Submission to the Financial System Inquiry 4. Sources and Management of Systemic Risk

This Chapter outlines the concept of systemic risk and the factors that might give rise to it. Sources of systemic risk in the Australian financial system – entities, activities and markets – are examined as well as the measures in place to manage it. The key messages from this Chapter are:

• Systemic risk is the risk of financial system disruption so widespread or severe that it causes material damage to the real economy.
• The four major banks are important sources of systemic risk in Australia because of their size and interconnections with the real economy and the rest of the financial system. While on the one hand, the systemic risk posed by the major banks is heightened due to the similarity of their business models and funding structures, on the other hand, their business models are relatively low risk.
• Compared with other assets, housing is not particularly risky in most countries. The housing market nonetheless poses systemic risk because of its size, importance to the real economy, and interconnection with the financial system, including through the sizeable exposure that banks have to housing. While not as large, the commercial property market also poses systemic risk through its cyclicality and strong connections to the banking system; historically it has been one of the main sources of loan losses during episodes of banking distress globally.
• Financial market infrastructures (FMIs) are critical to the smooth functioning of financial markets, but can also be a source of systemic risk because of their size, strong connections with banks and other financial institutions, and the lack of substitutes in the markets they serve.
• The Reserve Bank closely monitors systemic risk in the Australian financial system, including how risk can propagate. Public policy does not seek to eliminate systemic risk – this would require removing certain risk-taking altogether, which would not be optimal for society. However, effective public policy design and other mitigation strategies can substantially reduce systemic risk and the impact on the financial system and real economy.

4.1 What is Systemic Risk?

There is no consensus definition of systemic risk, but a reasonable working definition is that it is the risk of financial system disruption so widespread or severe that it causes, or is likely to cause, material damage to the economy.^[1] Such disruptions can have significant implications for employment, wages, prices and activity, including for people and firms outside of the financial sector. That is, systemic risk and financial system disruptions can impose large negative externalities (external costs) on the rest of the economy. These externalities imply that a financial system without regulation to address these risks is not socially optimal. Although some definitions of systemic risk do not emphasise negative effects on the real economy,^[2] the public policy case for seeking to manage or mitigate systemic risk stems from these; the experience of the global financial crisis highlights that the effects on the real economy can be severe.

Systemic risk differs from financial risk (the various types of which are outlined in ‘Box 4A: Types of Financial Risk’). Financial risk involves both good and bad potential outcomes for those that assume such risk, whereas systemic risk is concerned solely with negative outcomes for the economy. Moreover, because systemic risk is borne by the economy as a whole, it cannot be fully diversified away or transferred to others like financial risks can be. However, systemic risk can be reduced by certain policies.

Several aspects can be identified that may lead to a particular risk being considered systemic.

• An entity or market might be so large that adverse outcomes for it would have significant effects on the economy. Thus even if the risk of this outcome is very low, the entity or market is considered to pose systemic risk because of its sheer size: in other words, it is systemically important. In Australia, the major banks and the housing market both fall into this category.
• In other settings, risks might propagate from one entity or part of the financial system to others because of their interconnections, such as borrowings or payments. That is, a shock to one institution can be quickly propagated to others through direct connections. Examples of networks of entities that borrow from one another, and that in principle could propagate distress through chains of failures, include the banking system and the non-financial businesses that extend trade credit to one another. Actual or expected failure to settle payments in a high-value payment system or in insurance and other risk transfer markets can also propagate distress. These are all examples of ‘contagion’ within the financial system.^[3]
• Systemic risk might also arise in cases where individual entities or markets, including those that are small in size, tend to fail or experience financial distress all at once. This occurs because they are affected by the same shocks or behave in the same way in particular circumstances; that is to say, they display correlation. This also falls within the term contagion as it is generally used. One example is the reactions that can occur in financial markets when there is uncertainty about a firm's financial position, especially in distressed market conditions. Creditors may extrapolate bad news about one institution to other institutions (particularly those with similar business or funding models) and ‘rush to the exit’.
• Systemic risk often stems from cycles in financial risk-taking by entities and market participants; this mechanism can be referred to as procyclicality, although it is not only a function of the business cycle (Kindleberger and Aliber 2011). Under certain circumstances, investor exuberance can be self-propagating, as rising prices caused by greater risk-taking are wrongly interpreted as signals of favourable future financial and economic conditions. Similar behaviours can also accelerate downturns, as purely financial distress, and the changes it causes, are interpreted as signals of less favourable future conditions. One manifestation of procyclical risk-taking that is critical for systemic risk is leverage cycles, such as those in financial markets and property markets (Fostel and Geanakoplos 2013). Leverage plays a key role in many risk amplification processes; losses that less-leveraged entities could absorb may render more-leveraged ones unable to meet obligations to creditors.

In practice, many sources of systemic risk involve more than one of these aspects. For example, the effects of procyclicality may be amplified through the various channels of correlation and interconnection. And a highly leveraged large institution in financial difficulty is, other things equal, more of a problem than a small one.

Most researchers accept that a variety of shocks can lead to systemic disruption, both shocks from outside the financial system such as a recession or natural disaster, and those generated by the activities of the sector itself. Examples of internally generated shocks include lending booms and lax lending standards that lead to broad-based credit problems, the propagation of stress through the global financial system via derivatives and wholesale funding markets, as well as simple but large shocks to prices of financial assets (BCBS 2004; ECB 2009a; Bernanke 2010). Consistent with this, a shock that triggers systemic disruption may initially affect a large number of institutions or only a single institution (De Bandt, Hartmann and Peydro-Alcalde 2010). The Group of Ten (2001) acknowledged that, in some concentrated financial systems, the collapse of a single market or financial institution may itself constitute systemic disruption.

The costs of systemic financial crises can be severe, as is clear from the deep recessions and slow recoveries in many of the economies most affected by the recent crisis, such as the United States, the euro area, and the United Kingdom.^[4] A number of mechanisms produce these costs (Cecchetti, Kohler and Upper 2009), although how they operate is still a topic of research. One important channel is a decline in the availability of credit. Whether this reflects lenders' responses to deterioration in the balance sheets and creditworthiness of their customers, or deterioration in their own financial health, the outcome is that funding of productive activity becomes constrained.^[5] Another mechanism is the wealth effect on household consumption when asset prices decline, as they often do in periods of financial disruption.^[6] The severity of such effects in large part depends on the strength of any feedback loops between the financial and non-financial sectors of the economy.

4.2 Systemic Risk in the Australian Financial System

Systemic risk varies across the Australian financial system, reflecting that some parts of the system are more important to market functioning and economic activity than others. Systemic risk also varies over time due to changes in risk appetite and the operating environment.

4.2.1 Banks

Within the financial sector, banks are the entities that are widely regarded as posing the greatest systemic risk. There are at least three reasons for this:

• First, banks provide financial services that are critical to the functioning of the economy, in particular payment facilities, deposit-taking and allocating credit to productive activities. Disruption to these types of services would be expected to impose significant costs in terms of foregone economic activity.
• Second, banks are inherently unstable because of their asset-liability structure. Banks assume liquidity risk by engaging in maturity transformation and providing liquidity services. These inherent functions of banks mean they are exposed to liquidity shocks, such as funding market disruptions. If a bank's creditors seek to withdraw their funds at the same time, the bank may be unable to meet its obligation to return these funds due to the difficulty of liquefying its assets. Banks use fixed-price liabilities (including deposits) to fund most of their assets, so they are highly leveraged; that is, their capital is equivalent to only a small proportion of their total assets. This gives depositors and other creditors only a small buffer against losses on a bank's assets, increasing their incentive to quickly withdraw their funds in times of financial distress or uncertainty.
• Third, problems in one or two banks – whether internally generated or arising from external shocks – have the potential to quickly spread to others. Contagion in banking markets can stem from interconnections between banks through borrowing and lending activities, payment or derivative transactions, as well as correlated exposures and similar funding structures. In certain circumstances, a lack of transparency and uncertainty surrounding parts of banks' balance sheets can be enough for some creditors to run on an entire banking system rather than simply one or two banks.

4.2.1.1 The four major banks

Some entities and markets can pose risk simply because of their size or importance to the economy, even if they are low risk in an individual sense. The four major banks in Australia – Australia and New Zealand Banking Group (ANZ), Commonwealth Bank of Australia (CBA), National Australia Bank (NAB) and Westpac Banking Corporation (WBC) – can be reasonably considered to fall into this category. Their stand-alone credit ratings across several credit rating agencies and low funding spreads show that the market considers them to have a low risk of failure. Such an assessment is consistent with their capital positions and business models, as well as, more broadly, Australia's sound regulatory and macroeconomic environment. In the unlikely event that one of these entities were to become distressed, however, this would be likely to have a marked and negative effect on the real economy.

This recognition motivated their recent designation as domestic systemically important banks (D-SIBs) by the Australian Prudential Regulation Authority (APRA) (Chapter 3). In determining the systemic importance of the Australian banks, APRA (2013a) examined four broad indicators – size, interconnectedness, substitutability and complexity – and found a clear distinction between the major banks and other banks.^[7]

• The four major Australian banks are much larger in the domestic market than other banks, with their resident assets each representing between 16 per cent and 23 per cent of resident banking system assets, compared with 2 per cent for the next largest bank (Table 4.1).
• Compared with other banks, the major banks have deeper connections to other financial institutions through their borrowing and lending activities. Using data on large exposures to individual counterparties (which capture risk concentrations arising from both on- and off-balance sheet transactions), Tellez (2013) found that the Australian banking system network is characterised by a large number of financial institutions being exposed to the four major banks and, to a lesser extent, a few other large banks.
• Substitutability can be regarded as the ease with which other banks can replace the services that were provided by a failed bank. The four major banks dominate the provision of financial services considered most critical to the economy: they each account for a large share of lending to households and businesses and, together, the bulk of payments activity.
• The complexity of a bank's activities is often measured by its holdings of over-the-counter (OTC) derivatives, trading assets and structured financial assets. The activities related to these can increase the opacity of a bank's financial position and thus the difficulty in resolving it in the event of failure or if it encounters severe financial stress. Although the four major banks are less engaged in these activities compared with many of the largest global banks, their complexity is generally measured to be much higher than most other banking institutions in Australia.

There are other, less direct, ways in which the four major banks are important to the Australian financial system and the economy. In particular, the household sector has a sizeable equity (and debt) exposure to them via superannuation and other investment portfolios. For example, ABS data indicate that domestic retail investors and other domestic investors (mainly superannuation funds) own around three-quarters of the major banks' equity. The major banks' share of equity market capitalisation in Australia is around 25 per cent, which is higher than the (listed) banking sectors in a number of other large countries (Graph 4.1). Given this sizeable equity exposure, a substantial decline in the major banks' market value could result in a material negative wealth shock to the household sector, albeit one that is unevenly distributed across households.

In addition to their individual systemic importance and interconnections with the economy, the systemic risk posed by the major banks might be heightened due to their similarity (or perceived similarity). In the terminology used above, they are correlated in ways that could increase the chance of simultaneous credit losses, joint funding stress, or sentiment-driven contagion. A range of balance sheet metrics and banking indicators support the assertion that the major banks' businesses are quite similar.^[8] They have: similar sized balance sheets; broadly comparable shares of lending and deposits; relatively large shares of housing loans; similar capital positions; and the same credit ratings (Table 4.2). Furthermore, non-resident assets comprise a fairly similar proportion of their balance sheets at between 15–35 per cent, with New Zealand being the largest offshore business in most cases.

Although it could be argued that similarity increases the probability of joint bank financial distress, it could still reduce systemic risk if it reflects the joint adoption of business models that are relatively low risk. This is a consideration in Australia, given that the major banks' businesses focus heavily on traditional commercial banking – that is, servicing household and business customers – in Australia and New Zealand. The major banks have been able to earn good returns from this business, giving them less incentive to pursue potentially high-yielding, but higher risk, assets offshore. They had little exposure to the complex structured credit products that were the catalyst of the global financial crisis; they are also not significantly involved in the types of investment banking activities that faced difficult market conditions and significant structural change of late. Indeed, loans represent the bulk of their assets, whereas trading assets and securities are only a small share, including compared to most large banking groups globally (Ellis 2013). And within their loan portfolios, housing loans – which typically pose less risk than most other forms of lending – account for over half of total loans.

