Gareth Vaughan argues there are bigger issues at play behind ANZ NZ's censure by the RBNZ, well beyond calling for John Key to go as the bank's chairman

Gareth Vaughan argues there are bigger issues at play behind ANZ NZ's censure by the RBNZ, well beyond calling for John Key to go as the bank's chairman

By Gareth Vaughan

The recent call for former Prime Minister John Key to resign or be removed as chairman of ANZ New Zealand after the bank was censured by the Reserve Bank may make a great headline, but it masks bigger issues.

The call came from ex-BNZ chairman Kerry McDonald. It was made after the Reserve Bank revealed it has revoked ANZ's accreditation to model its own capital requirements for operational risk, citing a persistent failure in controls and the director attestation process at the country’s biggest bank that date back five years. Key joined ANZ's board in October 2017 and became chairman in January 2018.

For me calling for the heads of Key, ANZ CEO David Hisco - now on extended sick leave  - or anyone else overlooks two key questions.

Is the internal ratings based (IRB) capital approach New Zealand's big four banks use fit for purpose? 
And; Is the Reserve Bank's bank director attestation regime fit for purpose?

Both the IRB capital approach and bank director attestation regime arguably outsource regulation to the regulated.

This means we have a situation where there's a combination of a DIY regulatory capital regime, and an idiosyncratic, light touch regulator as highlighted in the International Monetary Fund's 2017 Financial Sector Assessment Program report on New Zealand. The IMF pointed out the Reserve Bank's broad statutory objectives requiring it to promote the maintenance of a sound and efficient financial system have allowed it to develop "a particular hands-off supervisory philosophy."

The Reserve Bank's three pillar approach to prudential regulation consists of self, market, and regulatory discipline. The self-discipline pillar relies on directors’ attestations to the fact that their bank has adequate risk management systems in place. There's no direct Reserve Bank access to bank records or files, and very limited on-site interaction with banks.

The Reserve Bank is, of course, currently undertaking a high profile and controversial review of banks' regulatory capital requirements. A key aspect of this proposes major changes to the IRB capital approach, which the big four banks - ANZ, ASB, BNZ and Westpac - are fighting tooth and nail against.

The Reserve Bank reviewed its director attestation regime in 2017 after the IMF urged it to more rigorously test director attestations. This saw the prudential regulator commission a report from Deloitte, which the Reserve Bank ultimately released to interest.co.nz after I went to the Ombudsman. The Deloitte report calls for a more prescriptive approach to oversight of bank risk management.

Fit for purpose or a relic from the 1990s?

The attestation regime was introduced in 1996. This was part of a regulatory shift in emphasis by the Reserve Bank away from detailed prudential regulation to a greater emphasis on market disciplines. The requirement for all directors to sign quarterly attestations was described as "perhaps the single most important part of the [bank] disclosure provisions." What this means in practice is bank directors are required to attest to, i.e. sign-off on, the accuracy of information contained in their banks' quarterly or half-yearly disclosure statements. 

Thus, as the Reserve Bank has put it, directors are responsible and accountable for the integrity of bank financial reporting. Hence when the Reserve Bank talks of "persistent failure in controls and the director attestation process" at a bank it's a serious matter. The Reserve Bank argues attestations are important because they strengthen the incentives for directors to oversee, and take ultimate responsibility for, the sound management of their bank. If a disclosure statement is found to be false or misleading, directors can face potential criminal and civil penalties including a jail term of up to three years.

However, there are no specific rules around how a bank must meet attestation requirements, with the Reserve Bank accepting the attestations without auditing the process. 

Banks' compliance requirements have changed out of sight, mushrooming dramatically, since 1996. For example the IRB capital approach, of which there'll be more shortly, was introduced for the big four banks in 2008. The Byzantium world of the Anti-Money Laundering and Countering Financing of Terrorism Act came into force for banks in 2013. And now banks are being required to step up their game on the conduct and culture front, including by removing pay incentives linked to sales.

For its part Deloitte's report highlights concerns the directors' attestation regime is not well suited to identifying or dealing with a range of issues including cyber risk, conduct risk, progressive risk appetite creep and complacency. 

The stakes have risen significantly since 1996 given there are now more areas where a bank can get things wrong. Thus any visions of crusty old bank directors sitting around the board table swilling gin and arranging their next round of golf as they sign off on the latest disclosure statement should long since be removed from reality.

The more prescriptive approach Deloitte recommends would see the Reserve Bank set out its risk management expectations along the lines of the Australian Prudential Regulation Authority's Prudential Standard CP2 220. This Prudential Standard requires an Australian Prudential Regulation Authority-regulated institution to have systems for identifying, measuring, evaluating, monitoring, reporting, and controlling or mitigating material risks that may affect its ability, or the ability of the group it heads, to meet its obligations to depositors and/or policyholders. 

After the Deloitte report was done the Reserve Bank declared the attestation regime "largely effective." However, the ongoing review of the Reserve Bank of New Zealand Act should be used to test, publicly, whether the attestation regime is fit for purpose in the 21st century.

