Moscow, July 15, 2021 — Business Systems Consult. I am often asked how Covid-19 is reflected in loan portfolios. Well, now basing on the data some lessons may be learned. 

Recently, when we renewed our seminar on the effective IFRS 9 implementation and methodology improvement, these questions were asked several times. 

Generally speaking, the economic consequences of the Covid-19 pandemics is a very complicated issue, even in the allowances in accordance with IFRS 9. Auditors voluntarily require to recalibrate or even reengineer the models of credit risk to make them reflecting the credit risk grown. The logic behind this requirement is straightforward. As Adam Smith put, “The interest of money is always a derivative revenue which, if it is not paid from the profit which is made by the use of the money, must be paid from some other source of revenue” (Book One, Chapter VI of his famous “An Inquiry…”). The problem is, as usually, in the details. First of all, model development is an industry with its own quality requirements (in part of data collection Basel III bcbs239 is applied). The plain wish of an auditor, even from the respected auditing company, is not enough. Data are collected with time lag (it is quite normal), and this wish of an auditor does not correspond to the principle of “reasonable and supportable information that is available without undue cost or effort at the reporting date” (see p. 5.5.17 of the Standard). The last but the most, the capital in the times of crises is under the pressure, and an idealistic requirement for an auditor is the pro-cyclical threat for the business.

The key to the efficient negotiations with an auditors is to deeply understand the behaviour of loan portfolios in crises. Data available for any bank or non-banking lending enterprise provide a solid basis for it and allows to concentrate managerial efforts on the improvement of portfolio management procedures.

According to IFRS 9 and general economic motivation, the behaviour of a loan portfolio is a combination of the following effects:

•    natural portfolio maturation (through-the-cycle behaviour);

•    macroeconomics (M-factor);

•    initial quality of loans (Q-factor);

•    early repayment (P-factor);

•    deepening the terms of past due which is a result of macroeconomics as well as collection procedures (MM-factor).

This combination is referred to as Babikov’s decomposition.

All these factors except the first one are subject to Covid-19. Their schematic description before and during 2020 is provided in the charts. This is the short sum of the observations of the loan portfolios with comparatively high quality:

•    In the spring of 2020 and by the June the additional frequency to moving the loans to past due grew sharply;

•    at this time the frequencies of deepening the past due terms also grew: the borrowers with difficulties could not solve their financial problems with higher probability than before the crisis;

•    however, the declines of early prepayments remained in seasonal trends;

•    since the July of 2020 when lockdown was ended (in Russia), the additional past due frequencies fell although remained positive (as compared to the pre-crisis level);

•    the early repayment frequencies exploded: precautionary savings is the dominating idea in the borrowers’ bahaviour;

•    banks should pay close attention to the quality of their models aimed at discriminating the borrowers (scoring cards): the observed higher volatility of the influence of the new loans to the frequency of loan transition to past due state (Q-factor) is a very alarming sign (it may be observed in many real credit portfolios).

To finalise, it should be noted that all the illustrations to this article are schematical and do not reflect the real situations of our clients. However, they qualitatively summorise our observations, and we can digitise these effects for your portfolios too.