Systemic risk can, nonetheless, be heightened by perceptions that systemically important banks are ‘too big to fail’. If creditors assume that the public sector will not allow such banks to fail, they will assess the risk of failure as lower than that justified by a bank's stand-alone financial health. This can result in competitive distortions, allowing these entities to grow larger and become even more systemically important. It may also encourage moral hazard as banks that are perceived to be too big to fail may take on excessive risk with the assumption that the public sector will step in to cover the costs of downside risks, whereas the bank itself is the sole beneficiary on the upside. These dynamics can lead to large costs for the public sector in the event of failure. It is very difficult to quantify the extent to which the too big to fail issue exacerbates systemic risk in Australia or other countries more generally, and thus to calibrate policies to account for apparent lower costs of funding or more risky behaviour.^[9]

4.2.1.2 Non-major banks and other lenders

In addition to the major banks, the Australian banking system includes a range of other authorised deposit-taking institutions (ADIs): smaller Australian-owned banks (including mutual banks), foreign-owned bank subsidiaries, foreign-owned bank branches, credit unions and building societies. The designation of only the four major banks as D-SIBs reflects the comparatively smaller direct impact that the failure of a single non-major ADI would be expected to have on the domestic financial system and economy. However, the correlated behaviour of bank creditors during periods of severe financial stress means the failure of any single institution could transmit financial distress to other institutions that are perceived to be similar; a degree of systemic risk could emerge from such situations. Relatedly, some of these entities face higher levels of individual financial risks than the major banks, given their loan books are typically more concentrated, both geographically and in particular business lines. Their funding structures are also less diverse, with some funding markets not available to particular institutions.

In addition, it may be the case that smaller institutions aggressively pursuing growth can heighten procyclicality and thus overall systemic risk within an economy. There is some international evidence that foreign bank branches expand at a relatively fast pace during buoyant times and then reduce lending relatively quickly during downturns (Hoggarth, Hooley and Korniyenko 2013). Lending by foreign bank branches in Australia has been quite procyclical over the past decade, and may have influenced some asset prices (such as commercial property prices) in instances where it has been provided to more marginal borrowers (Graph 4.2).

Registered financial corporations (RFCs), comprising finance companies and money market corporations, are another set of financial institutions involved in credit intermediation in Australia. As these entities are not prudentially regulated by APRA, they can be considered ‘shadow banking’ entities under the Financial Stability Board's (FSB's) definition (Chapter 3). RFCs do undertake maturity transformation, and some are highly leveraged (RBA 2012a). Nonetheless, the RFC sector currently poses limited systemic risk given its small size and minimal connections to the regulated banking system. While there are currently over 300 RFCs, in aggregate, their share of total domestic financial system assets is small and has been declining over time (Graph 4.3). In addition, RFCs' overall borrowing from and lending to banks are each equivalent to less than 1 per cent of banking system assets. Within the RFC sector, a number of finance companies issuing unlisted retail debentures have failed over recent years, most notably property lender Banksia Securities Limited in late 2012. These failures have had no adverse impact on financial stability, although they have raised investor protection concerns.

4.2.1.3 Bank funding liabilities and structures

As explained above, banks are inherently exposed to liquidity shocks, such as funding market disruptions. An individual bank can experience a funding shortfall because the behaviours of its creditors are correlated – specifically, creditors can seek to withdraw their funds at around the same time. Creditors can also extrapolate bad news about one bank to other banks, perhaps due to similarities between bank balance sheets or business models, or general uncertainty about banks' prospects. Such correlation in creditor behaviour, in the form of runs on banks by depositors, has been a feature of private banking systems since their establishment (Gorton 2012).

Although bank liquidity crises have historically arisen from depositor runs, short-term wholesale funding, including its secured (e.g. repurchase agreements) and unsecured interbank forms, proved to be less stable than short-term customer deposits during the global financial crisis (IMF 2013). Apart from the existence of insurance for many retail deposits, wholesale creditors' behaviour may be more correlated than that of depositors because of their lower switching costs and the greater substitutes for their funds. The lower stability of short-term wholesale credit may also stem from creditors' procyclical risk appetite – in good times, when risk-taking is rising, wholesale creditors will tend to fund a bank at lower spreads, whereas in times of financial stress their appetite for bank credit risk can retreat rapidly (Gorton and Metrick 2012). In considering the liquidity risks associated with a bank's use of wholesale funding (and its funding composition more generally), it is important to consider the diversity and maturity of its funding. For instance, a bank could reduce its liquidity risk by regularly issuing longer-term wholesale debt and ensuring its maturity profile is spread over time (Stewart, Robertson and Heath 2013).

Since the onset of the financial crisis, Australian banks (like many banks internationally) have responded to market pressures to boost stable funding sources by increasing their deposit funding (for further discussion of bank funding composition and recent trends, see Chapter 5). Although the Australian banking system continues to source a high share of its funding from wholesale markets relative to a number of other banking systems, a relatively large share of Australian banks' wholesale debt has a remaining duration of longer than one year (Graph 4.4).^[10] As such, the share of short-term wholesale funding in Australian banks' total funding is not unusual. Indeed, Australian banks have increased the average maturity of their wholesale funding over recent years, and there are indications that recent issues have involved a wider range of investors than a few years ago. Increased diversity could lower a bank's funding risks if it reduces the correlation of creditor behaviour, all other things being equal.

Within the wholesale investor base, it is generally assumed that offshore wholesale creditors are more likely than domestic creditors to withdraw their funds during times of uncertainty or stress; such investor behaviour was demonstrated in the financial crisis, and the recent experience of some European countries shows that this can also apply to non-resident depositors (RBA 2012b). Consistent with this, Australian banks' use of offshore wholesale funding markets is often regarded by observers as a potential source of systemic risk, and periodic bouts of turbulence in global capital markets over recent years have created wholesale funding pressures for Australian banks.

As with other forms of offshore funding by non-financial sectors, a ‘sudden stop’ in offshore bank funding in the future – as occurred during the second half of 2008 – cannot be ruled out.^[11] In general, a lower exchange rate would be likely to be part of the adjustment mechanism to any disruptive event of this nature. There are a number of other possible adjustment mechanisms in response to a reduction in offshore demand (i.e. wider funding spreads) for Australian banks' paper, including slower credit growth, a change in the composition of bank funding, and large non-financial corporations directly accessing capital markets. Such adjustments need not be disorderly – indeed, the experience with Australian banks' lower use of offshore funding over the past few years supports this.

The majority of Australian banks' offshore funding is denominated in foreign currency, primarily US dollars. According to the 2013 Foreign Currency Exposure survey published by the ABS, Australian banks hedge more than 90 per cent of their foreign currency denominated debt securities, with the proportion of long-term debt being hedged even higher (Graph 4.5). Moreover, the maturities of the derivatives used to hedge against foreign currency risk are matched to the maturities of the underlying debt securities (Rush, Sadeghian and Wright 2013). Hedging foreign currency risk through derivatives removes exchange rate risk but exposes banks to the risk that the counterparty to the transaction will be unable to honour its obligations, forcing the bank to find a replacement swap or obtain foreign currency on the spot market at prevailing rates (IMF 2012b). Australian banks manage their counterparty credit risk exposure by entering into most transactions with large internationally active banks with high credit ratings, as well as by requiring bilateral collateralisation on transactions. Because derivative exposures increase Australian banks' interconnections with offshore banks, they are, in principle, a channel in which offshore financial shocks can transmit to the Australian banking system. However, collateralisation of these transactions, including the flows of collateral to Australian banks that occur if an offshore financial shock causes a depreciation in the Australian dollar, should ensure effective management of this risk.

4.2.1.4 International banking

Banks conduct business internationally either by undertaking cross-border transactions directly from their home location or by establishing foreign branch or subsidiary offices in foreign locations. Typical motivations for such activity include to assist their home country customers in international transactions and to pursue higher risk-adjusted returns than are available from solely domestic business (CGFS 2004). Although international expansion potentially increases and diversifies a bank's earnings, operating across borders and under different economic and market conditions may entail higher levels of financial risks, such as credit, market (including foreign exchange risk), and particularly operational risk. If these risks significantly materialise they could weaken the credit or liquidity position of the parent bank, and therefore pose systemic risk in the home market. This risk increases with the size and unfamiliarity of international exposures, as well as the strength of the interconnections (e.g. funding, legal and operational links) between the parent and its foreign offices. From a regulatory perspective, international banking activity can be harder for individual authorities to monitor (Song 2004), and problems with internationally active banks also tend to be more difficult for individual countries to resolve (IMF 2010a).

The large Australian banks are less internationally active on the asset side of their balance sheet than on the liability side, given their lending business is primarily focused on the domestic market, while some of this activity is funded offshore. A significant share of their international exposures are in New Zealand; all four major banks have sizeable operations there (Table 4.3). These tend to be similar to the major banks' operations in Australia, that is, focused on housing and business lending. The similarity of these operations to their Australian businesses, as well as their long-standing presence in New Zealand, alleviate the normally heightened operational, credit and market risks faced by banks operating in overseas jurisdictions. At the same time, the traditional banking focus in New Zealand, and the positively correlated performance of the Australian and New Zealand economies, mean that the majors receive little diversification benefit from these operations. As a practical example of how stress might transmit across the Tasman, the majors significantly increased their funding of their NZ subsidiaries during the global financial crisis prior to the announcement of government guarantees in both countries.

Domestic banks' international activities are not the only channel through which international banking can transmit financial stress. Foreign-owned banks have the potential to spread problems from their home country to the economy they operate in (Peek and Rosengren 2000). Again, the systemic risk posed by a foreign-owned bank will depend on its size in the host market and the strength of its interconnections with its parent bank. The strength of interconnections are likely to vary depending on whether the foreign-owned banks operate under a branch or subsidiary licence; foreign branches do not have a legal identity separate from the parent bank and thus tend to be more reliant on the parent for funding, lending and operational matters. That said, there have been numerous cases across different countries over recent years where foreign subsidiaries have encountered local funding difficulties as a result of severe problems at the parent bank.

In Australia, there have been some cases since the onset of the financial crisis where otherwise sound foreign-owned banking businesses operating in Australia have been destabilised by difficulties at a foreign parent (Laker 2010b). Even so, the systemic risk posed by external shocks transmitted to Australia by foreign branches and subsidiaries is limited by their small size. The largest foreign bank accounts for less than 2 per cent of domestic banking system assets, and foreign banks in aggregate only account for 12 per cent.

4.2.2 The housing market

The housing market can be considered to pose systemic risk because of its size and importance to the real economy, its interconnection with the financial system through banks' sizeable exposure to housing, and the concentration of risk in certain other players in the housing finance chain, such as lenders mortgage insurers (LMIs).

4.2.2.1 Bank lending

Banks and other ADIs provide almost all housing credit in Australia (about 97 per cent). While the banks' share of housing credit has varied a little in the post-Wallis period, it has increased following the crisis, as non-bank lenders were affected by the dislocation in residential mortgage-backed securities markets.

The low share of securitised housing credit and the (related) low penetration of non-bank housing loan providers are factors that contribute to Australian banks having a high share of housing loans in their asset portfolios. Aggregate housing loans represent around 40 per cent of total domestic banking system assets; the available data indicate that this is well above the share for many other large banking systems (Graph 4.6).^[12] In some overseas jurisdictions, public and private mechanisms exist that transfer housing loans off banks' balance sheets. These include government-sponsored mortgage insurance and support for mortgage securitisation, as well as legislative support for group structures that separate covered bond issuance and other banking activities. In these jurisdictions, the mortgage market could be just as large but less connected and important to the banking system. The Australian major banks' housing loan share is a little higher than the Australian industry total, at 45 per cent of their assets.