DIY bank capital

Now onto the IRB capital approach. From 2008 the Reserve Bank granted ANZ, ASB, BNZ and Westpac the right to use this. What this means is these four banks are allowed to set their own models for measuring credit risk exposure and then get them approved by the Reserve Bank. Other NZ banks must use what's known as the standardised approach where the banks have their credit risk exposure prescribed by the Reserve Bank. Users of the standardised approach include the NZ-owned banks Kiwibank, TSB, SBS Bank and The Co-operative Bank.

Banks use risk-weighted assets (RWA) to work out the minimum amount of regulatory capital they must hold, with the capital requirement based on a risk assessment for each type of bank asset. For residential mortgages, for example, the key risk weighting standardised banks must use is 35%. 

The risk weightings used by the four IRB banks differ from one another, with ANZ's risk weightings having typically been the lowest. An example from my January article shows ANZ with a residential mortgage risk weighting of 19%, ASB 27%, BNZ 30% and Westpac 28%. Remember these banks, and the standardised banks, are lending to home owners in the same streets, towns and cities across New Zealand.

Lower risk weights ultimately means less capital is held against those loans. Holding capital costs banks thus that translates into a profitability advantage for the IRB banks. This is borne out in a return on equity (ROE) comparison between the four IRB banks and the four NZ owned banks mentioned above. The average March quarter ROE across ANZ, ASB, BNZ and Westpac was 15.5%. Across Kiwibank, TSB, SBS Bank and The Co-operative Bank it was 4.9%.

Now scale and efficiency doubtless assist the IRB banks. But so to does their use of the IRB approach. The cost to banks from holding capital is what that capital could earn them if it was invested elsewhere. Thus bank owners earn a greater return on their investment by using less of their own money and borrowing more.

As I reported in December, the Reserve Bank's consultation paper on its proposals to increase bank capital includes a probe of the big four banks' use of the IRB approach. This shows them with RWA equivalent to 69% of the standardised approach used by all other banks for residential mortgages. Across all types of lending combined, ANZ, ASB, BNZ and Westpac have RWA of between 67% and 86% of the comparable standardised calculations, with an average of 76%, as shown in the Reserve Bank chart below.

Challenging for both regulator & regulated

In response to the Reserve Bank's move to revoke ANZ's accreditation to model its own capital requirements for operational risk, ANZ has said this was just one of 45 models it operates. This gives you an idea of the scale of the task for both banks operating these models, and the Reserve Bank in overseeing the four banks and having to check and approve their models. It also shows the potential for both honest mistakes and gaming of the system.

Indeed use of the IRB approach has posed challenges for both regulated and regulator. In 2017 Westpac was forced, by the Reserve Bank, to significantly increase its capital after a series of errors in its capital modelling dating back as far as nine years were revealed. Last year BNZ restated historical regulatory capital adequacy ratios after disclosing a data error dating back to 2003. And ASB was in breach of its banking registration for nine years, albeit in its case capital ratios were higher than they should have been.

These examples are, of course, in addition to ANZ's censure that comes with the requirement for the bank to increase its operational risk capital requirement by $277 million.

In a combined submission on the Reserve Bank's capital proposals, Kiwibank, The Co-operative Bank, SBS Bank and TSB argue they require 45% more capital for for residential mortgages than the IRB banks, and 64% more capital for lending to small businesses. This, they argue, means a loan to the same borrower secured by the same house, in the same street requires 45% more capital to be held by a standardised bank than an IRB bank. This means banks competing in the same marketplace for the same business have vastly different settings for their key regulatory requirement, - capital.

The Reserve Bank is proposing ANZ, ASB, BNZ and Westpac's RWA requirements be increased to 90% of those required by standardised banks, and for them to calculate their requirements using the standardised approach as well, thus enabling proper comparisons across banks using the two capital methodologies.

Attitude matters

In a break from its hands off approach, the Reserve Bank has told ANZ to increase its risk weighted assets by more than $10 billion after reviewing the bank's capital adequacy for farm lending and residential mortgage lending. The increase in risk weighted assets comes with a minimum regulatory capital requirement of $800 million. This action is being taken because ANZ's modelling has been significantly out of whack with the other IRB banks and the regulator doesn't believe ANZ's loans are of sufficiently better quality to justify lower risk weights, and thus less capital.

There's ongoing debate over whether the Reserve Bank is adequately staffed and resourced to fulfil its duties, with Governor Adrian Orr telling me last year it needs more prudential regulation staff. Whilst having enough money and people to do the job required is clearly important, so is having the right attitude. There must be a willingness to act when doing so is required. Whether the recent moves on ANZ are the start of a more hands-on approach from the Reserve Bank, or just the regulator making a point whilst debate rages over its capital proposals, remains to be seen. 

The IMF made the important point that effective banking supervision needs the catalyst of "political will that cannot be measured nor evaluated externally." This is crucial given there's never likely to be many votes for politicians in providing more money for banking regulation. And nor is it likely to get much media attention unless done in response to a crisis, i.e. after the horse has bolted.

