On November 1, 2017, an accounting asteroid hit the balance sheets of Canadian banks. Although markets have known about this momentous event for years, the reaction has been muted because most participants are underestimating its impact. The metaphorical asteroid, called IFRS9, is a new accounting standard that requires banks and other entities to change the methodology used for measurement of credit losses such that loss allowances are made immediately upon origination of a loan rather than when incurred (the current standard). Adoption of IFRS9 is mandatory for the 2018 fiscal year, which means private debt funds with a December 31, 2018 year-end, will first reflect the new standard in their January 2018 NAVs. The impact will not be catastrophic but still deep, and the shockwaves to lenders will be substantial and across several dimensions.
The most fundamental change under IFRS9 is that private lenders such as banks will now be required to take a provision on all loans as opposed to the current standard that requires objective evidence that the loan has become impaired before a reserve is established. Prior to the new standard, lenders would simply hold the loan at cost less amortization for repayments, with losses taken only when incurred. Now, lenders will be forced to provision for expected losses immediately based on the probability of the loan defaulting within 12-months. If, at each reporting period (which for some private debt funds could be monthly), the loan has experienced a significant increase in credit risk, then the provision would be increased based on the probability of default over the lifetime of the loan. Lenders will need to show some external validation for these loss provisions and incorporate macroeconomic and industry-related statistics into their analysis. Under IFRS9, loan loss provisions will increase and become more volatile.
IFRS9 establishes a three-stage approach for loan impairment tied to whether underlying credit risk of the borrower has deteriorated since inception (see chart below).
At initial recognition of the loan, the loan is in Stage 1, and the lender recognizes a loss provision equal to the 12-month expected credit loss, which is essentially the credit loss that is expected to result from default events possible within 12 months. This is determined by multiplying the probability of such default events by the loss that would occur given default within 12 months. If, at the reporting date, there has been no significant increase in credit risk, the exposure continues to be classified in Stage 1 (performing) and a loss allowance equal to 12-months expected credit loss is provided. If, however, there has been a significant increase in credit risk (Stage 2), an allowance equal to the “lifetime” expected credit loss is provided, that is, the credit loss that is expected to result from default events possible within the loan’s term. IFRS9 provides a non-exhaustive list of indicators relevant to assessing a significant increase in credit risk. These indicators can include: a decline in a borrower’s revenue; changes in contractual terms that would be made if the financial asset was newly-originated; adverse changes in general economic or market conditions; changes in borrower’s regulatory, economic, or technological environment; changes in value of collateral or guarantees; and expected or potential covenant breaches.
If there is objective evidence as at a reporting date of a detrimental impact on the estimated cash flows from a loan (for example, the borrower has filed for bankruptcy), then the loan is classified to be in Stage 3 and is in default or impaired. An allowance equal to lifetime expected credit losses is provided and interest income is recognized based on the impaired loan amount (i.e., gross carrying amount of loan less the lifetime expected credit losses).
A loan can migrate between stages based on whether there has been a significant increase in credit risk, which is a source of volatility in loan loss provisions and reported earnings (or returns for private debt funds). However, it is important to note that a loan that is overcollateralized may not require a loan provision even if default probability is high, depending on the liquidity and marketability of the collateral. All of TEC’s loans are overcollateralized at inception, and we do not anticipate any large swings in the returns of our loan portfolio as a result of IFRS9. We think the lenders most vulnerable to IFRS9 will be cash-flow-based lenders, which comprise the vast majority of the lending market, and regulated financial institutions such as banks that have capital adequacy requirements.
IFRS9 will have a major impact on bank capital levels. Higher and more volatile provisions will get recognized through the income statement, flow-through to reduce retained earnings, and thereby capital ratios. A turn in the credit cycle, which we have posited is becoming more likely, will decrease retained earnings faster than in previous downturns, just as capital ratio requirements increase causing banks to quickly curtail lending and leading to a self-fulfilling cycle. The differences between capital, stress testing, and accounting treatments will prevent banks from trying to underprovision or otherwise manipulate expected losses. A survey of global banks by Deloitte indicates that under IFRS9 loan loss reserves could increase by up to 50% for some banks. KPMG reported that “credit losses are expected to increase” along with “the number and complexity of judgments”.
The transition to the new accounting standard will impose systems challenges on smaller financial institutions and private debt funds. The expected loss model will increase the complexity of these lenders’ credit risk systems and require building new sets of models. Larger banks will try to extend existing Basel capital models, but because significant judgment is necessary to determine loan losses at each reporting period, IFRS9 introduces new operational risks. Smaller private debt funds will not have the resources to make the necessary investments in systems, and will not be equipped to coordinate origination, credit, and risk functions in the provisioning exercise to meet stated investment objectives. This will promote consolidation and favor larger private debt funds like ours that have the scale to reduce risk and maximize returns. IFRS9 will be a game changer for the lending markets.
We expect banks will be inclined to allocate capital to the lowest risk activities and reduce lending activities generally. Our own internal study of commercial loan growth at the five largest chartered banks in Canada showed that since 2015, banks expanded their book at an average rate of 11% per year, or about $30 Billion in the past twelve months. That impressive growth has coincided with a rapidly improving economy and benign default environment. We visit frequently with the special accounts units of banks and all of them are quiet or empty with little or no bad loans to manage. IFRS9 forces banks to look at different economic conditions and default situations, and acknowledge the pro-cyclicality of lending that can cause complacency in good times. Moody’s Analytics survey of banks found that more than 90% of respondents plan to integrate IFRS9 scenario analysis into their capital planning and origination activities. As Canadian banks reassess the way they model credit loss impairments, it is inevitable that they will slow the pace of commercial loan growth. The implications of IFRS9 to lenders will be particularly profound during the next credit downturn.