Commercial banks’ 2018 results have shown that adoption of international financial reporting standards version nine (or IFRS9) has had little impact on their income statements from an impairment adjustment perspective.
IFRS9 introduces a frontview (or expected) approach to recognising impairment of financial assets. Impairment of an asset simply refers to the amount by which the carrying amount of a cash-generating asset exceeds its recoverable amount.
Usually, any cash generating asset should be able to pay for itself and produce some profit on top. Failure to do so means that its carrying amount would have to be reduced to reflect this loss of utilisation.
Consequently, a business (in this case a bank) would have to recognise this realty in its financial statement by booking an impairment loss (and not loss provision).
Under IFRS9’s predecessor, known as IAS 39, banks derived this impairment recognition based on historical factors (commonly referred to as incurred loss model), which global regulators thought was not robust enough in helping banks create sufficient buffers against future shocks.
Under the current regime, financial institutions have to build various price points that would help them assign a credit-risk rating to borrowers, which then determines their probability of a default.
The street’s expectation was that adoption of IFRS9 would result into banks passing significant impairment losses through income statements. That didn’t happen. Instead, 2018 impairment losses through income statements declined by 22 percent year-on-year, and this was down to three reasons.
First, IFRS9 came with an exception to the extent that it allowed banks to pass any additional impairment recognitions through capital, and not income statement. They called it Day One rule. Essentially, in the first instance, institutions were allowed to capitalise the losses.
However, subsequent losses are to be passed through income statement. Additionally, the Central Bank of Kenya, in a guidnace note it issued in April 2018, provided an additional soft landing to banks by allowing them to spread the recognition of the additional impairments over a period of five years, ostensibly to limit the impacts on capital ( a move which I found to be rather unnecessary).
Second, under CBK prudential guidelines, banks were always required to set aside a rainy day fund in respect to good loans to take care of future unforeseen risks. In essence, there was a chance that the goodness could be illusionary (the Bikini puzzle). Essentially, as old bankers would tell you, rumble strips were just around the corner. It was only a question of timing.
That rainy day fund was pegged at one percent of all good loans and appeared to provide some cushion in the course of the migration.
Finally, IFRS9 incorporated the realisable value of a collateral into the impairment adjustment process. Essentially, in the case of an impairment, then the loss recognition ends where the realisable collateral value starts. However, just like its predecessors, IFRS9 has proved not to be immune from risk subjectivity. What do I mean?
Given a non-performing obligor with an exposure to multiple banks-say 10 of them, there is a high chance that, at the end of the day, each of these will recognise the impairment level of the same obligor differently, largely based on how they perceive its underlying credit risk.
This is called subjectivity of risk assessment. Going forward, and based on this subjectivity argument, I still expect more and more creativities around risk migrations. In essence, the much talked-about golden goose might not be invincible after all.
[email protected] @GeorgeBodo