The sizeable exposure of the major banks and many other banks in Australia to loans secured by housing collateral represents a concentration risk in their asset portfolios. Were a significant part of the national housing market to suffer a severe downturn, banks' asset performance could deteriorate significantly. APRA's stress tests, which involve severe macroeconomic downturns featuring large rises in unemployment and falls in house price (Section 4.3.2), give some comfort about the banks' ability to withstand the direct credit losses they would incur on their housing exposures under such a scenario. Nonetheless, the impact on the banking system could still be severe if bank creditors became worried enough about a deterioration in banks' housing exposures to seek to withdraw their funds.

Although Australian banks' asset books have a high share of housing loans, concerns about such a concentration need to be weighed against the alternatives: the systemic risks posed by banks having different asset compositions and business models. Housing lending is rightly viewed as lower risk than many other forms of lending, including lending for commercial property investment and development, corporate lending, small business lending, and credit card and other personal lending. And as demonstrated during the global financial crisis and the euro area crisis, banks' non-loan assets have the potential to generate larger losses than housing loans.

Australian banks recorded very low loss rates on their housing loan portfolios during the financial crisis (Graph 4.7), as well as over previous decades. In contrast, loss rates on some other loan portfolios have been elevated at times. Although some housing loan losses have been covered by mortgage insurance, the large difference in loss rates primarily reflects differences in underlying risk. This pattern of relative loss rates is also evident in other countries and enshrined in international rules on bank capital requirements.

The RBA has previously discussed a number of industry risk management and regulatory factors that have contributed to the relatively strong performance of housing loans in Australia (for example, see RBA (2009) p 21). Of critical importance from a risk management perspective has been that banks have generally maintained prudent lending standards. While some more risky lending practices emerged in the years leading up to the financial crisis, these practices have been much less prevalent in recent years. By and large, banks have maintained a focus in their credit assessments on the borrowers' capacity to repay over the life of the loan, including by verifying borrowers' incomes and other financial obligations. Australian borrowers' ability to service their mortgages across the interest rate cycle has also been assisted by the practice of Australian banks applying an interest rate add-on to their lending rate when assessing borrowers' loan-servicing capacity; this practice is not always prevalent in other jurisdictions (APRA 2013b).

Lending that does not involve proper assessment of borrowers' capacity to repay the debt assumes that the borrower will be able to refinance or sell the property in the future to avoid default, and therefore that house prices will not fall (Ellis 2012). This sort of ‘asset-based’ financing became especially prevalent in the United States in the mid 2000s; widespread lending of this type can increase systemic risk by increasing the procyclicality of the housing market, as well as the correlation between banks' housing loan performance and the housing market.

A feature that distinguishes the Australian mortgage market from those in many other countries is that mortgages in Australia are predominantly variable-rate mortgages with the rate set at the lenders' discretion. This structural characteristic has a number of implications for risk:

• While the variable-rate mortgages offered in Australia expose households to ongoing interest rate risk, this necessitates that lenders (and borrowers) include in their approval process an assessment of whether it can be serviced at higher (assumed) interest rates.
• Compared to fixed-rate loans, variable-rate mortgages offer much greater flexibility in allowing households to prepay their debt. This is a particularly appealing feature for many owner-occupier borrowers in Australia because the interest payments on owner-occupier mortgages are not tax deductible (Ellis 2006). From a lenders' perspective, mortgage prepayment should act to reduce the credit risk on the mortgage. This is because borrowers have built up a financial buffer that can be drawn upon in difficult times (Stewart et al 2013); prepayment also means that banks' losses on any defaulted mortgages would be smaller than otherwise.
• Although households can guard against interest rate risk by taking out a fixed-rate mortgage, the typical terms used by borrowers in Australia are relatively short at between two and five years. If interest rates rise substantially during the term of the fixed-rate loan, households would be exposed to a negative repayment shock at its expiry. Such a large immediate repayment shock could be more difficult for households to adjust to than incrementally higher repayments, especially because the ability to build up a financial buffer by prepaying fixed-rate loans is contractually limited in Australia.^[13] The Australian situation is different from the US and Danish mortgage markets which are unique in that they provide borrowers with the ability to fix an interest rate over the life of the mortgage, yet still prepay their loan early without penalty (Frankel et al 2004).

4.2.2.2 The role of lenders mortgage insurers

Housing finance in Australia is facilitated by the use of mortgage insurance, which is a specialist type of insurance that protects mortgage lenders in the event that a borrower cannot repay their loan. In Australia, this cover is offered by prudentially regulated institutions known as lenders mortgage insurers. These institutions charge an upfront premium to lenders, which usually pass on the cost to borrowers.

Lenders generally use mortgage insurance for riskier loans, such as those loans originated with loan-to-valuation ratios (LVRs) 80 per cent. In addition to insuring individual loans, LMIs also insure pools of loans as a form of credit enhancement for securitisation, although the use of such policies has fallen in recent years. Overall, more than one-quarter of Australian housing loans are estimated to be covered by mortgage insurance.

Importantly, the use of LMI does not eliminate the systemic risk arising from a severe downturn in the housing market; it merely shifts some of the underlying credit risk to a different class of prudentially regulated entity. As discussed in RBA (2013b), the correlated nature of mortgage insurance policies means that LMIs would likely face substantial claims under these circumstances, potentially weakening their creditworthiness. Distress in the LMI sector could hinder the payment of claims to lenders, thus imposing losses on them, although the direct financial impact of this on the Australian banking system is unlikely to be substantial. However, there could be significant indirect effects. For example, without the ability to transfer risk to LMIs, lenders might be unwilling to write high LVR housing loans, which are disproportionately used by first home buyers. Any such pullback in housing credit would in turn affect the broader economy and confidence in the financial system.

It is possible that the use of mortgage insurance in housing finance could either reduce or increase systemic risk (BCBS, IAIS and IOSCO 2013). Mortgage insurance can help promote financial stability to the extent that it dampens swings in lending standards (and therefore the housing market cycle) – this might occur because LMIs' risk appetite is lower than some (marginal) lenders, or because they are a ‘second set of eyes’ in loan origination processes across the industry. On the other hand, lenders could respond to the use of mortgage insurance by relaxing lending standards because they believe the LMI is assessing the risk, or because they believe that any loss is an LMI loss.

The structure of the Australian LMI industry differs from other countries where mortgage insurance is used extensively, in that the mortgage insurers are all privately owned and operate without government guarantees (RBA 2013b). In addition to the widespread use of private mortgage insurance in the Australian mortgage market, the systemic importance of the largest LMIs in Australia is increased by the high concentration in the industry. There are only five active LMIs, of which two – Genworth Australia and QBE LMI – account for around three-quarters of the industry's annual premium pool. The other three LMIs are owned by major banks or other prudentially regulated lenders, which reduces the extent to which default risk is transferred out of the banking system.

4.2.2.3 The role of investors in the housing market

A feature of the Australian housing market is its extent of investor involvement. The share of occupied dwellings that are rented suggests that investors hold around a quarter of the housing stock (Graph 4.8). Investors account for a higher share of outstanding housing debt than this (over 30 per cent), reflecting their higher average gearing rate than owner-occupiers and possibly some ownership of holiday rental and second properties.

Substantial investor involvement in the housing market could have implications for systemic risk. Investor housing loans – over half of which are taken out on interest-only terms – tend to amortise more slowly than owner-occupier loans, as interest deductibility and negative gearing reduce the incentive for property investors to pay down their loans faster than required. These borrowers could therefore be more likely than owner-occupiers to experience negative equity during a sharp housing market downturn, and thus generate loan losses for lenders in the event of default. However, this risk is mitigated by the lower prevalence of investor loans with high LVRs at origination; in recent years, around 10 per cent of investor loan approvals had LVRs greater than 90 per cent, compared with over 15 per cent of owner-occupier loan approvals. Moreover, investors tend to have higher incomes and levels of (non-housing) wealth. Consistent with this, the performance of investor housing loans has historically been in line with that of owner-occupier housing loans (Graph 4.9).

A more indirect, but more likely, channel by which investors can pose systemic risk is through their influence on housing prices and housing price expectations. If investors' decisions to invest in housing are more correlated than those of owner-occupiers, then greater investor participation in the housing market could increase its procyclicality – that is, the amplitude of housing price cycles. This could occur because an investor's investment decision relies more heavily on relative expected returns on housing and alternative investments, and less on non-financial factors. Strong house price upswings might lead to unrealistic house price expectations by property purchasers, who respond by increasing leverage. It may also encourage more marginal (owner-occupier) borrowers to overstretch themselves in order to purchase a home before it is ‘out of reach’. These types of behaviours would increase the chance of a sharp downturn in housing prices and significant loan losses for lenders.

4.2.3 The commercial property market

The commercial property market poses significant systemic risk despite being smaller in size than the housing market. There are strong connections between the commercial property market and the banking system because a substantial portion of commercial property investment and development, including development of both residential and non-residential property, is financed by lending from banks. Moreover, the characteristics of the commercial property market and features of banks' loan contracts can mean that their commercial property portfolios are riskier than their housing loan portfolios, or even their general corporate loan portfolios. Indeed, banks' commercial property exposures have historically been one of the main sources of loan losses during episodes of banking sector distress internationally. This was the case in the US savings and loan crisis, banking crises in Japan, Scandanavia, and more recently, Ireland, Spain and the United Kingdom. In Australia, commercial property loans were a major contributor to sizeable banking system loan losses, and the failures of a few state government-owned and foreign-owned banks, during the early 1990s (Gizycki and Lowe 2000).^[14]

Ellis and Naughtin (2010), and Ellis, Kulish and Wallace (2012), among others, provide several reasons why fluctuations in the commercial property market propagate rapidly through financial institutions' balance sheets, particularly when compared to those for the housing market:

• Exposure to the construction cycle. Relative to the housing loan book, commercial property lending is more concentrated in financing new development. This type of lending tends to be more risky because the value of the project is not realised until completion, as well as being more cyclical.
• Commercial property developments are lumpy and can have long construction lags. Commercial property developments are large relative to existing stocks. Because they impact local supply, vacancy rates can remain high long after a downturn. The long construction phase means that there is a risk that demand falls before the completion of a property.
• Default is cyclical and borrowers face fewer disincentives to default. Commercial property investors face the risk of lost income during an economic downturn. If the value of the property has also fallen, such that they are in negative equity, borrowers might make themselves better off by defaulting. This is in contrast to an owner-occupier who still derives benefit from the property regardless of equity.
• Commercial property financing is procyclical. Lending is often short term and needs to be refinanced. Changes in market conditions can cause borrowers to breach covenants, making refinancing difficult.

4.2.4 Financial market infrastructures (FMIs)

FMIs are key components of the financial system, delivering services critical to the smooth functioning of financial markets, such as the settlement of payments and the clearing and settlement of transactions in securities and derivatives.

Well-designed and reliable FMIs can be a source of both financial stability and operational efficiency. Indeed, this has been the experience in Australia and internationally. As noted in Chapter 8, FMIs were a source of stability during the crisis, operating reliably throughout and retaining the confidence of market participants. FMIs act as a coordinating device, bringing a network of counterparties together to support trading liquidity and the netting of exposures and settlement obligations. They also establish secure arrangements for the timely clearing and settlement of obligations between counterparties; assist institutions in the management of counterparty credit risks; and help to coordinate actions in the event of a market participant's default.

Many of these benefits derive from the size and breadth of the network that an FMI controls. Accessing a large network of counterparties with a single operational connection delivers not only operational and informational efficiencies, but also greater scope for netting of offsetting exposures. Accordingly, there is a tendency towards a single FMI (or few FMIs) providing services in any given market (Pirrong 2011). This concentration can, however, be a source of systemic risk, since it creates a high level of market dependence on the single provider. Any shortcomings in the design or risk management framework of the FMI can, in the event of a financial or operational shock, have spillover effects across the system.

Given their typically large size, their lack of substitutability in the markets they serve, and strong connections with banks and other financial institutions, FMIs are generally systemically important. This means they require sound design, and high standards of operational and financial resilience. International policymakers have focused increasingly on FMI resilience in recent years, especially in light of the international initiative to expand the scope of central counterparty (CCP) clearing to OTC derivatives markets, discussed in Chapter 3 (Tucker 2013; Coeuré 2014).