Nonetheless a modernising of the director attestation regime, the proposed changes to the IRB bank capital regime, and a regulator awake to its responsibilities and willing to act when and where required, would significantly bolster the oversight of New Zealand's banks.

*This article was first published in our email for paying subscribers early on Tuesday morning. See here for more details and how to subscribe.

We welcome your help to improve our coverage of this issue. Any examples or experiences to relate? Any links to other news, data or research to shed more light on this? Any insight or views on what might happen next or what should happen next? Any errors to correct?

We welcome your comments below. If you are not already registered, please register to comment.

Remember we welcome robust, respectful and insightful debate. We don't welcome abusive or defamatory comments and will de-register those repeatedly making such comments. Our current comment policy is here.

9 Comments

Comment Filter

Highlight new comments in the last hr(s).

good article and hits right at the heart of the matter re bank capital.

I think you could make a case that given the 'hands off' approach to bank supervision from the RBNZ that perhaps internal models isn't appropriate.

Personally... Id like to have some proper analysis done that compares the potential cost to bank customers of moving to a standard method with the cost to taxpayers of properly staffing the RBNZ so it can perform its oversight function properly and administer the internal models regime. Its what is sorely missing in this whole discussion.

If we are all going to have to pay more for bank services to make them safer then I want to know how much we save from letting people go at the RBNZ.

Thanks Andy, and thanks for your contribution to the bank capital debate. I expect there's a bit more water to flow under the bridge yet!

What RBNZ imposts are in place to stop banks creating their own regulatory capital?

5.2.1. How to create your own capital: the Credit Suisse case study

The link between bank credit creation and bank capital was most graphically illustrated by the actions of the Swiss bank Credit Suisse in 2008. This incident has produced a case study that demonstrates how banks as money creators can effectively conjure any level of capital, whether directly or indirectly, therefore rendering bank regulation based on capital adequacy irrelevant: Unwilling to accept public money to shore up its failing capital, as several other major UK and Swiss banks had done, Credit Suisse arranged in October 2008 for Gulf investors (mainly from Qatar) to purchase in total over £7 billion worth of its newly issued preference shares, thus raising the amount of its capital and thereby avoiding bankruptcy. A similar share issue transaction by Barclays Bank was “a remarkable story of one of the most important transactions of the financial crisis, which helped Barclays avoid the need for a bailout from the UK government”. The details remain “shrouded in mystery and intrigue” (Jeffrey, 2014) in the case of Barclays, but the following facts seem undisputed and disclosed in the case of Credit Suisse, as cited in the press (see e.g. Binham et al., 2013):

Link Section 5.2.1

They mention only the Operational Risk IRB model has been censured - bad enough for sure (and quite embarassing for ANZ) but the larger banks would have a range of IRB modelsso I dont think you can draw a conclusion about the 'state' of NZ IRB models just from this. They do mention that Westpac and BNZ also had problems with the capital they reported - but was this to do with IRB models (these only came in wih Basel2 about 10 years ago) or bad data quality. If Reserve Bank truly lack the resources to monitor these things and keep the banks inline then the Standardised Approach is perhaps the only way to go (or simply add a buffer on to the IRB models in question to allow for additional 'unknown' risk). I personally don't care about the co-ops (etc) argument that all the players should utilise Standardised approach to create equal playing field - if IRB works its great because in theory it should lower costs (?) for consumers and reward banks for having better internal controls and understanding of their assets.... but if we are saying that IRB in NZ is a facade that big banks just use to make more money and bemuse regulators then thats not right.

"light handed" and "self" regulation approaches don't come from any technical analysis. It's more from the cultural Kiwi horror of conflict that is open.
Other countries can do it differently. Folk can look you in the eye and say "that's bullshit" and amazingly you can still be friends and work together.
Time for the RBNZ to go "italian" or better"dutch"

KH - you are preaching to the converted here. I look at examples like the Bain case where repeated Kiwi appeals simply regurgitated the status quo but when it was taken to London they laughed at the in bred kiwis and he was freed in an instant.

What?

Gareth, while I am absolutely supportive of the removal of internal models regime, for many reasons, not least being that it is fundamentally unfair to NZ, I am a little more sanguine about director attestation.

I worry about becoming prescriptive in approach. While I can understand why firms such as Deloitte and co would love it, as it creates an lucrative auditable and consulting universe, what it does not create is something that is fit for purpose. Rather focus on outcomes, and make the directors squirm, and don’t, like they did with capital allow the auditors to weasel out of their obligations in attesting and affirming compliance as well.

As an insider, who has seen how directors react when being asked to attest, and there have been known issues, I know how much they agonise over signing......if there is the right leadership in the board.

Perhaps, like in the UK, persons who have specific roles should be positively vetted and reviewed as whether they are fit for purpose, both in skill, and attitude. That’s what needs to change.

Get the culture right at the top, and highly structured, consultant friendly processes become less relevant.

It becomes an argument about outcome, and objectives, rather than compliance and form filling.

Much harder on the directors and management, but so much more effective.

Good comment Anon annoy anono, cheers.