot prevent poor management, nor can they be a fully effective proxy for the market discipline exerted by sophisticated investors.

2.22 Risk is a critical feature of the financial system. Indeed, the continuing international development of financial markets and the ever increasing innovation and sophistication of risk transfer mechanisms within the financial system are readily observable trends. Such developments are likely to continue to test the ability of our regulators to maintain the strength of the regulatory framework at the level expected or demanded by the community.

2.23 Australia has a relatively modern, flexible, and competitive financial system, which engages closely with the rest of the world. Its development does not appear to have been hindered by the absence of explicit guarantee schemes.

2.24 On balance, notwithstanding the recent experience with HIH, Australia’s existing regulatory system has a very favourable international reputation. Indeed, aspects of it (such as creation of a multi-sector prudential regulator separate from the central bank) have been embraced as a model for reform in other economies.

2.25 An important question thus arises as to whether Australia’s strong track record of financial stability can be partly attributed to the way in which our existing regulatory framework allows efficient risk-taking in conjunction with appropriate risk sharing by all stakeholders.

2.26 Internationally, however, there has been increasing use of explicit financial system guarantees as an important component of regulatory systems.

International experience with financial system guarantees

2.27 There is a considerable wealth of international literature and opinion on guarantee schemes, but this focuses primarily on deposit insurance. It includes material from the Financial Stability Forum (FSF),9 the World Bank and the International Monetary Fund as well as practitioner associations and a number of leading academics in the field. This includes detailed guidance on the issues to address in considering and developing deposit insurance schemes as well as surveys of international practice.

2.28 Deposit insurance and insurance policyholder protection schemes are becoming more widespread in financial systems around the world and are fairly common amongst OECD member countries. For example, 28 of the 30 OECD countries now have in place or are implementing explicit deposit insurance schemes.10

2.29 Twenty one OECD countries are reported to have in place or are implementing schemes for life and general insurance products. Many focus on protecting claimants under compulsory classes of insurance 11 while 9 of the 21 schemes extend beyond compulsory classes (Yasui 2001).12 These include Canada, France, Ireland, Japan, Korea, Norway, Poland, the United Kingdom (UK) and the United States (US) (OECD 2001).

2.30 A number of countries have implemented guarantee schemes to cover situations where an employer, who has occupational superannuation commitments, becomes insolvent. These include Germany, the US and UK13 and in many cases these countries run defined benefit schemes.14 Under the schemes, employers are required to have insurance cover in place to ensure adequate protection in the event of employer insolvency. Australia’s superannuation and retirement income arrangements15 differ from many countries making international comparisons difficult.

2.31 Table 2.1 provides additional statistical information on the adoption of explicit schemes for depositors. Table 2.2 provides some statistical information on international insurance schemes. Newly emerging schemes for life and non-life insurance are less well researched with data on coverage features less readily available.

Table 2.1: Key statistics on deposit insurance schemes — globally and for OECD countries

Variables
 
Globally
OECD
Type
Explicit scheme
68
28
 
Implicit or no scheme
110
2
Fund
Pre-funded
58
21
 
Post-funded
10
7
 
Coinsurance
17
9
 
No coinsurance
51
19
Premiums
Risk-adjusted premiums
23*
9*
 
Flat rate premiums
45
19
Fund sources
Public funding
1
Nil
 
Private funding
15
8
 
Joint funding
51
20
Membership
Compulsory membership
55
27
 
Voluntary membership
13
1
Administration
Official administration
33
12
 
Private administration
11
8
 
Joint administration
24
8

* Since the survey, Canada appears to have introduced differential premiums. In addition, contributions under the French scheme are now risk-based. Hong Kong introduced legislation into parliament in 2003 to establish a deposit protection scheme intended to become operational in 2005. There may be other changes since this time that have not been included.
Source: World Bank Deposit Insurance Database 2000.

