Accounting Matters: Will IFRS 9 cause a new crisis?

• The IASB’s IFRS 9 standard evolved in response to the 2008-09 financial crisis
• IFRS 9 was designed to account for forward-looking expected loss models
• COVID-19 threatens to test the standard’s resilence

And so, in January 2009 the FASB voted to add a comprehensive project on financial instruments to its agenda. The IASB matched the FASB move in April 2009 with a statement outlining “plans to publish proposals within six months” to develop a single converged financial instruments accounting standard to replace International Accounting Standard 39 and its incurred-loss impairment model with a new, forward-looking expected-loss model. 

The two boards fell out in 2012, on the issue of impairment, and the IASB issued International Financial Reporting Standard 9, Financial Instruments (IFRS 9). A compelling address to the IFRS Foundation’s World Standard Setter’s conference in September 2009 by FASB member Tom Linsmeier demonstrated why the boards went their separate ways. 

He explains: “In recent financial institution crises, when we talk about… the savings and loan crisis in the US, the banking crisis in Japan, and the current crisis, hundreds, and perhaps even more accurately thousands, of financial institutions have failed with positive net worth and even higher positive regulatory capital just prior to their failure.”

He laid the blame at the door of loan-loss reserve accounting. “The failures of well-capitalised banks,” he argued, “and trillions of dollars of lost investor wealth and taxpayer money indicates the current amortised cost model for loans is broken….Some have characterised the amortised cost model as being consistent with the delay-and-pray-strategy: delay recognition of losses and pray the economy recovers before the bank fails.”

“Not only mustbonuses, dividend distributions and share buybacks stop, but so too must coupon payments on AT1 instruments” - Sven Giegold

Linsmeier’s concerns have been long shared by the Local Authority Pension Fund Forum. In late 2015, the forum’s chairman, Keiran Quinn, fired off a strongly worded letter to the EU commissioner for financial stability, financial services, Jonathan Hill. In it, he argued that IFRS 9 was a lame-duck standard. He also warned that defects in the IFRS 9 endorsement process had left both the European Commission (EC) and the UK’s Financial Reporting Council, open to the threat of legal action from investors.

It is undeniable that IFRS 9 requires banks to set aside an allowance on loan inception representing total foreseeable credit losses over a 12-month period. Where a loan is underperforming, IFRS 9 says you have to move it to stage 2 of the model where you book total lifetime losses. And, if the loan is credit impaired, it moves to stage 3, where the holder recognises not only total lifetime losses but a lower level of interest income. 

Such were the concerns about its impact on Europe’s fragile banks that the European Parliament in November 2017 approved a five-year plan to allow the banks to add back in up to 90% of their IFRS 9 impairments to their tier-1 core equity.

Fast forward to this March, and the COVID-19 pandemic, European governments responded by locking down societies crippling their economies. Our saviours were the banks who were to act as the conduit for delivering financial aid to businesses and households.

But the problem was that the banks were reluctant to load their balance sheets with loans that they feared might never be repaid. The forecasts that feed into the calculation of loan loss allowances under the new standard were truly dire. Something had to give. 

In April, the EC annouced measures to grant temporary relief to EU-based banks from the full impact of accounting and regulatory rules that many feared were holding back lending. It hoped to minimise the impact of the expected-credit-loss model without undermining investor confidence.

In March, the European Securities and Markets Authority (ESMA) issued a statement in which it addressed the accounting implications of the pandemic. It contained the observation that “the principles-based nature of IFRS 9 includes sufficient flexibility to faithfully reflect the specific circumstances of the COVID-19 outbreak and the associated public policy measures”. Issuers and their auditors, ESMA said, should take note.

Essentially, the measures, which should become law in June, will allow the banks to look through any deterioration in credit and assume the loan will be recoverable over its lifetime. The EC adds in an accompanying Q&A document that banks are also “encouraged to implement the IFRS 9 transitional arrangements that will reduce the impact of IFRS 9 ECL [expected credit loss] provisioning on banks’ regulatory capital”.

For Green Party MEP and ECON committee member Sven Giegold, the measures lack conditionality. “If the banks now get equity relief,” he says, “they have to do everything they can to maintain their equity. Intervention by the state during this crisis cannot be a one-way street. 

“Not only must bonuses, dividend distributions and share buybacks stop, but so too must coupon payments on AT1 instruments. Legislators must make this clear and not leave decisions on distributions to the banks.… The preferential treatment of public guarantees in the rules governing impaired loans exacerbates the problematic sovereign-banking nexus.”

Others, however, such as former ECON chair Sharon Bowles, argue that extraordinary times call for unprecedented measures: “I agree with Sven Giegold’s point about restricting distributions and bonuses, and I can also see the sense in including AT1 restrictions – halting coupon payments on those is a half-way measure before a full trigger of conversion.”

Where she disagrees is on his criticism of sovereign guarantees for bank lending. “Yes, it does exacerbate the sovereign-banking nexus, but if it is prohibited, then it will exacerbate and increase other macroeconomic imbalances, which have not been addressed,” she says.

Some, however, argue that regulators have managed to unpick what little progress IFRS 9 delivered. One such critic is financial instruments specialist and former Lombard risk-management expert Cormac Butler. He says the effect of recent regulatory interventions renders the move to IFRS 9 largely cosmetic.

He says: “Under IFRS 9, banks can continue to hide losses if the directors believe that while the recoverable value is less than the amount advanced, the change is not significant enough to justify making a provision.” As regulators are now telling banks that they no longer need to treat missed loan payments as an impairment trigger, that, he believes, leaves us effectively back at IAS 39.

He warns that the 2008 Irish banking crisis suggests that the banks’ second-quarter results will mask their difficulties with non-perfoming loans. “If you take a 20-year mortgage, calculating one year’s expected credit loss will not necessarily capture all losses,” he argues.

“There is the risk that banks will rely on overvalued assets to borrow funds from central banks, which is precisely what happened in Ireland when the ECB lent substantial sums to Anglo Irish Bank – even though it was clearly insolvent but showing healthy accounts.

“So, the starting point for getting out of this mess is for banks to bring asset values down to realistic levels and reveal all losses. Then it’s a matter for governments if they want to refinance it or not.”