The measurement of expected credit losses is a complex calculation that involves a significant number of interrelated inputs and assumptions and the allowance is not sensitive to any one single factor. The key drivers of in expected credit losses include the following:

To reflect relevant risk factors not captured in our modelled results, we applied expert credit judgment in determining the measurement of our weighted allowance for credit losses. The measurement of expected credit losses, including scenario design and weightings, determining significant increases in credit risk since origination and application of expert credit judgment, is overseen by a senior management committee that includes representation from Finance, Group Risk Management and Economics.

Internal risk ratings are assigned according to the risk management framework outlined under the headings Wholesale credit risk and Retail credit risk of the Credit risk section of Management’s Discussion and Analysis. Changes in internal risk ratings are primarily reflected in the PD parameters, which are estimated based on our historical loss experience at the relevant risk segment or risk rating level, adjusted for forward-looking information.

Our estimation of expected credit losses in Stage 1 and Stage 2 considers five distinct future macroeconomic scenarios. Scenarios are designed to capture a wide range of possible outcomes and are weighted according to our expectation of the relative likelihood of the range of outcomes that each scenario represents at the reporting date. We weight each scenario to take into account historical frequency, current trends, and forward-looking conditions which will change over time. Scenario weightings take into consideration the extent to which the base case scenario includes both favourable and unfavourable economic expectations, and upside and downside risks to the base scenario materializing in the future. The base case scenario is based on forecasts of the expected rate, value, or yield for each relevant macroeconomic variable. The upside and downside scenarios are set by adjusting our base projections to construct reasonably possible scenarios and weightings that are more optimistic and pessimistic, respectively, than the base case. Two additional downside scenarios capture the nature of potential credit losses across our portfolios. When the economy is at or near equilibrium, the severity of the downside scenario generally reflects an adverse event typical for a business cycle and both the downside scenarios reflect an outcome that is materially more adverse than the downside scenario.

The impact of each of our five scenarios varies across our portfolios given the portfolios have different sensitivities to movements in each macroeconomic variable.

The impact of weighting these multiple scenarios increased our ACL on performing loans, relative to our base scenario, by $945 million as at October 31, 2024 (October 31, 2023 – $868 million).

The PD, LGD and EAD inputs used to estimate Stage 1 and Stage 2 credit loss allowances are modelled based on the macroeconomic variables (or changes in macroeconomic variables) that are most closely correlated with credit losses in the relevant portfolio. Each macroeconomic scenario used in our expected credit loss calculation includes a projection of all relevant macroeconomic variables used in our models for a five-year horizon, reverting to averages generally within the 2 to 5 year period. Depending on their usage in the models, macroeconomic variables are projected at a country, province/state or more granular level. These include one or more of the variables described below, which differ by portfolio and region.

Our allowance for credit losses reflects our economic outlook as at October 31, 2024. Subsequent changes to this forecast and related estimates will be reflected in our allowance for credit losses in future periods.

Our base scenario reflects rising unemployment rates in the near-term in Canada and the U.S. Central bank policy interest rate cuts are expected to continue as inflation declines. Central bank policy interest rate cuts in Canada are expected to be larger than in other regions due to slower economic growth and higher unemployment rates.

Downside scenarios, including two additional and more severe downside scenarios designed for the real estate and energy sectors, reflect the possibility of a more severe macroeconomic shock beginning in calendar Q1 2025 relative to our base scenario. In these scenarios, conditions are expected to deteriorate from calendar Q4 2024 levels for up to 18 months, followed by a recovery for the remainder of the period. These scenarios assume monetary policy responses that return the economy to a sustainable growth rate within the forecast period.

The upside scenario reflects slightly stronger economic growth than the base scenario, without prompting a further offsetting monetary policy response as compared to our base scenario, followed by a return to a sustainable growth rate within the forecast period.

We reduced weight to our downside scenarios relative to October 31, 2023, in to reflect the likelihood of recessions as reflected in our downside scenarios.

 

 

 

 

 

Increases in the following macroeconomic variables will generally correlate with higher expected credit losses: Canadian and U.S. unemployment rates, Canadian overnight interest rates, Canadian and U.S. 10 year BBB corporate bond credit spreads, Canadian and U.S. 10 year government bond yields, and U.S. 10 year BBB corporate bond yield.

Increases in the following macroeconomic variables will generally correlate with lower expected credit losses: Canadian and U.S. housing price indices, Canadian and U.S. GDP, Canadian consumer confidence index, Canadian and U.S. commercial real estate price indices, and oil and natural gas prices.

Transfers between Stage 1 and Stage 2 are based on the assessment of significant increases in credit risk relative to initial recognition, as described in Note 2. The impact of moving from 12 months expected credit losses to lifetime expected credit losses, or vice versa, varies by product and is on the expected remaining life at the date of the transfer. Stage transfers may result in significant fluctuations in expected credit losses.

The following table illustrates the impact of staging on our ACL by comparing our allowance if all performing loans were in Stage 1 to the actual ACL recorded on these assets.

 

 

 

The following table presents the gross carrying amount of loans measured at amortized cost, and the full contractual amount of undrawn loan commitments subject to the impairment requirements of IFRS 9. Risk ratings are based on internal ratings used in the measurement of expected credit losses as at the reporting date, as outlined in the internal ratings maps for Wholesale and Retail facilities in the Credit risk section of Management’s Discussion and Analysis.