FMIs have, however, long had a central role in the financial system. High-value payment systems grew out of early arrangements for the settlement of interbank claims. As central banks emerged as the providers of the ultimate settlement asset, operating these systems became an integral function of central banks (Norman, Shaw and Speight 2011). Central banks also have a natural interest in ensuring that payment and settlement systems operate safely and effectively, so as to underpin the role of money as a medium of exchange and to facilitate central banks' provision of liquidity to the financial system both in normal times and in periods of stress (CPSS 2005). CCPs also have a long history. They have existed for more than a century, initially clearing exchange traded derivative markets (Moser 1998), but more recently expanding to other asset classes. The need to ensure robust design, operational resilience and sound risk management has therefore always been there; the reforms have not created ‘new’ systemically important FMIs, but rather amplified their existing systemic importance.

Since FMIs necessarily concentrate payments, clearing and settlement processes, they need to manage carefully the risks that may arise from their activities. The remainder of this section discusses the potential systemic consequences if these processes are not well designed and associated risks not appropriately managed.

4.2.4.1 Sources of systemic risk in FMIs

Table 4.1 identified a number of indicators of systemic importance for banks: size, substitutability, interconnectedness and complexity. With the possible exception of complexity, these features generally also apply to FMIs. Table 4.4 considers these indicators for the key FMIs operating in Australia, which are discussed further in Chapter 8. There is typically only a single FMI providing payments, clearing or settlement services in each market or product class. Each FMI has a broad direct or indirect participant base and clears or settles a high value of financial market activity every day.

While, as noted in Section 4.1, financial risks may typically be either diversified or transferred away, this may not be feasible in the case of the risk a participant bears from its use of an FMI. Since many financial markets or product classes are served by a single FMI, a market participant wishing to undertake activity in a given market may have no choice but to accept the risks associated with participating in the FMI that serves that market. To the extent that such risks are a common source of vulnerability for participants, or that they may trigger contagion, the financial risks associated with an FMI's activities may become sources of systemic risk.

Of the sources of financial risk identified in ‘Box 4A: Types of Financial Risks’, the key risks to an FMI that could lead to contagion elsewhere in the system are credit, liquidity, operational and business risks (Manning, Nier and Schanz 2009).

• Credit risk: The risk that a participant in the system defaults on its obligations to the FMI, imposing direct unanticipated losses on other members. Certain variants of credit risk are also relevant:

  – In a securities settlement facility (SSF), a relevant credit risk is principal risk; this is the risk that one party fulfils its obligation to pay funds or deliver securities, while the other does not.

  – For CCPs, the relevant credit risk is replacement cost risk; that is, the risk that a counterparty defaults prior to the intended settlement date, requiring that the non-defaulting counterparty replace the trade – potentially on less favourable terms. Particularly in derivatives markets, pre-settlement periods may be long, and in some cases – for example, in interest rate swap contracts – there may be a number of interim cash-flow obligations through the life of the contract. The fundamental role of a CCP (discussed in more detail in Chapter 8) is to manage this source of risk.

• Liquidity risk: The risk that one or more participants – or where the FMI assumes risk as principal (e.g. a CCP), the FMI itself – has insufficient liquidity in the settlement asset to meet its obligations. This could disrupt the flow of liquidity in the FMI and, in some extreme circumstances, lead to delay in or failure of other participants to meet their obligations. In an SSF, an analogous concept is liquidity in the security to be delivered.
• Operational risk: The risk of losses arising directly or indirectly from technical failure or other forced interruption to the operations of an FMI.
• Business risk: The risk that the operator of an FMI suffers commercial losses that threaten its viability, leading to the suspension or termination of its services and flow-on disruption or losses for its participants.

Importantly, a participant default on obligations arising in a payment, clearing or settlement system will typically be the result of a credit event that arises outside of the system; the system then becomes the mechanism by which the loss is transmitted. As noted in the Wallis Report, ‘the payments system may be the transmission mechanism for systemic instability’ (Financial System Inquiry 1997, p 363). How widely losses are ultimately transmitted depends on the FMI's design and the role the FMI plays in the financial system.

The remainder of this section describes how shortcomings in the design of an FMI or an interruption to an FMI's services can give rise to flow-on losses in the financial system or disruption to other markets and services, thereby creating systemic risk through contagion. Regulation and oversight against robust risk management and operational standards, alongside well-articulated recovery and resolution arrangements, can help to mitigate the various risks described (discussed further below).

FMI design

A fundamental design feature in a payment system or SSF is its settlement model. A well-designed settlement model can be a source of stability, instilling confidence in the market that obligations will be met on a timely basis. The frequency of settlement is one aspect that has received considerable attention both domestically and internationally over the past two decades, and indeed featured in the Wallis recommendations.

• In high-value payment systems, it was increasingly appreciated in the 1980s and 1990s that a ‘deferred net settlement’ model – in which interbank obligations accumulate over a period and settle on a net basis – could be a source of systemic risk: if banks credit incoming payments to customer accounts before final interbank settlement has occurred, credit exposures could build up between banks, potentially triggering a cascade of defaults should one participant in the system fail (Humphrey 1986). Under the alternative of ‘real-time gross settlement’, by contrast, payments are settled individually with finality in real time intraday, eliminating the scope for such a build-up of unintended interbank credit exposures.
• In SSFs, the length of the settlement cycle gives rise to similar concerns. However, since securities are subject to market risk, the counterparties to the trade need to manage the associated replacement cost risk (via a CCP if one operates in the relevant market). The complexity of replacement cost risk management can increase with the length of the settlement cycle. It is also well understood that in a securities transaction non-simultaneous exchange of funds for securities (or in a foreign exchange transaction, one currency for another) is a source of principal risk.

An FMI's risk management framework is also central to its design. This is clearly illustrated in the case of a CCP. In contrast to a typical high-value payment system or SSF, a CCP assumes financial risk as principal (Chapter 8). As the legal counterparty to both the buyer and the seller of a given contract, the CCP maintains a balanced book: obligations due from the CCP to one party are fully covered by its receipts from the other. In the event that one party fails, however, the CCP must still fulfil any obligations to the surviving party. In such circumstances, the CCP must enter into a replacement contract with a new counterparty to rebalance its position; this may entail a financial loss. To manage its financial exposures, a CCP establishes arrangements for margining and risk mutualisation among participants.^[15] While such arrangements protect the CCP and provide essential clarity around participants' obligations, in extreme circumstances they could trigger flow-on liquidity or capital stress. Such procyclicality was identified more generally in Section 4.1 as a source of systemic risk. In the FMI context, procyclicality could arise even in the absence of default if, as market conditions deteriorated, a CCP increased margin rates or other collateral requirements (Rehlon and Nixon 2013).

Other potential channels for transmission of shocks have been identified, arising from observed and anticipated changes to the FMI landscape (Hermans, McGoldrick and Schmeidel 2013).

• Concentration in the provision of OTC derivatives clearing services. Smaller OTC derivative market participants may not be eligible, or may find it uneconomical, to participate directly in a CCP. Accordingly, they may seek access to CCPs as clients of direct participants that act as ‘clearing agents’. Provision of such clearing services is a scale business. Smaller market participants are therefore likely to channel their trades through a concentrated group of clearing agents that have the operational and balance sheet capacity to support such activity.
• Interdependencies between CCPs, or between CCPs and other FMIs. Links between CCPs may arise where CCPs establish arrangements between themselves that allow a participant of one CCP to trade with a participant of another. Such arrangements may lead to a build-up of credit exposures between the linked CCPs which if not appropriately managed could be a source of contagion and wider instability.

Interruption to FMI services

A lengthy interruption to an FMI's provision of services, for either operational or financial reasons, could cause widespread disruption and potential losses. For instance:

• In the event of an interruption to settlement in a high-value payment system, financial institutions would be unable to extinguish their obligations to each other, leading to the build-up of exposures between them. It would also disrupt central bank operations and, more broadly, the flow of liquidity between institutions. To the extent that other FMIs or retail payment systems relied on final settlement in the high-value payment system, the dislocation would have wider implications across the financial market infrastructure.
• Problems at an SSF or CCP could lead to the closure of underlying markets. An important benefit of CCP clearing is that it facilitates anonymous trading: the buyer and seller of a financial product do not need to monitor each other's capacity to meet settlement obligations; only the capacity of the CCP to do so. Should either a financial or operational shock disrupt a CCP's ability to stand between buyers and sellers, it is likely that the underlying market would cease to function, leaving market participants unable to establish new positions or manage existing exposures.^[16] Loss of confidence in an SSF's capacity to complete timely settlement could similarly interrupt activity in the underlying securities markets.

4.2.5 Insurers

Insurers – including general insurers and life insurers – assume the risk of financial loss from physical events, in exchange for an upfront premium.^[17] The use of insurance can support economic activities that might otherwise not be undertaken because the potential loss is deemed to be too large. By shielding household and business balance sheets against potentially severe financial loss, insurance may also contribute to financial stability.

Insurers pool and invest the premium revenue they receive from policyholders in financial assets, primarily highly rated sovereign, bank and non-financial corporate debt. They typically have long-term investment horizons reflecting that insurance claims may be paid out a number of years after premium is received. Insurers may therefore exert a stabilising influence on financial markets as they are less likely than other investors to sell their assets when prices are falling (ECB 2009b).

Although insurers are not immune from failure, the insurance business model has traditionally been viewed as relatively stable. Insurers' liabilities tend to be well diversified because they are often tied to a range of uncorrelated physical events. Moreover, the upfront funding of their liabilities protects insurers against the risk of liquidity shortfalls that are inherent in banking markets (IAIS 2011). Not being reliant on debt funding also reduces insurers' interconnections within the financial system, in turn lowering their contagion risk.

A failed insurer can typically be resolved in an orderly fashion; existing policy liabilities can be run off over time, limiting the potential for a ‘fire sale’ of their assets that may destabilise financial markets. The systemic impact of a failed insurer will partly depend on the extent to which other insurers are able to replace their services. For instance, insurance risks that are specialised or where remaining insurers already have high concentrations would likely be more difficult to replace than other insurance business. An example of substitutability problems in the insurance market was the failure of Australia's second largest insurer, HIH, in 2001. HIH had underpriced risks in some business lines that it dominated (e.g. builders warranty insurance), so competitors only re-entered the market with substantial premium rate increases, and the disruption to insurance supply imposed significant costs on parts of the economy, particularly the building industry (IAIS 2011; ‘Box 2A: The Collapse of HIH Insurance’).

Participation in non-traditional insurance and non-insurance activities – such as derivatives business – have the potential to significantly increase the systemic risk posed by an insurance group. This was demonstrated by AIG, the US insurance group, during the financial crisis. Because AIG was connected to many banks and other financial institutions through its provision of credit protection on structured credit products and its securities lending, its failure had the potential to cause cascading losses in the financial system (FCIC 2011). Reflecting the potential for greater systemic risk, non-traditional insurance business is the ‘impact factor’ that receives the highest weight (45 per cent of the total) under the International Association of Insurance Supervisors' (IAIS') recently released methodology for identifying global systemically important insurers (G-SIIs) (IAIS 2013).

The largest Australian insurers make up a small part of the global insurance market. Accordingly, no Australian insurers were among the nine insurers identified as G-SIIs by the FSB using the IAIS' methodology. From a domestic perspective, the systemic risk posed by the large Australian insurers is lessened by their traditional insurance business models and, with the exception of LMIs, limited connections to the rest of the financial system. For example, only 2 per cent of banks' funding is sourced from general and life insurers.

4.2.6 Managed funds

Managed funds invest and hold funds on behalf of their clients, in the process providing debt and equity funding to financial markets and other sectors. Managed funds can differ greatly in their investment strategies and the types of financial instruments they hold – for instance, some focus mainly on investing in short-term bank paper, whereas others invest in long-term infrastructure debt. Chapter 7 discusses the specific systemic risks posed by superannuation, which is the largest part of the managed fund industry in Australia, and the one that is subject to prudential regulation.

