Systemic risk

The GFC highlighted that focusing on the soundness of individual institutions, without stepping back to consider the overall financial system, is not sufficient to ensure financial stability. Externalities and spill-over effects can mean that, even where an individual institution’s capital, risk exposure and liquidity look sound, systemic risks can be building up. Yellen described it as “akin to caring for an entire ecosystem rather than individual trees”.31

The issue is that, if a regulatory system is primarily focused on institutions, it can be hard to both monitor and respond to systemic risk. This is particularly true when data on institutions outside the prudential regulatory perimeter is scarce. Australia’s regulatory framework deals with this issue by being designed to enable a focus on the system as a whole, yet allowing for tools to be applied to individual institutions to achieve systemic goals.

Preliminary assessment


APRA, the RBA and ASIC have a well-established system for monitoring systemic risk in Australia.32 APRA considers system-wide issues as part of its approach to supervising financial institutions. The RBA complements this with its own analysis of the financial system, with an eye to emerging systemic risks. The two agencies coordinate and share their analysis with each other and other agencies through the CFR, as well as through day-to-day contact between the agencies. The RBA’s semi-annual Financial Stability Reviews make this analysis available to the broader public. ASIC’s monitoring of FMI and non-prudentially regulated financial service providers contributes to identifying systemic risks outside the prudential perimeter.

APRA’s mandate for pursuing financial system stability was codified in the 2006 amendment to the APRA Act, which required the regulator “to promote financial system stability in Australia”. The RBA takes its mandate to promote financial stability as implied under the Reserve Bank Act 1959. Financial stability is also explicitly included in the Statement on the Conduct of Monetary Policy agreed between the Treasurer and the Governor of the RBA, most recently in October 2013.33

Significant reforms to improve the transparency of over-the-counter (OTC) derivatives since the GFC strengthen the monitoring of these instruments. Uncertainty about the derivatives exposures of institutions in the GFC negatively affected market confidence, increased perceptions of counterparty risk and made resolution of institutions with large derivatives books difficult. Reforms have included work to: standardise OTC derivative contracts; mandate that some OTC derivatives be traded on platforms and cleared through central counterparties; and record transactions on trade repositories.

As markets continue to develop, financial and technological innovations are emerging rapidly. These developments result in new products and services to enable consumers to manage and conduct their financial activities. Allowing activity outside regulatory perimeters encourages innovation and competition. However, new products can bring new risks. This was highlighted by the GFC, where the proliferation of new and complex asset-backed securities structures, such as collateralised debt obligations, contributed to risk transparency problems in the United States.

History tells us that the next financial crisis never looks quite like the last.34 It is not possible to know for certain where systemic risks will arise; however, it is perhaps more likely that they will emerge outside the prudentially regulated part of the financial system, such as in the shadow banking sector, where oversight is more limited. It is therefore important that monitoring arrangements are adequate for identifying risks wherever they might develop. This will require regulators to have adequate access to data.35

Global regulators have expressed their concern about the increasing use of shadow banking and its potential to generate systemic risks.36 The CFR recently considered developments in shadow banking in Australia, concluding that risks to stability remain limited. This is consistent with the size of the shadow banking sector as a share of financial sector assets declining substantially since the GFC (Chart 5.2). However, the CFR still recommended ongoing monitoring of emerging risks outside the perimeter.

Chart 5.2: Shadow banking sector(a)

Share of financial system assets, end December

This chart shows the shadow banking sector in Australia as a share of total financial assets from 2003 to 2013. The share was around 14 per cent from 2003 to 2007 but following the GFC has steadily declined to be around 7 per cent in 2013.

(a) Based on the FSB’s definition of ‘broad’ and ‘narrow’ estimates of shadow banking. Estimate excludes assets that are part of prudentially-regulated banking groups and assets related to equity trusts and self-securitisation.

Sources: ABS, APRA, RBA.

Current tools to address systemic risks are mostly applicable within the prudential perimeter, as discussed in the next section. This could be problematic if such risks are identified in other parts of the financial system, since the most effective way to manage these risks may be to subject the relevant institutions or activities to heightened regulatory and supervisory intensity. This is a gap in the current arrangement.

Macroprudential powers


A number of jurisdictions have implemented new macroprudential toolkits to assist with managing systemic risks. The effectiveness of these for a country like Australia is not yet well established, and there are significant practical difficulties in using such tools.

In conducting prudential policy, APRA has an eye to both the soundness of individual institutions and the financial system more broadly. APRA’s existing prudential tools to deal with emerging systemic risks include supervising particular institutions or lines of business more intensively, and changing capital or other prudential requirements for individual institutions. As part of the Basel III implementation, APRA can also activate a counter-cyclical capital buffer for the whole deposit-taking industry.

