By FINTECH Books Contributor, Panos Skliamis The “IFRS 9 Financial Instruments” is the new standard which will replace the existing IAS 39. It contains a new impairment model resulting in earlier recognition of Expected Credit Losses (ECLs). Trade receivables is one of the subjects of the above standard and, although the approach of determining the relevant ECL is considered “simplified”, it is still too demanding compared with the current regulation. ECLs are calculated for all credit transactions at the time of origination and not only on impairment. Exposures (transactions) should be grouped by similar risk characteristics. Both historical (actual) losses and prudent forward-looking perspectives should be analyzed in order to derive probability-weighted loss rates per group. The above calculations must take into account the time value of money as well as potential Credit Mitigants (Credit Insurance, Letters of Credit, Letters of Guarantee, Factoring etc.).  Fulfilling some of the above requirements may become extremely challenging for a company, since specific historical data should be consistently stored in databases and complex statistical processes are required to run in a continuous basis. In addition, given that Trade Receivables are dynamically changing assets, traditional empirical methodologies or heuristics, especially if they applied in unsafe forms (e.g. spreadsheets), are far from appropriate for meeting the new regulatory requirements. Besides, the extent of the credit exposures that are susceptible to Credit Loss Provisions is now the entire Company’s Credit Portfolio, impaired or not, and the impact to the required provisions’ amount may be quite significant. RegTech comes to the rescue. A solution employing Big Data, Predictive Analytics and, ideally, Artificial Intelligence and Machine Learning techniques not only ensures meeting the regulatory requirements but can also significantly improve the Credit Risk Management Function.