Managed funds can face liquidity risk. Liquidity transformation usually occurs because funds offer at-call or at-short-notice withdrawals to clients, while some assets may not be able to be sold quickly. This liquidity mismatch exposes funds to the risk of investor runs – where concerns about a fund lead to large-scale withdrawals by investors. If withdrawals exceed the amount of available cash, the fund may need to liquidate other assets at unfavourable prices, reducing the value of the units in the fund. This creates an incentive for individual investors to rush to withdraw while a fund is still liquid. If the fund is updating unit prices to properly account for losses caused by withdrawals, however, this is a weaker incentive than that faced by bank depositors (who can withdraw their funds at par while a bank remains liquid). Australian mortgage funds experienced high rates of withdrawals during 2008–09 and, in line with the requirements of the Corporations Act 2001, many suspended withdrawals due to insufficient liquidity. While the suspension of withdrawals may have preserved some of the unit value, it may also have contributed to the decline in popularity of the suspended funds and the broader mortgage fund industry, as investors came to better appreciate the risks involved. The risk of a run can be exacerbated when funds are more complex or opaque.

Although managed funds are connected with the rest of the financial system through their investments, it is unlikely that most individual managed funds would pose systemic risk because of their small size. However, high correlation in investment strategies across individual managed funds could magnify the impact of this sector on the financial system. This could occur because of ‘herding’ behaviour among fund managers – that is, where decisions to buy or sell a particular asset prompt other fund managers to do the same (Office of Financial Research 2013).

More complex individual managed funds could also give rise to financial instability. For example, hedge funds may pose risks to the financial system due to their use of leverage and other linkages to the banking system, and in some cases dominant positions in complex and less liquid financial markets (Kambhu, Schuermann and Stiroh 2007). In Australia, these more complex parts of the sector are relatively small and less leveraged than in Europe or the US. For example, hedge funds only account for about 3 per cent of the aggregate assets held by the managed funds industry in Australia. Australian hedge funds also have quite limited levels of leverage, suggesting that they pose little systemic risk to the financial system as a whole (ASIC 2013).

4.3 The Process for Monitoring Systemic Risk

The financial stability responsibilities of the Reserve Bank, APRA and the Australian Securities and Investments Commission (ASIC) require ongoing monitoring of systemic risk. This section sets out the processes used within the Reserve Bank to monitor systemic risk, and the interactions between this work and that done by other agencies.^[18]

4.3.1 Analysis at the Reserve Bank

Monitoring systemic risk requires the use of multiple sources of data, analytical frameworks and models. One major component of systemic risk monitoring at the Reserve Bank is analysis of balance sheets of households, businesses and financial intermediaries. The size of each sector's balance sheet indicates its potential impact upon other sectors, and balance sheet structure can indicate resilience to shocks. Importantly, monitoring is done at aggregated (sector and sub-sector) levels, and with disaggregated data. Such disaggregated analysis is important for the monitoring of systemic risk, as it allows the identification of risk concentrations that may be concealed by aggregate data. Financial distress often develops in the tails of the distribution of borrowers or lenders, and significant financial disruption is likely to manifest prior to the distress of the median or average case in a sector.^[19]

Given its importance as a source of systemic risk, the banking sector is monitored closely by the Reserve Bank. Much of this is based on confidential regulatory data collected by APRA.^[20] On the asset side, there is a focus on the composition of banks' lending books, given that some types of lending are riskier than others, as well as the repayment performance of the existing stock of loans. On the liabilities side, the amount and quality of banks' capital is a focus, as it is the primary buffer through which adverse shocks can be absorbed. Banks' other funding sources are also monitored closely: amounts of deposits, short-term and long-term debt used, and similarities in funding structures across banks, are major determinants of susceptibility to funding and liquidity crises, including widespread withdrawals of funding across the banking system. Analysis of banks' profitability – perhaps a more forward-looking measure of financial health – is also undertaken. Levels of direct exposure between banks, one indicator of the potential for contagion, are also considered (see, for example, Tellez (2013)). Analysis includes the banking system's international exposures, in order to obtain an overview of the risks banks face across the whole corporate group (RBA 2013a).

Many other indicators of risk are also examined, including financial market prices. Residential and commercial property prices are analysed in the context of market conditions, as well as changes in intermediaries' standards for lending against property, given the potential of such developments to harm financial stability. The size, growth and composition of the non-prudentially regulated share of the financial system is also monitored, as the global financial crisis has clearly shown the economic harm that can be caused by distress in sizeable ‘shadow banking’ systems (FSB 2011b).

In conjunction with these quantitative indicators, qualitative data on attitudes and behaviours in the financial system are monitored. A range of indicators, including investor behaviour and various dimensions of lending standards (that is, loan features and the loan approval process), are closely monitored for signs of excessive risk-taking and exuberance, as these add to procyclicality in the financial system. The Reserve Bank also watches for signs of governance problems, including conflicts of interest and other incentive problems, which can give rise to rent-seeking.^[21] These types of behaviours tend to emerge in more complex, opaque parts of the financial system, such as structured credit markets in the years before the crisis.

The global financial crisis spurred much new development of ‘direct’ quantitative measures of systemic risk that rely on sophisticated statistical or analytical techniques. The resulting measures can be grouped into those that attempt to measure systemic risk (incorporating assessments both of the likelihood of systemic crises and their impact), and measures that quantify individual institutions' contributions to systemic risk. This latter group of measures gives policymakers extra tools to identify institutions that should be more closely monitored or subject to higher levels of regulation. The total number of ‘direct’ quantitative measures has grown significantly in the wake of the crisis: Bisias et al (2012), for example, survey 31 different measures. While these are useful tools, the RBA does not place overt reliance on any one such measure. The lack of a proven track record in predicting or quantifying systemic risk makes exclusive reliance on a particular metric inappropriate at this stage. Further research may, over time, mitigate this concern.

4.3.2 Systemic risk monitoring at APRA

APRA takes an explicitly system-wide approach to the supervision of financial institutions. This is made operational through APRA's risk-based approach to supervision, specifically its Probability and Impacts Ratings System (PAIRS) and Supervisory Oversight and Response System (SOARS) (APRA 2012). The first of these combines an estimate of the probability of an institution's failure with an estimate of the likely impact of its failure on the financial system, giving both an equal weight, to provide an overall assessment of the risk it poses. SOARS ensures supervisory resources are concentrated on the institutions that receive the highest risk scores under this approach. In effect, this means that the largest banks in Australia have dedicated teams of supervisors, while a single person often supervises multiple smaller institutions.

This systemic approach is supported by APRA's thematic approach to risks in each of the industries it regulates. This involves both ‘horizontal reviews’ in which an individual institution's financial metrics are compared to those of its peers, as well as the maintenance of industry risk registers that facilitate the monitoring of issues APRA sees as involving risks for each industry as a whole. Examples of risks to the banking industry that have received attention in recent years include credit quality in commercial property lending and the effect of low interest rates on residential mortgage lending standards (Laker 2010b; APRA 2013b).

The macroeconomic stress tests that APRA has conducted since the early 2000s are another component of its systemic risk monitoring (Laker 2010a, 2012). These cover the vast majority of the Australian banking system by assets, and involve the application of a common scenario and common risk parameters (e.g. loss rates) across banks. In recent years, these stress tests have focused mainly on determining whether banks have adequate capital to withstand a severe macroeconomic downturn that also involves large falls in property prices.

Consistent with their overlapping roles in monitoring systemic risk, APRA and the Reserve Bank share much of their monitoring and analysis. This is done both through formal mechanisms such as the Council of Financial Regulators (CFR) and the RBA-APRA Coordination Committee, as well as on a less formal basis at the working level.

4.4 Policy Mitigants and Responses to Systemic Risk

Public policy does not seek to eliminate systemic risk – this would require removing certain financial risk-taking altogether, which would not be optimal for society. However, effective public policy design and other mitigation strategies can substantially reduce systemic risk. Effective responses to systemic risk events may also lower their impact on the financial system and the economy.

4.4.1 Sound macroeconomic policies

Sound macroeconomic policies – specifically, monetary, exchange rate and fiscal policies – underpin financial system stability. Monetary policy provides a nominal anchor to the economy. A policy that achieves a low and stable rate of inflation, while responding to current economic conditions, can help to promote financial stability. In particular, it can reduce the risk of unanticipated transfers of wealth between creditors and debtors, and ensure that price signals (including the price of risk) reflect economic costs.

Overly tight monetary policy redistributes financial wealth from debtors to creditors and can lead to increased defaults (corporate and household), falling asset prices (as the supply of assets for sale often increases), and a reduced appetite for new debt. This has occurred previously, during periods such as the Great Depression, and is thought to increase the financial fragility of firm and bank balance sheets and the risk of systemic financial instability (Eichengreen and Grossman 1994; Bernanke 1995).

Periods of excessively loose monetary policy can also lead to an increase in systemic risk. In this case, abnormally low interest rates can stimulate the demand for credit, boost asset prices and increase leverage in the economy. These developments tend to be accompanied by a rise in marginal lending and inadequate pricing of risk (that is, a misallocation of credit). Loose monetary policies, and the relaxation of lending standards, have been discussed as a contributing factor to the global financial crisis (see Woodford (2010) and the references cited therein).

A floating exchange rate regime can be an important element of financial system stability because it allows monetary policy to be directed at domestic circumstances. It can also induce domestic agents to actively manage foreign exchange risk, reducing the chance that foreign exchange mismatches build up in domestic financial portfolios and liabilities. Some jurisdictions with more constrained monetary policies, given their fixed exchange rate regimes or membership of currency unions, have pursued ‘macroprudential policies’ to deal with the resultant capital account volatility and credit expansion (Crowe et al 2013).

Sound fiscal policy is important for mitigating systemic risk. Unsustainable fiscal policies can create systemic risk because of the strong interconnections between sovereign and financial sector balance sheets. Concern about sovereign creditworthiness can induce losses on financial institutions' holdings of sovereign debt; it can also lower the value of collateral banks use to raise funding (CGFS 2011). Problems in the financial system have the potential to feed back to sovereign finances as tighter financial conditions adversely affect the economy. Adverse feedback loops between sovereign and bank balance sheets have been evident in the euro area over recent years.

Taxation policies that do not distort investment and risk-taking decisions also help to promote financial stability. Tax systems that provide incentives to engage in leveraged speculation, either through property or other financial assets, or that encourage a high level of debt when financing investment projects, can result in higher systemic risk. Prudent design of the tax system, including appropriate capital gains taxation, and depreciation allowances that are consistent with economic costs, can be helpful in this regard.

In addition to prevention, sound macroeconomic policy means there are more options available to policymakers if faced with a financial crisis. In particular, co-ordinated expansions in monetary and fiscal policy, when appropriately managed, can reduce the magnitude of a crisis and result in a faster economic recovery, which in turn benefits financial stability.

In sum, given the importance of sound macroeconomic policies for financial stability, and the significant costs of financial crises, the macroeconomic policy framework and settings in a particular jurisdiction can and should be managed with consideration of society's appetite for bearing systemic risk.

4.4.2 Prudential regulation and supervision

Financial markets are subject to consumer protection and market integrity regulation to promote fairness and efficiency. Prudential regulation is a more intense form of regulation that aims to minimise financial and systemic risk, and protect depositors, by ensuring a higher level of safety and confidence in certain financial services and products, appropriate to the nature of the promises embedded in those products. The Wallis Inquiry concluded that some level of prudential regulation should apply to those financial promises that are inherently difficult to honour, where the likelihood of the promise being met is difficult to assess and where a breach of the promise has significant adverse consequences for the recipient of the promise and the financial system.

All developed countries have in place well-established systems of prudential regulation. In Australia, APRA is charged with the prudential regulation and supervision of banks and other deposit-takers, general and life insurance companies, and the bulk of the superannuation fund sector. APRA's mandate is to ensure that, under all reasonable circumstances, the financial institutions that it oversees are able to honour their financial promises, within a stable, efficient and competitive financial system. Protecting the financial interests of depositors, policyholders and superannuation fund members (which together it defines as ‘beneficiaries’) and promoting the stability of the financial system are at the heart of APRA's mandate. It fulfils this mandate using a combination of standard-setting and supervision of the institutions it oversees, while at the same time maintaining an insistence that the primary responsibility for the safety and soundness of regulated institutions remains with their boards of directors and senior management. Although APRA's prudential activities are designed to ensure that institutions act in a manner that minimises the likelihood of financial losses to beneficiaries, they are not intended to eliminate failures entirely – to do so would be very burdensome for financial institutions and would obstruct competition and innovation.