Communication is important as it can mould risk perceptions and affect risk-taking behaviour.37 In normal times, the RBA primarily relies on its public communications — such as speeches and the semi-annual Financial Stability Review — to highlight its concerns about any build-ups in systemic risk and potentially shape the expectations and actions of other parties. The RBA also takes any such risks into consideration in formulating monetary policy. In addition, in a crisis, the RBA can supply the financial system with liquidity and is the lender of last resort.

Unlike some other jurisdictions, Australia has not explicitly introduced macroprudential tools in the wake of the GFC.38 These tools are designed to address build-ups of systemic risk. They are typically applied in a counter-cyclical fashion at a system-wide level, rather than tailored to different institutions. Often, they take the form of limits on loan-to-value ratios (LVRs) in mortgage lending and maximum debt-servicing-to-income ratios.

Another example is the provision in Basel III to allow time-varying capital requirements for banks. This builds up capital in good times which can be run down during stress periods. This is similar to the dynamic provisioning used by countries such as Spain before the GFC, where banks build up larger provisions for impaired assets during good times, in recognition that regular provisioning under international accounting rules often does not reflect the actual experience in a downturn.

Some studies suggest that these tools can be of benefit in the right circumstances.39 Their purpose is to manage systemic risks by, for example, curbing the excesses in asset price inflation, risk appetite or credit growth. In doing so, they aim to limit the incidence and severity of financial crises in the future. However, the RBA’s submission notes that evidence on the effectiveness of such tools is still limited, especially on their application in advanced economies.

There are practical difficulties to using macroprudential tools.40

  • Deciding when to use macroprudential tools involves identifying unsustainable trends such as asset price or credit bubbles and build-ups of risk in the financial system. Although the problem is usually obvious in hindsight, identifying these trends ex ante has proven extremely difficult for regulators, central banks and investors globally.
  • In many ways, the macroprudential tools being proposed and implemented today are not too different from the quantitative restrictions used in Australia prior to financial deregulation in the 1980s. That experience showed that such restrictions can have large negative effects on efficiency and cause financial activity to move to the unregulated part of the system; for example, the shadow banking sector.
  • Many macroprudential tools are focused on the housing market, such as LVR caps, but will have little effect on other possible sources of systemic risk. Having the right tools to address all potential sources of risk would require granting a substantial degree of power to the responsible regulator.
  • Whereas monetary policy is relatively transparent and predictable, with a clearly articulated target, the targets of macroprudential policy are more numerous and less clearly defined. This makes it harder for the public to predict the use of these tools, adding additional uncertainty to the financial system. It may also send conflicting policy signals; for example, if monetary policy is being loosened while maximum LVR ratios are being lowered.

Box 5.2: Macroprudential tools in other countries

Approaches to using macroprudential powers vary significantly among countries. In some instances they are used by countries as a substitute for independent monetary policy — often the result of a managed exchange rate regime.

Approaches include:

  • Authorities in Singapore have focused on measures to limit house price inflation. These include mortgage LVR caps, mortgage tenure restrictions, increased property stamp duty and increased land allocated to residential development.41
  • The Reserve Bank of New Zealand (RBNZ) has identified four macroprudential tools that it may use: the (Basel III) counter-cyclical capital buffer; adjustments to the minimum core funding ratio; sectoral capital requirements; and restrictions on high-LVR residential mortgage lending.42 Concerned that the housing markets posed a growing threat to financial stability, in 2013 the RBNZ restricted the portion of banks’ new residential mortgage lending that could have an LVR greater than 80 per cent.
  • In the United Kingdom, the Financial Policy Committee (FPC) can issue directions on sectoral capital requirements and set the counter-cyclical capital buffer, and has asked for the power to impose a time-varying leverage ratio.43 It can recommend other changes to the United Kingdom Treasury or other regulators to address systemic risks, including changes to regulation such as when existing rules are out of date or activity has moved beyond the regulatory perimeter.
  • China has intensive state involvement in the financial sector, both through state-owned banks and its regulatory regime. Authorities use caps on LVRs, credit growth ceilings, counter-cyclical capital requirements, reserve requirements, taxes and property ownership limits.44

Policy options for consultation

Assess the prudential perimeter

Along with needing to identify risks outside the prudential perimeter, regulators must also be able to respond to those risks. In the current system, where the majority of the tools available to address systemic risks reside with APRA, and can only be applied within the prudential perimeter, this may be problematic. The Inquiry welcomes stakeholder views on the most appropriate way to ensure such risks can be managed.

