The procyclicality of loan loss provisions: a literature review

The recent introduction of expected credit loss (ECL) accounting standards under International Financial Reporting Standard 9 Financial Instruments (IFRS 9) and US Generally Accepted Accounting Principles (US GAAP) (Current Expected Credit Losses (CECL)) has impacted the amount and timing of loan loss provisions (LLPs) relative to the previous incurred loss (IL) standards. ECL standards require banks to recognise credit losses projected to crystallise in the future and credit losses already incurred. Recognition of such future losses, however, was generally not permitted under IL standards, which placed significant constraints on this practice. Many contend that the constraints under IL accounting led to a possible 'too little, too late' problem that reinforced the inherent procyclicality of the banking sector and amplified the depth and duration of the 2007–09 financial crisis.

The purpose of this literature review is to shed light on the role that credit loss accounting standards play in affecting procyclicality as viewed from the lens of a prudential policymaker. Accordingly, we take as our starting point the concept of 'procyclicality' considered by the Financial Stability Forum and BCBS as being related to the reinforcing interaction between the functioning of the banking sector and the real economy, leading to excessive economic growth during upturns and deeper recessions in the downturns. In this case, procyclicality is the idea that the banking sector, through a variety of channels or 'causal' links with the real economy, can exacerbate economic cycles. This interaction is a major policy consideration, since it can hinder the efficient allocation of resources in the economy and adversely affect credit growth and financial stability. Developing a better understanding of the degree to which IL and ECL standards support (ie strengthen or weaken) these causal links – and, therefore, procyclicality – is a key aim of this literature review.

With this concept of procyclicality in mind, we note that a key expectation of prudential policymakers is that the move from IL to ECL standards should, in fact, address the 'too little, too late' problem and benefit financial stability and the broader economy. This intended effect, however, depends on bank behaviour under the ECL standards, as well as the extent to which ECL standards improve (relative to IL standards) the timeliness and accuracy of loss recognition and increase the transparency of bank balance sheets. These effects are still not well established, making it difficult to assess ex ante the impacts of ECL standards, including the risk of unintended effects. This uncertainty has prompted some to question whether ECL standards might exacerbate procyclicality relative to IL. This question has become especially prominent in light of the coronavirus disease (Covid-19) crisis and its potential consequences for banks' LLPs. One could argue that actions taken by regulatory authorities around the globe to moderate the impacts of ECL standards and facilitate banks' ability to support economic activity during the Covid-19 crisis to some extent acknowledge this question. It is also important to bear in mind that the characteristics of the unforeseen Covid-19 shock plus the additional support measures introduce further challenges for the evaluation of the procyclicality of ECL standards. More robust evidence needs to be established on loss recognition practices under IFRS 9 and CECL and the extent to which these impact bank lending behaviour before a need for regulatory intervention to address procyclicality stemming from accounting standards can be evaluated.