4.4.2.1 Licensing and prudential standards

APRA is responsible for setting prudential standards and licensing institutions to conduct business,^[22] including requirements for capital, liquidity, governance, risk management and information systems. For ADIs, there is generally no distinction in these requirements between Australian-owned and foreign-owned ADIs that are locally incorporated. In contrast, there is a distinction in the requirements for foreign-owned banks that are licenced to operate in Australia as a branch. As per international practice, they are subject to their home regulators' standards on a consolidated basis. One of the key rules to reduce the risk to Australian depositors of these institutions is that they are prohibited from accepting initial deposits of less than $250,000 from individuals and non-corporate institutions.

The capital standards that apply to locally incorporated banks, other deposit-takers and insurers specify a minimum capital requirement that depends on the (measured) risk profile of their portfolios. At the heart of a risk-based regulatory capital framework is the principle that the amount of capital needed for a given activity should reflect the risk of that activity; it also recognises that there are incentives for financial institutions to boost returns by taking on greater risk for a given amount of capital employed (Byres 2012).

Commercial property loans are one example of a portfolio that has tended to experience greater losses during market downturns. Consistent with this, Australian banks' regulatory capital requirements for commercial property exposures can be much higher than for other forms of lending.

In addition to capital standards, prudential regulators generally require financial institutions to have specific policies to manage concentration risk – for example, concentrations in particular types of counterparties, industries, countries and asset classes. This reflects that risk concentrations are likely to increase the chance of correlated losses. For these reasons, APRA imposes prudential limits on banks' and insurers' exposures to both single unrelated counterparties and single related entities (usually termed ‘large exposure rules’). Because these single counterparties tend to be other banks or insurers, these measures also act to reduce interconnectedness in the financial system.

As discussed in Chapter 3, the international response to the global financial crisis centred on a regulatory reform agenda to address key sources of systemic risk. Implementation of Basel III and ‘too big to fail’ reforms has largely been the responsibility of APRA, with significant changes to APRA's prudential standards as a result.

4.4.2.2 Supervision

In addition to licensing financial institutions to conduct business and setting prudential standards, APRA is tasked with supervising institutions. Supervision in a narrow sense involves the continuous monitoring of individual institutions to ensure they comply with prudential standards, are in a sound financial condition and more generally conduct their affairs in a prudent manner. However, because the financial system is complex and dynamic, ensuring individual institutions meet prudential rules is alone not enough to safeguard financial stability. This has been evident in the financial crises that have occurred recently in those overseas jurisdictions that previously adopted a more ‘light touch’ approach to prudential supervision.

To effectively address systemic risk, supervision needs to be forward-looking (such as by assessing emerging risks for institutions) and have a system-wide focus (such as by analysing the way different parts of the system interact, or are likely to interact, with each other). Good supervision also involves a willingness to act when risks are identified, such as by taking timely and effective action, intruding on decision making and questioning common wisdom, even in the face of external criticism (IMF 2010b). Of course, identifying risks and acting on them requires the necessary legal authority and resources to be available to the prudential supervisor. A prerequisite to having all of these elements of good supervision is support from government and parliament (Littrell 2013).

APRA also liaises with supervisors in other jurisdictions, to assist with ongoing supervision of internationally active banks, as well as crisis management arrangements. One component of these relationships is participation in supervisory colleges for some foreign-owned banks operating in Australia. Given the importance of the major Australian banking groups for financial stability in Australia and New Zealand, banking supervisors (and representatives of other regulatory authorities) in both countries meet regularly through the Trans-Tasman Council on Banking Supervision.^[23] The Australian and New Zealand banking regulators are also legally required to consider financial stability in the other country in their regulatory actions.

4.4.3 Regulation and oversight of FMIs in Australia

In a similar vein, regulation and oversight promote best practice in the design and operation of FMIs with the aim of ensuring that they remain a source of strength and resilience in the financial system, rather than a source of instability. As the role of FMIs has expanded, most notably with the G20's commitment to centrally clear all standardised OTC derivatives, there is an increasing awareness internationally of the importance of high regulatory standards for FMIs (Chapter 3).

Historically, payment and settlement systems have been overseen by central banks. Central bank oversight emerged with the growing realisation that the safety and efficiency of such systems was integral to central banks' financial stability objective as well as their monetary operations (CPSS 2005). The primary regulatory role in relation to CCPs and, often, the broader functioning of SSFs has typically fallen to securities commissions, sometimes with accompanying oversight by the central bank.

Over recent years, in several jurisdictions central banks have assumed a more formal regulatory role in respect of CCPs and SSFs (jointly CS facilities) alongside securities commissions. A deeper role for central banks in the regulation or oversight of CS facilities recognises the systemic importance of these entities and the value of integrating this activity with the broader financial stability activities of the central bank. This shift took place in Australia following the Wallis Inquiry, with the Reserve Bank granted the responsibility for setting financial stability standards in the regulation of CS facilities, alongside ASIC's licensing responsibilities (Chapter 8).

The Principles for Financial Market Infrastructures (PFMIs), which set international standards for the design and operation of FMIs, were published by the Committee on Payment and Settlement Systems (CPSS) and the International Organization of Securities Commissions (IOSCO) in April 2012 (CPSS-IOSCO 2012) (Chapter 3 and Chapter 8). Alongside the PFMIs, CPSS and IOSCO also developed a set of ‘responsibilities’ for central banks, market regulators and other relevant authorities for FMIs. These responsibilities emphasise the consistent application of the PFMIs to systemically important FMIs, the power to enforce corrective action to ensure observance of the PFMIs, and the importance of cooperation among authorities responsible for particular FMIs.

Building on earlier CPSS work on oversight principles (CPSS 2005), the responsibility for cooperation among authorities promotes effective mutual assistance so as to avoid duplication of regulation and minimise the burden on the regulated entity. Accordingly, good cooperation is relevant both domestically and internationally. Domestically, cooperation between the central bank and securities commission is essential, given the typical joint responsibility for CS facilities. Clarity of roles, both to the authorities themselves and to regulated entities, is particularly important. In the Bank's view, the current arrangements in Australia deliver both good cooperation and clarity. As the cross-border reach of many FMIs' activities expands, international cooperation is also becoming crucial, giving host authorities both regulatory influence and access to information (Chapter 8).

4.4.4 Other measures to mitigate systemic risk

4.4.4.1 Other banking related measures

The ‘four pillars’ policy reflects the view that any mergers between the four major banks in Australia, or their takeover by a foreign entity, would be against the national interest and hence should not be allowed. The policy has been confirmed by successive Australian governments since the Wallis Inquiry in 1997; before this time it was a ‘six pillars’ policy additionally involving two insurers. In addition to its possible effects on competition in the banking sector, the policy also has financial stability effects. The four major banks are each systemically important due to their size and interconnectedness with other financial institutions. Mergers between these banks would increase systemic risks arising from concentrations in the banking sector and add to the complexity and costs of a potential bank failure. Additionally, foreign ownership of a major bank may lead to increased risk from greater exposure to international financial markets, or changes in the business models of Australian banks towards more complex and risky banking activities. On the other hand, however, mergers between major banks and large international banks could have the benefit of providing more diversified income streams, and additional sources of capital, that are not related to the performance of the Australian economy.

4.4.4.2 Consumer protection and market integrity measures

Some regulations aimed at the protection of consumers of financial services and the integrity of financial markets have the associated benefit of reducing systemic risk. These fall mainly within the ambit of ASIC, and include regulations underlying the licensing and conduct of credit and financial services providers, regulation of disclosure and conduct in financial markets, as well as regulations surrounding corporate governance and auditing. These rules act to address the issues of information asymmetry and agency problems, and by reducing the financial risk faced by many individual parties, can mitigate systemic risk.

Consumer credit rules are probably the example that has had the greatest impact on the evolution of systemic risk in Australia. Since the mid-1990s, these have required lenders to show that retail borrowers can reasonably be expected to repay loans without substantial hardship, including without having to sell underlying collateral.^[24] They explain part of the good performance of residential mortgages in Australia; indeed, mortgages originated by loan brokers, who were until 2009 not bound by these rules, performed worse than those made directly by large banks prior to that time (Battellino 2008). The partial absence of responsible lending rules of this type has been identified as a factor in the poor performance of residential mortgages in the United States (Gerding 2009).

ASIC licenses and supervises conduct in financial markets in Australia, including equity, derivatives and futures markets. ASIC's work in such markets is aimed at ensuring they are efficient and fair, and covers the information market participants must disclose and the prevention of manipulative practices. Such regulations improve the confidence of market participants and make it easier for businesses to raise needed funding.

4.4.4.3 Public communication

Public communication can be used to promote financial stability. It operates mainly by informing the risk perceptions of investors, financial analysts and the broader community, in order to influence behaviour and, ultimately, prevent future adverse financial system outcomes. Communication about financial stability is particularly important as a tool for non-regulatory central banks such as the Reserve Bank (and overseas peers, see Sveriges Riksbank (2014)). But it is also employed by central banks with regulatory powers, such as the Bank of England (Bank of England 2013).

The Reserve Bank's main platform for public communication about financial stability is the Financial Stability Review, which is published at the end of March and September of each year. Other forms of communication include speeches by senior Reserve Bank staff and articles in the quarterly RBA Bulletin. Other Australian regulatory authorities also use such communication to achieve their goals.^[25]

Sometimes public communication is used to address issues that represent a stability threat in the near term; the responses of Australian authorities to rapid increases in house prices and household indebtedness in the early 2000s are a good example of this.^[26] Probably more frequently, communication is used to highlight risks surrounding developments that, while not currently a threat, have the potential for adverse consequences in the longer term.

4.4.5 Crisis management and resolution

4.4.5.1 Liquidity support

The fundamental intermediation and liquidity provision functions of the financial sector mean it bears significant liquidity risk, and this is socially desirable. But the inherent susceptibility of banks and other financial institutions to runs and other panics (and the economic consequences of such financial distress) show there is a cost to this structure. It has been recognised since at least the 19th century that there is a role for central banks to supply extra liquidity to the market during periods of stress, and to act as a ‘lender of last resort’ for entities unable to obtain funding elsewhere (Goodhart 1988).

Providing liquidity support to the financial system during times of stress is a central part of the Reserve Bank's role. During the period 2007 through 2009, there were considerable frictions in the interbank market resulting from turmoil in global credit markets, and the Reserve Bank significantly increased its supply of liquidity to the system to accommodate greater demand. The Bank made some changes to its operating procedures for liquidity provision, but these were easily accommodated within its existing market operations framework (Debelle 2008). For example, the pool of eligible collateral was widened and operations were conducted at longer maturities. The fact that the Bank had dealt with a wide range of counterparties over an extended period also contributed to the effectiveness of its liquidity support during the crisis.

The provision of liquidity support by the central bank to an individual institution – rather than the entire system – is a related, but separate form of intervention which central banks can make. The widely accepted doctrine for such actions is Walter Bagehot's (1873) exhortation to central banks to ‘lend freely against good collateral at a high rate of interest’. Notable components are the availability of good collateral (inseparable from the solvency of the institution), and a high rate of interest, which is intended to act as both ex ante and ex post deterrence from reliance on this type of central bank support. However, lender of last resort operations are unlikely to be as simple as Bagehot's dictum suggests (Stevens 2008). Assessing the solvency of an institution and the quality of its collateral is likely to be very challenging in real time. In cases such as Australia, where supervision lies outside the central bank, a good flow of information from the prudential supervisor on the health of the institution requesting support is essential; this is something that would be facilitated in Australia by the close working relationship between the Reserve Bank and APRA. The Reserve Bank's role as the lender of last resort does not extend to supporting insolvent institutions;^[27] while such interventions should not be absolutely ruled out, they are decisions that can only be properly taken by governments (Stevens 2008; Lowe 2013).