One option would be to allow for some permeability of the prudential perimeter. That is, on the rare occasion a systemic risk is identified outside the prudential perimeter, some mechanism allows for affected institutions or activities to be brought within APRA’s remit and subjected to more intensive regulation and supervision. This may include when an institution or activity becomes of such size, interconnectedness, complexity or market importance that it poses a risk to overall system stability.45

Currently, legislation must be passed to enact such a change. This has the advantage of strong accountability, but can also take a significant amount of time. A more timely option could be to allow the appropriate Minister to designate institutions or activities to be brought into APRA’s purview, on systemic risk grounds, on advice from the RBA or CFR. This may be advantageous if significant delays in addressing the risk may let it build up further. Any such mechanism would need a high threshold for activation, with clear processes to ensure accountability for the decision.

Heightened regulatory and supervisory intensity for activities that pose systemic risk is not out of step with international practice. Some jurisdictions are setting thresholds for defining systemically important institutions. The Dodd-Frank Act authorises the Financial Stability Oversight Council to issue rules to require prudential supervision of systemically important non-bank entities or FMI entities.

The Inquiry seeks views on the costs, benefits and trade-offs of the following policy options or other alternatives:

  • No change to current arrangements.
  • Establish a mechanism, such as designation by the relevant Minister on advice from the RBA or CFR, to adjust the prudential perimeter to apply heightened regulatory and supervisory intensity to institutions or activities that pose systemic risks.

The Inquiry seeks further information on the following areas:

  • Is new legislation the most appropriate mechanism to adjust the prudential perimeter to respond to systemic risks, or could a more timely mechanism be of benefit? What alternative mechanisms could be used?
  • What accountability processes would be necessary to accompany any new mechanism?
  • What criteria could determine when an institution or activity was subject to heightened regulatory and supervisory intensity?

Additional macroprudential powers

APRA and the RBA have argued on a number of occasions that APRA’s existing toolset is adequate to undertake macroprudential supervision.46 APRA has the ability to vary its intensity of supervision, prudential standards and capital requirements through the economic cycle. Although APRA has not publicly stated under what circumstances it will use it, the counter-cyclical capital buffer enshrined in Basel III is also available. The RBA does not have specific macroprudential tools in the traditional sense, but it does extensively use public communications to address concerns and manage particular risks.

For two decades, Australia has effectively navigated systemic risk without the kinds of tools being introduced in some countries. Although there is no guarantee this will always be the case, Australia should be cautious regarding unproven tools while empirical evidence of their effectiveness remains limited.

The Inquiry would value views on the costs, benefits and trade-offs of the following policy options or other alternatives:

  • No change to current arrangements.
  • Introduce specific macroprudential policy tools.

The Inquiry seeks further information on the following areas:

  • Are there specific macroprudential tools that Australia should adopt to manage systemic risk?
  • What agency or agencies should have these macroprudential tools?

Stress testing

In its recent Financial Sector Assessment Program, the IMF recommended that Australian regulators enhance their stress-testing capabilities. In particular, that APRA devotes more resources to stress testing and the RBA develops the capability to undertake in-house macroeconomic stress tests.

A number of other jurisdictions make extensive use of stress testing. In some cases, such as in the United States, stress tests are a key input into supervision, helping to assess institutions’ compliance with prudential standards. Enhanced capabilities could aid APRA in supervising institutions, and assist both APRA and the RBA in monitoring systemic risk. Given the differences in their mandates and areas of focus, having both agencies conduct independent stress tests would provide a useful check and balance on the process. It would also encourage cross-agency discussion and deeper analysis to understand any differences in outcome.

However, robust stress testing has a resource cost for both regulators and industry. During stress testing, industry participants are typically required to provide data or respond to questions — some large banks involved in United States and European Union stress tests reported up to 100 staff members being used in the process.47 Stress testing is likely to be on the lower end of the cost spectrum compared to other options for improving financial stability.

The Inquiry would value views on the costs, benefits and trade-offs of the following policy options or other alternatives:

  • No change to current arrangements.
  • Australian regulators make greater use of stress testing with appropriate resourcing.

Box 5.3: A high-level scenario analysis

The FSI Secretariat, in conjunction with the Treasury and the RBA, conducted a qualitative scenario analysis as a simple stress test. The test was a useful tool to examine systematically, at a high level, the effect of several hypothetical financial shocks on the financial system. It did not examine the robustness of individual institutions’ balance sheets to these shocks or the adequacy of capital.

