Financial Stability Review – October 2016 Box C: Recent Developments in Australian Banks' Capital Position and Return on Equity
A bank's capital represents its ability to absorb unexpected losses; all else equal, the higher its capital, the lower the risk that a bank might become insolvent. Ensuring that banks maintain adequate capital is therefore central to reducing risks to financial stability and macroeconomic performance, given the large negative effects that bank failures, or even just fears of bank failure, can have on the real economy. However, high capital levels do not ensure a stable and resilient banking system on their own. Regulators also need to take into account a range of factors affecting banks' risk profiles and, in turn, the likelihood that banks will experience unexpected losses.
The global financial crisis revealed that banks in many countries were not holding enough capital for the risks they were taking. In response, the Basel Committee on Banking Supervision (BCBS) introduced internationally agreed requirements for higher and better-quality capital for banks globally under the Basel III framework. It included a new minimum requirement for Common Equity Tier 1 (CET1) capital – the highest quality form of capital – as well as regulatory capital buffers and a non-risk-weighted leverage ratio. More recently, the BCBS has been finalising these post-crisis regulatory reforms by reviewing the calculation of risk-weighted assets. The ongoing implementation of these measures has contributed to a material rise in bank capital globally and a reduction in return on equity (ROE), an important measure of profitability. This box outlines the recent history of capital ratio trends for Australian banks and current challenges associated with a decline in ROE.
The Recent Capital Reform Agenda in Australia
The Australian Prudential Regulation Authority (APRA) has been considering recommendations from the 2014 Financial System Inquiry, which have been endorsed by the Australian Government, in determining its approach to bank capital. One of these recommendations was that domestic capital standards be set so that Australian banks are ‘unquestionably strong’, so that banks remain resilient and continue to extend credit following an adverse shock, and that investors maintain their confidence in the Australian banking sector. Another recommendation was that the average mortgage risk weight under the internal ratings-based (IRB) approach to credit risk be increased to narrow the gap between the mortgage risk weights of banks using their own risk-weight models and those using standardised mortgage risk weights. This recommendation was intended to address concerns about competition in the mortgage market.
APRA implemented higher risk weights on Australian mortgages measured under the IRB approach on 1 July 2016.[1] Given that housing loans account for around two-thirds of total loans at the major banks, the increase in risk weights is expected to have a large effect on their CET1 ratios, reducing them by an estimated 0.7 to 1.1 percentage points, all else equal. This measure also has implications for the relative amount of equity funding these banks use for different types of lending. For example, the change in mortgage risk weights is estimated to increase the ratio of equity funding used for housing lending compared with business lending from just over one-quarter to a little under half.
Reforms to the international capital framework that are due to be finalised this year will also influence domestic standards. The BCBS has proposed a number of measures aimed at reducing excessive complexity and variability in IRB risk weights across banks, motivated by evidence that some banks were calculating markedly different risk weights for exposures with similar characteristics. These measures include restrictions on modelling risk weights for some exposures that the BCBS considers cannot be accurately modelled, as well as potentially adopting capital floors for the IRB approach relative to the standardised approach. In addition, the risk weights used under the standardised approach are being reviewed. While these changes are not intended to raise aggregate capital requirements significantly, APRA has indicated that it would be prudent for Australian banks to continue to plan for the likelihood of strengthened capital requirements in some areas.[2]
Banks' Response
In response to these developments, the major banks have significantly increased their capital ratios. The major banks' Tier 1 capital ratio was about 12 per cent at June 2016, around one and a half times the level during the global financial crisis, and their CET1 ratio has increased by almost 2 percentage points since this metric was introduced in 2013 (Graph C1). At 10 per cent of risk-weighted assets, the major banks' CET1 ratio is well above the standard regulatory requirement, and a buffer is expected to be maintained even after taking into account the effect of higher mortgage risk weights in the second half of the year.