4.4.5.2 Resolution of distressed financial institutions

Most of the policy measures discussed above are designed to reduce systemic risk by lowering the chance that a financial institution will fail, under reasonable circumstances. But this does not mean that financial institutions are immune from failure (nor should they be). Policy measures are therefore needed to ensure a failing financial institution can be resolved effectively.

Financial institutions that have failed cannot be liquidated in the same way that non-financial corporate entities can, because doing so could significantly disrupt the financial system and impose large costs on the economy. In other words, liquidation may increase systemic risk. Specifically, there is the potential for a liquidation to result in: economic costs and hardship arising from the loss of systemically important financial services (such as deposit and payment services); fire sales of financial assets that destroy value; investor panic; and the loss of confidence in otherwise sound financial institutions. Such disorderly outcomes could increase the possibility of ad hoc government support for a failing institution, which would not be consistent with the principle of responsibility.

Resolution frameworks therefore need to be designed in ways that allow non-viable institutions to exit the market in an orderly manner, while minimising the adverse costs to society. According to the FSB's Key Attributes (2011a), an effective resolution framework should extend to all types of financial institutions that may be systemically important or critical in the event of failure. They should also:

• ensure continuity of critical financial services, such as payment and banking services
• predefine protection of certain depositors and policyholders from financial loss
• have clear resolution responsibilities for authorities, and the legal powers to carry out these responsibilities
• avoid unnecessary destruction of value in the resolution process
• impose losses in a way that respects the hierarchy of creditor claims
• not rely on public solvency support, or create the expectation that such support will necessarily be available.

In addition to minimising systemic risk once a financial institution fails, a well-designed resolution framework can also lower ex ante systemic risk by incentivising those market participants that are well placed to assess risk to do so. As discussed in Chapter 3, resolution arrangements in Australia have been strengthened over recent years, and have recently been assessed by the FSB as generally consistent with international best practice (FSB 2013).

Depositor protection is another policy that provides both ex ante and ex post mitigation of systemic risk. In Australia, two mechanisms work directly to protect depositors in ADIs from loss: depositor preference and deposit insurance in the form of the Financial Claims Scheme.^[28] Depositor preference – the legislated seniority of depositors to other unsecured creditors in the event of ADI failure – is a long-standing feature of the Australian banking system. The Financial Claims Scheme was introduced in 2008 and currently provides cover for deposits up to $250,000 per person per ADI (Chapter 3).

Footnotes

See Group of Ten (2001), ECB (2009a), IMF, BIS and FSB (2009) and EU (2010) for similar definitions. [1]

See Kaufman and Scott (2003) and Billio et al (2011) for examples of definitions that do not emphasise harm to the real economy. [2]

See De Bandt, Hartmann and Peydro-Alcalde (2010) for a general discussion of contagion. [3]

Cerra and Saxena (2008) found that economic downturns were more severe and persistent if they were preceded by a banking crisis. Reinhart and Rogoff (2009) estimate that 13 systemic banking crises between 1977 and 2001 plus the Great Depression were associated with an average 7 per cent rise in unemployment (over 4.8 years) and a fall in output of 9.3 per cent (over 1.9 years). [4]

See Jacobs and Rayner (2012) and La Cava (2013) for attempts to estimate the credit supply channel in Australia. Bernanke and Gertler (1989) and Bernanke, Gertler and Gilchrist (1999) discuss the effects of declining creditworthiness of borrowers. Gertler and Kiyotaki (2010) and Adrian, Colla and Shin (2012) discuss the effects when it is lenders whose creditworthiness declines. [5]

See Christelis, Georgarakos and Jappelli (2011) for the US experience and Windsor, Jaaskela and Finlay (2013) for an examination of this channel in Australia. [6]

The International Monetary Fund (IMF) (2012a, p 5) came to the same conclusion in a paper produced as part of its 2012 Financial Sector Assessment Program review of Australia, stating that ‘defining domestic systemic importance for financial institutions is relatively straightforward in Australia … the four largest banking groups stand head and shoulders above the rest’. [7]

The IMF (2012a, p 19) commented about the major banks that ‘with similarities in business models and common exposures, shocks are likely to impact these banks in a similar way.’ [8]

See IMF (2012a) for one attempt at calculating the funding advantage the major banks potentially receive from implicit government support. [9]

Stewart et al (2013) also provide a comparison of Australian banks' funding structures to those in selected other countries. They highlight that differences in statistical definitions might contribute to the higher share of wholesale funding for Australian banks. For example, in Australia, certificates of deposit and offshore intragroup deposits are treated as short-term wholesale funding rather than deposit funding, based on an assessment of how sticky they would be during times of stress; this is not necessarily the case elsewhere. [10]

The abrupt withdrawal of foreign capital inflows to banks or other recipients is sometimes referred to as a ‘sudden stop’ (see Brei (2007), for example). [11]

Many banks in Australia have an additional exposure to housing collateral because small business loans are often secured by residential property. The four major banks' total residential mortgage exposure is estimated to be roughly 50 per cent of their assets once residentially secured small business loans are included. [12]

It is common practice for lenders in Australia (and many other countries) to charge break fees for early repayment of fixed-rate loans, or prepayment of fixed-rate loans each year over a specified amount. These fees aim to recoup the economic cost of the bank hedging its interest rate risk over the life of the loan. [13]

Rising property prices also featured in the run-up to banking crises in Australia in the 1890s and 1930s, driving speculation and a view that ‘one couldn't lose money by investing in land’ (Fisher and Kent 1999). [14]

Margin is a ‘defaulter pays’ tool and covers both observed market movements and, to a high level of confidence, potential future price moves. To provide a further buffer in more extreme market circumstances, a CCP also maintains a pool of additional financial resources, comprising a mix of the CCP's own capital and participant contributions. This is a pre-funded mutualised resource. Should its pre-funded financial resources be depleted, the CCP may make additional calls on surviving participants. [15]

Fortunately, there are relatively few examples of CCP failures. Hills et al (1999) cite three instances: the failure of Caisse de Liquidation in 1974, the Kuala Lumpur Commodity Clearing House in 1983, and the Hong Kong Futures Guarantee Corporation in 1987. All three instances involved shortcomings in the risk management arrangements of the CCPs and all involved some interruption to activity in underlying markets. [16]

Life insurers also commonly offer investment policies that are more akin to funds management business than insurance; references to life insurers in this section refer to true ‘risk’ insurance business. [17]

This section draws heavily on APRA and RBA (2012). [18]

See Ellis (2010) for an example of misleading aggregate data in a financial stability context. [19]

The Reserve Bank and its staff are required to preserve the confidentiality of such data in accordance with the requirements of the Australian Prudential Regulatory Authority Act 1998 and the Reserve Bank Act 1959. [20]

Rent-seeking can be roughly defined as manipulating the environment to obtain a private benefit without adding economic value to society, although in practice such activity is usually detrimental to others. [21]

More generally, institutions providing financial products (whether they be licenced by APRA or not) are also subject to ASIC's financial services licensing regime. [22]

See CFR (2013) for more information. [23]

The relevant rules are the National Consumer Credit Code (which was introduced in 2009), and its predecessor, the Uniform Consumer Credit Code (which was introduced in 1996). [24]

For examples, see Price (2013) and Tanzer (2013). [25]

See Bloxham, Kent and Robson (2010) for a case study that focuses on the Reserve Bank's efforts in the early 2000s to draw attention to the risks associated with large, ongoing increases in housing prices and household borrowing. [26]

These arrangements were included in the 2010 Statement on Conduct of Monetary Policy and reaffirmed in the 2013 version. [27]

The Financial Claims Scheme also covers policyholders of licenced general insurers. [28]

Box Footnotes

Malz (2011) provides a detailed overview of financial risks. [1]

References

Adrian T, P Colla and HS Shin (2012), ‘Which Financial Frictions? Parsing Evidence from the Financial Crisis of 2007 to 2009’, NBER Macroeconomics Annual 2012, 27(1), pp 159–214. Available at <http://www.jstor.org/stable/10.1086/669176>.

APRA (Australian Prudential Regulation Authority) (2012), ‘Probability and Impact Rating System’, June.

APRA (2013a), ‘Domestic Systemically Important Banks in Australia’, Information Paper.

APRA (2013b), ‘Loan Serviceability Standards in Housing Lending’, APRA Insight, Issue 2, pp 40–54.

APRA and RBA (Reserve Bank of Australia) (2012), ‘Macroprudential Analysis and Policy in the Australian Financial Stability Framework’, September.

ASIC (Australian Securities and Investments Commission) (2013), ‘The Australian Hedge Funds Sector and Systemic Risk’, Report 370, September.

Bagehot W (1873), Lombard Street: A Description of the Money Market, 1912 reprint, Smith, Elder & Co, London.

Bank of England (2013), Financial Stability Review, November.

Battellino R (2008), ‘An Update on Household Finances’, RBA Bulletin, November, pp 76–84.

BCBS (Basel Committee on Banking Supervision) (2004), ‘Bank Failures in Mature Economies’, BCBS Working Paper No 13, April.

BCBS, IAIS (International Association of Insurance Supervisors) and IOSCO (International Organisation of Securities Commissions) (2013), ‘Mortgage Insurance: Market Structure, Underwriting Cycle and Policy Implications’, Bank for International Settlements, April.

Bernanke B (1995), ‘The Macroeconomics of the Great Depression: A Comparative Approach’, Journal of Money, Credit and Banking, 27(1), pp 1–28.

Bernanke B (2010), ‘Causes of the Recent Financial and Economic Crisis’, Statement before the Financial Crisis Inquiry Commission, Washington, DC, 2 September.

Bernanke B and M Gertler (1989), ‘Agency Costs, Net Worth, and Business Fluctuations’, The American Economic Review, 79(1), pp 14–31.

Bernanke B, M Gertler and S Gilchrist (1999), ‘The Financial Accelerator in a Quantitative Business Cycle Framework’, in JB Taylor and M Woodford (eds), Handbook of Macroeconomics, Volume I, Elsevier, pp 1341–1393.

Billio M, M Getmansky, AW Lo and L Pelizzon (2011), ‘Econometric Measures of Connectedness Systemic Risk in the Finance and Insurance Sectors’, Journal of Financial Economics, 104(3), pp 535–559.

Bisias D, M Flood, AW Lo and S Valavanis (2012), ‘A Survey of Systemic Risk Analytics’, Office of Financial Research Working Paper No 0001.

Bloxham P, C Kent and M Robson (2010), ‘Asset Prices, Credit Growth, Monetary and Other Policies: An Australian Case Study’, RBA Research Discussion Paper No 2010-06.

Brei M (2007), ‘The Impact of Sudden Stops on Bank Lending: Are there Cross-Sectional Differences?’, Social and Economic Studies, 56(4), pp 21–46.

Byres W (2012), ‘Regulatory Reforms – Incentives Matter (Can We Make Bankers More Like Pilots?)’, Remarks to the Bank of Portugal Conference on ‘Global Risk Management: Governance and Control’, Lisbon, 24 October.

Cecchetti SG, M Kohler and C Upper (2009), ‘Financial Crises and Economic Activity’, NBER Working Paper No 15379.

Cerra V and SC Saxena (2008), ‘Growth Dynamics: The Myth of Economic Recovery’, The American Economic Review, 98(1), pp 439–457.

CFR (Council of Financial Regulators) (2013), ‘Trans-Tasman Council on Banking Supervision’. Available at <http://www.cfr.gov.au/about-cfr/financial-distress-planning-management/trans-tasman-council-on-banking-supervision.html>.

CGFS (Committee on the Global Financial System) (2004), ‘Foreign Direct Investment in the Financial Sector of Emerging Market Economies’, Report, March.

CGFS (2011), ‘The Impact of Sovereign Credit Risk on Bank Funding Conditions’, Report, July.

Christelis D, D Georgarakos and T Jappelli (2011), ‘Wealth Shocks, Unemployment Shocks and Consumption in the Wake of the Great Recession’, Centre for Financial Studies Working Paper No 2011/27.

Coeuré B (2014), ‘Risks in Central Counterparties (CCPs)’, Speech to the Policy Panel during the Conference, ‘Mapping and Monitoring the Financial System: Liquidity, Funding and Plumbing’, Washington, DC, 23 January.