This work suggested that financial and economic shocks would have to be severe, and several would need to occur at once, to threaten the viability of the Australian financial system. This is similar to the conclusions made by APRA and the IMF in 2012 when they performed their own stress tests of the banking system.48

The exercise found:

  • A prolonged and pronounced deflation in housing prices would have a direct effect on the capital adequacy of the banking system, impairing the ability of banks to intermediate funds and leading to a large economic adjustment. Weaker economic outcomes could lead to a further deterioration of bank balance sheets, which would exacerbate already impaired bank lending conditions.
  • A severe shock to international financial markets that reduces or removes access to foreign funding for Australian financial institutions would sharply increase the interest rates on bank lending and significantly reduce bank lending across the economy. This would in part be offset by an easing in monetary policy. Weaker economic outcomes could increase non-performing loans and deteriorate banks’ asset quality, which would further increase the cost, and reduce the availability, of credit.
  • An offshore growth shock from a large trading partner would largely have an indirect effect on the financial system — weaker economic outcomes could lead to an increase in non-performing loans and a deterioration of bank balance sheets. The magnitude of the effect would depend largely on the degree to which financial imbalances had built up prior to the shock.

The scenarios considered mainly affected the financial system in two dimensions: through access to and management of liquidity and through compromising the capital base of the banks. However, many of the regulatory measures put in place since the North Atlantic financial crisis, as well as some of the measures and mechanisms financial institutions have implemented themselves, have served to increase the resilience of the financial system to shocks since the GFC.

31 Yellen, J 2009, Linkages between Monetary and Regulatory Policy: Lessons from the Crisis, presentation to the Institute of Regulation & Risk, North Asia, November, Hong Kong.

32 International Monetary Fund (IMF) 2012, Australia: Financial System Stability Assessment, IMF Country Report No. 12/308, IMF, Washington DC, November.

33 RBA and Treasury 2013, Statement on the Conduct of Monetary Policy, 24 October 2013.

34 Edey, M 2011, Basel III and Beyond, speech to Basel III Conference, 24 March, Sydney.

35 See the Regulatory architecture chapter for discussion of regulator data access.

36 The FSB has a workstream focused on the shadow banking sector — see Financial Stability Board (FSB) 2013, Global Shadow Banking Monitoring Report 2013, FSB, Basel.

37 Australian Prudential Regulation Authority (APRA) and Reserve Bank of Australia (RBA) 2012, Macroprudential Analysis and Policy in the Australian Financial Stability Framework, APRA and RBA, Sydney.

38 Aside from the counter-cyclical capital buffer contained in Basel III.

39 Lim, C, Columba, F, Costa, A, Kongsamut, P, Otani, A, Saiyid, M, Wezel, T and X Wu 2011, Macroprudential Policy: What Instruments and How to Use Them? Lessons from Country Experience, IMF Working Paper WP/11/238, IMF, Washington DC.

40 RBA 2014, First round submission to the Financial System Inquiry, contains a fuller discussion of some of the difficulties with operationalising these kinds of macroprudential tools.

41 Menon 2013, Securing Price Stability as Singapore Restructures, address to Asian Bureau of Finance and Economics Research Opening Gala Dinner, 21 May, Singapore.

42 RBNZ 2014, Macro-prudential policy — FAQ, RBNZ, Wellington, viewed 18 June 2014.

43 Bank of England (BoE) 2013, Macroprudential policy at the Bank of England, Bank of England Quarterly Bulletin, Quarter 3, BoE, London.

44 Lim, C, Columba, F, Costa, A, Kongsamut, P, Otani, A, Saiyid, M, Wezel, T and X Wu 2011, Macroprudential Policy: What Instruments and How to Use Them? Lessons from Country Experience, IMF Working Paper WP/11/238, IMF, Washington DC.

45 For example, see the proposals in Financial Stability Board (FSB) 2014, Assessment Methodologies for Identifying Non-Bank Non-Insurer Global Systemically Important Financial Institutions, consultative document, FSB, Basel.

46 See Australian Prudential Regulation Authority (APRA) and Reserve Bank of Australia (RBA) 2012, Macroprudential Analysis and Policy in the Australian Financial Stability Framework, APRA and RBA, Sydney; Ellis, L 2012, Macroprudential Policy: A Suite of Tools or a State of Mind?, speech to Paul Woolley Centre for Capital Market Dysfunctionality Annual Conference, 11 October, Sydney; and Edey, M 2012, Macroprudential Supervision and the Role of Central Banks, remarks to the Regional Policy Forum on Financial Stability and Macroprudential Supervision, 28 September, Beijing.

47 Moody’s 2013, Stress Testing: European Edition, Risk Perspectives, vol. 1, September, Moody’s Corporation, New York.

48 See the following publications: International Monetary Fund (IMF) 2012, Australia: Financial System Stability Assessment, IMF Country Report No. 12/308, IMF, Washington DC, and Laker, J 2012, The Australian banking system under stress — again?, AB+F Randstad Leaders Lecture 2012, 8 November, Brisbane.