2.22 Increasingly, explicit guarantees are being embraced as a legitimate and desirable component of the overall financial regulatory system. While the vast majority of financial sector guarantee schemes operating internationally relate to the protection of bank deposits, there has been a noticeable move towards implementing guarantee schemes in the insurance sector in recent years.

Table 2.2: Statistics on insurance schemes — OECD countries

Countries
Life /Non-life insurance
Funding
Canada
Life and non-life
Post
France
Life
Pre
Ireland
Life
Post
Japan
Life and non-life
Pre
Korea
Life and non-life
Pre
Norway
Non-life
Pre
Poland
Non-life
Post
UK
Life and non-life
Post
US
Life and non-life
Post

Source: OECD Workshop on insurance and private pensions in the Baltic States, February 2002.

2.23 Appendices 2.3 and 2.4 provide a more detailed comparison of selected countries’ guarantee systems for deposit-taking and insurance schemes.

2.24 A number of options are available to a country regarding protection of consumers in the case of financial institution failure (Garcia 1999).16 These range from explicit denial of protection (New Zealand), through discretionary approaches and implicit guarantees, to limited and comprehensive explicit guarantees.

2.25 The US has the oldest explicit deposit insurance scheme in existence, with the Federal Deposit Insurance Corporation (FDIC) commencing in 1934. Policyholder protection schemes for general insurance have existed in the US since the late 1960s. Elsewhere, explicit schemes have been in existence since the late 1960s and early 1970s, such as in Canada and the UK, while other countries such as Hong Kong and Korea have introduced schemes over the last five years. The schemes have generally been introduced in the aftermath of a major bank’s failure or more general crisis. One certainty exists — the range of guarantee schemes is highly diverse; no one scheme fits all circumstances.

2.26 Despite an apparent international movement towards their adoption, explicit guarantees may not be applicable or appropriate in every situation. Moreover, the institutional and product coverage of these guarantees is never assumed to be universal. There is ample recognition that introducing guarantees in the absence of an adequate supervisory framework, or introducing poorly designed guarantees, can weaken a country’s financial system over time and may give rise to significant fiscal exposures.

2.27 Various studies comparing international experience have highlighted a correlation between financial crises and the existence of explicit deposit insurance schemes, while the US Savings and Loan crisis in the 1980s is often cited as a case study of what may go (very) wrong. Notably, however, it is a combination of overly generous guarantees, inadequate supervisory and regulatory structures, and lack of market discipline (rather than guarantees per se) which appear to have been factors in explaining the adverse experiences with guarantee schemes.17 This is also the conclusion which emerges from the much more limited research into the performance and effects of insurance guarantee schemes (Bohn and Hall (1995, 1997) and Hall (1998)).

2.28 Few countries that have adopted explicit deposit insurance or insurance guarantee schemes have had the opportunity to consider the issue across the financial system as a whole, nor during times of relative stability.18 Most countries have implemented sector or product-specific schemes despite ongoing blurring of institutional boundaries and product overlap.

2.29 As noted by Garcia (1999), ‘if well-designed, an explicit deposit insurance system can be preferable to no insurance and can complement legal priority’. On the other hand, a poorly designed deposit insurance scheme is likely to exacerbate existing problems and possibly have systemic implications. While systemic concerns are lesser in relation to insurance, many of the concerns in relation to the need for well-designed schemes of deposit insurance would appear to be applicable for insurance schemes.

2.30 Underlying prudential structures, and the stability of institutions, are important in ensuring that a system of guarantees does not lead to adverse outcomes for bank stability. Demirguc-Kunt and Detragiache (2000) argue that a system of deposit insurance has the potential to lead to banking instability and higher chances of failure by reducing market discipline. Nevertheless, they argue that in countries with stable institutions ‘an effective system of prudential regulation and supervision is in place to offset the lack of market discipline [that may be] created by deposit insurance’.