The way that capital ratios have risen differs between the period since 2015 and the eight years prior. Over the earlier period, much of the increase was due to a reduction in average risk weights as the composition of banks' portfolios shifted towards mortgage lending (which tends to attract a lower risk weight than lending to businesses).[3] As a result, the leverage ratio was unchanged between 2010 and 2015, when risk-weighted capital ratios strengthened by around 2½ percentage points. In contrast, the more recent increase in the major banks' capital ratio has largely been due to an increase in capital; the major banks have raised around $20 billion of new equity and an additional $7 billion from retained earnings since the start of 2015. This has clearly increased the leverage ratio.
The recent strengthening of capital positions has improved the major banks' standing relative to international banks, even as the positions of international banks have also trended higher.
Assessing the capital strength of banks across jurisdictions is difficult because national regulatory authorities apply the Basel III international framework in different ways. However, APRA released a study in early 2015 that provided capital ratios of the major banks that could be compared with those of a large number of international banks as at June 2014. The study highlighted APRA's conservative application of the international capital framework, with the major banks' aggregate CET1 ratio around 300 basis points higher when reported on an internationally comparable basis. An update of the study showed that, as at December 2015, the major banks' combined CET1 capital ratio had moved into the top quartile of international banks, predominantly as a result of the capital raised in the second half of 2015. The major banks' combined leverage ratio also improved to be around the median. If the major banks want to maintain their relative position they will likely need to continue increasing their capital ratios given the upward trend in capital ratios globally.
Return on Equity and Adjusting to Lower Leverage
ROE is an important measure for assessing the profitability of a bank and its various divisions. Higher capital levels directly reduce ROE because the share of equity funding is greater for a given return on assets. The decline in leverage of Australian banks has therefore contributed to a fall in ROE of late, which has been compounded by lower profits due to a decline in income and an increase in bad debt charges (Graph C2).
Higher capital levels are expected to have a persistent effect on ROE. Indeed, analysts' expectations are for Australian banks' ROE to remain on average around 12½ per cent over the next couple of years. While this is high by international standards and appears to be above banks' cost of equity, it is lower than the returns to which Australian banks and their investors have become accustomed.
In theory, investors might be expected to accept the lower ROE that results from higher capital levels. This is because the reduction in leverage reduces volatility and risk in returns. If investors do accept lower returns, banks could adjust their target ROE lower. However, investors' expectations may not adjust immediately and banks may feel pressured to maintain historical levels of ROE.
One way that the major banks have so far responded to the reduction in ROE has been by repricing their loan books. Most lenders increased their standard variable housing rates by 15–20 basis points in the second half of 2015 after the announcement of higher risk weights on Australian mortgages, although some of this has since been offset by increased discounting for new loans. Another response has been for some banks to reduce their focus on divisions that have had lower returns than more traditional activities in the Australian market. This has resulted in some pullback from international portfolios: NAB divested its UK Clydesdale subsidiary earlier in the year and ANZ announced that it would narrow its focus in the Asian region by reducing low-return or higher-risk assets. Banks have also sought to divest underperforming parts of their wealth management portfolios recently.[4]
These responses will support ROE but they may not be enough to offset the impact of lower leverage. It would be a concern if banks were to attempt to restore their ROE to historical levels by taking on additional risk or by weakening the quality of their risk culture or governance. It will be important to continue monitoring how higher capital levels and lower ROE affect banks' incentives and behaviour in the period ahead.
Footnotes
See RBA (2015), ‘Box C: The Regulatory Capital Framework for Residential Mortgages’, Financial Stability Review, October, pp 52–55. [1]
See APRA (2016), ‘International Capital Comparison Update’, APRA Insight Issue Two. [2]
In addition, the major banks transitioned to IRB risk weights over this period, which lowered their risk-weighted assets relative to the Basel I standards. For more information, see Byres W (2014), ‘Seeking Strength in Adversity: Lessons from APRA's 2014 Stress Test on Australia's Largest Banks’, AB+F Randstad Leaders Lecture Series, 7 November. [3]
For more information on the major banks' wealth management activities, see Golat T (2016), ‘Banks' Wealth Management Activities in Australia’, RBA Bulletin, September, pp 53–59. [4]