CPSS (Committee on Payment and Settlement Systems) (2005), ‘Central Bank Oversight of Payment and Settlement Systems’, May.

CPSS and IOSCO (International Organization of Securities Commissions) (2012), Principles for Financial Market Infrastructures, April.

Crowe C, G Dell'Ariccia, D Igan and P Rabanal (2013), ‘How to Deal with Real Estate Booms: Lessons from Country Experiences’, Journal of Financial Stability, 9(3), pp 300–319.

De Bandt O, P Hartmann, JL Peydro-Alcalde (2010), ‘Systemic Risk in Banking: An Update’, in AN Berger, P Molyneux and JOS Wilson (eds), The Oxford Handbook of Banking, Oxford University Press, Oxford, pp 633–672.

Debelle G (2008), ‘Market Operations in the Past Year’, Speech to FTA Congress, Melbourne, 31 October.

ECB (European Central Bank) (2009a), ‘Special Feature B: The Concept of Systemic Risk’, Financial Stability Review, December, pp 134–141.

ECB (2009b), ‘Special Feature E: The Importance of Insurance Companies for Financial Stability’, Financial Stability Review, December, pp 160–168.

Eichengreen B and R Grossman (1994), ‘Debt Deflation and Financial Instability: Two Historical Explorations’, unpublished manuscript, University of California at Berkeley.

Ellis L (2006), ‘Housing and Housing Finance: The View from Australia and Beyond’, RBA Research Discussion Paper No 2006-12.

Ellis L (2010), ‘Remarks to Finsia Financial Services Conference’, Address to Finsia Financial Services Conference – Meeting of the Minds: Our Global Agenda, Sydney, 20 October.

Ellis L (2012), ‘Prudent Mortgage Lending Standards Help Ensure Financial Stability’, Address to the Australian Mortgage Conference, Sydney, 23 February.

Ellis L (2013), ‘Stability, Efficiency, Diversity: Implications for the Financial Sector and Policy’, Address to the Paul Woolley Centre for the Study of Capital Market Dysfunctionality Conference, Sydney, 18 October.

Ellis L and C Naughtin (2010), ‘Commercial Property and Financial Stability – An International Perspective’, RBA Bulletin, June, pp 25–30.

Ellis L, M Kulish and S Wallace (2012), ‘Property Market Cycles as Paths to Financial Distress’, in A Heath, F Packer and C Windsor (eds), Property Markets and Financial Stability, Proceedings of a Conference held by the Reserve Bank of Australia and the Bank for International Settlements, Sydney, pp 109–132.

EU (European Union) (2010), Regulation (EU) No 1092/2010 of the European Parliament and of the Council: European Union Macro-Prudential Oversight of the Financial System and Establishing a European Systemic Risk Board, 24 November.

FCIC (Financial Crisis Inquiry Commission) (2011), ‘The Financial Crisis Inquiry Report’, Final Report, January.

Financial System Inquiry (1997), Financial System Inquiry Final Report (S Wallis, chairperson), Australian Government Publishing Service, Canberra. Available at <http://fsi.treasury.gov.au/content/FinalReport.asp>.

Fisher C and C Kent (1999), ‘Two Depressions, One Banking Collapse’, RBA Research Discussion Paper No 1999-06.

Fostel A and J Geanakoplos (2013), ‘Reviewing the Leverage Cycle’, Cowles Foundation Discussion Paper No 1918.

Frankel A, J Gyntelberg, K Kjeldsen and M Persson (2004), ‘The Danish Mortgage Market’, BIS Quarterly Review, March.

FSB (Financial Stability Board (2011a), Key Attributes of Effective Resolution Regimes for Financial Institutions, October.

FSB (2011b), ‘Shadow Banking: Scoping the Issues’, April.

FSB (2013), ‘Thematic Review on Resolution Regimes’, April.

Gerding EF (2009), ‘The Subprime Crisis and the Link between Consumer Financial Protection and Systemic Risk’, Florida International University Law Review, 22 May.

Gertler M and N Kiyotaki (2010), ‘Financial Intermediation and Credit Policy in Business Cycle Analysis’, in BM Friedman and M Woodford (eds), Handbook of Monetary Economics, Volume 3, pp 547–599.

Gizycki M and P Lowe (2000), ‘The Australian Financial System in the 1990s’, in D Gruen and S Shrestha (eds), The Australian Economy in the 1990s, Proceedings of a Conference held by the Reserve Bank of Australia, Sydney, pp 180–215.

Goodhart C (1988), The Evolution of Central Banks, MIT Press, Cambridge, Massachusetts.

Gorton G (2012), Misunderstanding Financial Crises: Why We Don't See Them Coming, Oxford University Press, New York.

Gorton G and A Metrick (2012), ‘Securitized Banking and the Run on Repo’, Journal of Financial Economics, 104(3), pp 425–450.

Group of Ten (2001), ‘Report on Consolidation in the Financial Sector’, January.

Hermans L, P McGoldrick and H Schmeidel (2013), ‘Central Counterparties and Systemic Risk’, European Systemic Risk Board Macro-prudential Commentaries, Issue No 6, November.

Hills B, D Rule, S Parkinson and C Young (1999), ‘Central Counterparty Clearing Houses and Financial Stability’, Bank of England Financial Stability Review, June.

Hoggarth G, J Hooley and Y Korniyenko (2013), ‘Which Way do Foreign Branches Sway? Evidence from the Recent UK Domestic Credit Cycle’, Financial Stability Paper No 22.

Humphrey DB (1986), ‘Payments Finality and Risk of Settlement Failure’, in A Saunders and L White (eds), Technology and the Regulation of Financial Markets, Lexington Books/Salomon Brothers Center Series of Financial Institutions and Markets, New York, pp 97–120.

IAIS (International Association of Insurance Supervisors) (2011), ‘Insurance and Financial Stability’, November.

IAIS (2013), ‘Globally Systemic Important Insurers: Initial Assessment Methodology’, July.

IMF (International Monetary Fund) (2010a), ‘Resolution of Cross-Border Banks—A Proposed Framework for Enhanced Coordination’, June.

IMF (2010b), ‘The Making of Good Supervision: Learning to Say “No”‘, IMF Staff Position Note, May.

IMF (2012a), ‘Australia: Addressing Systemic Risk Through Higher Loss Absorbency—Technical Note’, IMF Country Report No 12/311.

IMF (2012b), ‘Australia: Financial System Stability Assessment’, IMF Country Report No 12/308.

IMF (2013), ‘Chapter 3: Changes in Bank Funding Patterns and Financial Stability Risks’, Global Financial Stability Report: Transition Challenges to Stability, October, pp 105–148.

IMF, BIS (Bank for International Settlements) and FSB (2009), ‘Guidance to Assess the Systemic Importance of Financial Institutions, Markets and Instruments: Initial Considerations’, Report to the G-20 Finance Ministers and Central Bank Governors, October.

Jacobs D and V Rayner (2012), ‘The Role of Credit Supply in the Australian Economy’, RBA Research Discussion Paper No 2012-02.

Kambhu J, T Schuermann and K Stiroh (2007), ‘Hedge Funds, Financial Intermediation, and Systemic Risk’, Federal Reserve Bank of New York Economic Policy Review, December, pp 1–18.

Kaufman GG and KE Scott (2003), ‘What is Systemic Risk, and Do Bank Regulators Retard or Contribute to It?’, The Independent Review, 8(3), pp 371–391.

Kindleberger CP and RZ Aliber (2011), Manias, Panics and Crashes: A History of Financial Crises, Palgrave Macmillan, Houndmills.

La Cava G (2013), ‘Inventory Investment in Australia and the Global Financial Crisis’, RBA Research Discussion Paper No 2013-13.

Laker J (2010a), ‘The Australian Banking System Under Stress?’, Address to the Australian Business Economists Conference, Sydney, 9 June.

Laker J (2010b), ‘Supervisory Lessons from the Global Financial Crisis’, AB+F Leaders Lecture Lunch, Sydney, 8 December.

Laker J (2012), ‘The Australian Banking System Under Stress – Again?’, AB+F Randstad Leaders Lecture, Brisbane, 8 November.

Littrell C (2013), ‘Macro Prudence vs. Macro-Prudential Supervision’, Speech at the 3rd Annual Risk Day Conference, Sydney, 22 March.

Lowe P (2013), ‘Some Tensions in Financial Regulation’, Address to the Institute of Global Finance Second Conference on Global Financial Stability and Prosperity, Sydney, 4 July.

Malz M (2011), Financial Risk Management: Models, History and Institutions, John Wiley & Sons, Hoboken.

Manning M, E Nier and J Schanz (2009), The Economics of Large-value Payments and Settlement: Theory and Policy Issues for Central Banks, Oxford University Press, Oxford.

Moser JT (1998), ‘Contracting Innovations and the Evolution of Clearing and Settlement Methods at Futures Exchanges’, Federal Reserve Bank of Chicago Working Paper No WP-98-26.

Norman B, R Shaw and G Speight (2011), ‘The History of Interbank Settlement Arrangements: Exploring Central Banks’ Role in the Payment System', Bank of England Working Paper No 412.

Office of Financial Research (2013), ‘Asset Management and Financial Stability’, September.

Peek J and S Rosengren (2000), ‘Collateral Damage: Effects of the Japanese Bank Crisis on Real Activity in the United States’, The American Economic Review, 90(1), pp 30–45.

Pirrong C (2011), ‘The Economics of Central Clearing: Theory and Practice’, International Swaps and Derivatives Association Discussion Paper Series, No 1, May.

Price J (2013), ‘The Regulator's Perspective on the Regulation of SMSFs’, Speech to The Institute of Public Accountants National Congress, Gold Coast, 8 November.

RBA (Reserve Bank of Australia) (2009), Financial Stability Review, September.

RBA (2012a), ‘Box D: A Closer Look at the Shadow Banking System in Australia’, Financial Stability Review, March, pp 69–72.

RBA (2012b), ‘The Global Financial Environment’, Financial Stability Review, September, pp 5–18.

RBA (2013a), ‘Box A: Australian Bank Activity in Asia’, Financial Stability Review, March, pp 36–38.

RBA (2013b), ‘Box C: Lenders Mortgage Insurance’, Financial Stability Review, September, pp 39–42.

Rehlon A and D Nixon (2013), ‘Central Counterparties: What Are They, Why Do They Matter and How Does the Bank Supervise Them?’ Bank of England Quarterly Bulletin, Quarter 2, pp 1–10.

Reinhart CM and KS Rogoff (2009), ‘The Aftermath of Financial Crises’, The American Economic Review, 99(2), pp 466–472.

Rush A, D Sadeghian and M Wright (2013), ‘Foreign Currency Exposure and Hedging in Australia’, RBA Bulletin, December, pp 49–58.

Song I (2004), ‘Foreign Bank Supervision and Challenges to Emerging Market Supervisors’, IMF Working Paper No WP/04/82.

Stevens G (2008), ‘Liquidity and Lender of Last Resort’, RBA Bulletin, May, pp 83–91.

Stewart C, B Robertson and A Heath (2013), ‘Trends in the Funding and Lending Behaviour of Australian Banks’, RBA Research Discussion Paper 2013-15.

Sveriges Riksbank (2014), ‘The Riksbank's Work to Prevent Financial Crises’. Available at <http://www.riksbank.se/en/Financial-stability/The-Riksbanks-work-to-prevent-financial-crises/>.

Tanzer G (2013), ‘The Australian Hedge Funds Sector and Systemic Risk’, Speech at the AIMA Australia Hedge Fund Forum, Sydney, 10 September.

Tellez E (2013), ‘Mapping the Australian Banking System Network’, RBA Bulletin, June pp 45–54.

Tucker P (2013), ‘Central Counterparties in Evolving Capital Markets: Safety, Recovery and Resolution’, Banque de France Financial Stability Review, No 17, April, pp 179–184.

Windsor C, J Jaaskela and R Finlay (2013), ‘Home Prices and Household Spending’, RBA Research Discussion Paper No 2013-04.

Woodford M (2010), ‘Inflation Targeting and Financial Stability’, Keynote address to the Future of Central Banking Conference, Rome, 30 September.