2.31 Laeven (2002) also found that ‘the effectiveness of deposit insurance has shown to be country specific’. Viewed another way, Demirguc-Kunt and Detragiache (2000) argue that in countries with weak institutions and poor regulatory frameworks, the benefit of a guarantee system is diminished and has the potential to lead to banking crises. As noted by Demirguc-Kunt and Kane (2002), ‘deposit insurance is neither always good nor always bad. It can be a useful part of a country’s overall system of bank regulation and financial markets’.

2.32 The Study has had regard to the range of international literature and best practices. Two key issues noted from the literature that have guided deliberations are:

  • the unique nature of the Australian financial system. Schemes are best designed around country specific requirements; and
  • the importance of a well-designed scheme. Poorly constructed schemes can be expensive and they can have perverse incentive effects.

1 In simple terms, an institution may be said to have failed when it cannot meet its financial promises to customers, employees or other creditors. Failure as discussed in this Study can be differentiated from ‘managed exits’ from the financial system which regularly occur on a voluntary basis, or as a result of a takeover (sometimes prompted by regulatory oversight of a troubled institution).

2 The experience of the early 1990s, however, when two significant sized State-government owned banks required recapitalisation and were subsequently sold, highlighted flaws in the then extant structure of supervisory arrangements. These institutions were not subject to the prudential framework that was then overseen by the Reserve Bank of Australia.

3 Figures on bank liquidations during this period are not available.

4 Money market corporations (investment and merchant banks) and finance companies (in some cases subsidiaries of banks), were sectors which also experienced problems and some failures.

5 A number of Australian States including NSW and Victoria operated ‘stabilisation funds’, which served liquidity support, failure resolution and deposit insurance type roles. Funded by credit union contributions their purpose was to provide resources to facilitate rehabilitation through transfers of business.

6 State governments who are responsible for the regulation of the statutory classes (Compulsory Third Party (CTP) motor vehicle insurance, Workers’ Compensation and Builders’ Warranty) also became involved in ‘bailouts’ of some statutory classes. The NSW and Queensland Governments announced separate rescue packages for CTP and the Western Australian and Tasmanian Governments for Workers’ Compensation.

7 The estimated cost to date of rehabilitating CNAL’s funds is around $17.5 million. Refer to Appendix 4.2 for details on arrangements for rehabilitating CNAL. Table 3.1 in Chapter 3 provides details on the operation of Part 23 of the Superannuation Industry (Supervision) Act 1993.

8 ‘Long-tail’ business involves considerable lags between a claimable incident and the settlement of the claim, and contains uncertainties over the amount, timing and potential length of payouts. Examples include CTP, professional indemnity and workers’ compensation.

9 The FSF is an international body comprising representatives from central banks and prudential supervisors which was created in 1999 to promote international financial stability, improve the functioning of markets, and reduce systemic risk.

10 Amongst OECD members, Australia and New Zealand are the only two countries which do not have some form of deposit insurance. Between 80 and 90 per cent of New Zealand banking assets are Australian controlled.

11 This type of scheme steps in when an insurer becomes insolvent as well as when an insurer cannot be found, similar to the way in which nominal defendant schemes operate in some of Australia’s States and Territories.

12 It appears as though Australia is counted among those countries with insurance guarantee schemes by virtue of existing State-based compensation arrangements for privately underwritten CTP, Workers’ Compensation and Builders’ Warranty insurance.

13 There is limited data available on these types of schemes.

14 See Chapter 5 for a discussion on defined benefit schemes.

15 Australia has a three pillar system comprising: the age pension, a government-funded social safety net; compulsory superannuation, funded by employer contributions; and superannuation and savings funded by the individual or voluntary employer contributions.

16 Garcia identifies six separate options: an explicit denial of protection; legal priority for depositors over other creditors; ambiguity regarding guarantee coverage; an implicit guarantee; explicit limited guarantees; and explicit full guarantees.

17 Demigurc-Kunt and Kane (2002) provide a valuable recent review of some of this international evidence and lessons which may be drawn from it.

18 The UK is one exception. Following its establishment, the UK Financial Services Authority undertook a comprehensive public consultation process concerning compensation arrangements across the UK financial sector.