Skip links

 

 

 

 

 

 

 

 

 

 

Expected Credit Loss -Based Loan Loss Provisioning Norms

  • Recently, the Reserve Bank of India (RBI) said that the banks will be given ample time to implement Expected Credit Loss (ECL)-based loan loss provisioning norms.
    • Loan-loss provision, as defined by the RBI, refers to the allocation of funds set aside by banks to cover losses incurred from defaulted loans.
    • In simpler terms, it is a reserve of cash that banks keep to mitigate the impact of losses resulting from borrowers’ failure to repay their loans.

    This new approach replaces the current “incurred loss (IL)” model. Under the proposed norms, financial institutions like banks will have to calculate expected credit losses (ECL) on their loans during each reporting period and make necessary adjustments to their profit-and-loss account even before a borrower may default on a certain loan. This is in contrast to the present accounting norms wherein banks incur credit losses in their books only after outstanding loans have been in a state of default over a certain number of days as stated in the rules laid down by the RBI.

    • A key drawback in the IL model was that usually banks made provisions with a significant delay after the borrower may have started facing financial difficulties, thereby increasing their credit risk. This led to systemic issues.

    Furthermore, the delayed recognition of loan losses resulted in an overstatement of banks’ income, combined with dividend payouts, which further eroded their capital base.

 


 

Telecom Regulatory Authority of India (TRAI)

  • It is a regulatory body set up by the Government of India under section 3 of the Telecom Regulatory Authority of India Act, 1997.
  • It is the regulator of the telecommunications sectorin India.
  • Composition:
    • It consists of a Chairperson and not more than two full-time members, and not more than two part-time members.
    • The chairperson and the members of TRAI are appointed by the Central Government, and the duration for which they can hold their office is three years or until they attain the age of 65 years, whichever is earlier.
  • Government Control over TRAI:
    • TRAI is not a completely independent telecom regulator.
    • Under section 25 of the Act, it has the power to issue directions which are binding on TRAI.
    • The TRAI is also funded by the Central Government.
  • Functions:
    • Making recommendations on various issues;
    • General administrative and regulatory functions;
    • Fixing tariffs and rates for telecom services; and
    • Any other functions entrusted by the Central Government.
  • The recommendations made by the TRAI are not binding on the Central Government.
  • Central Government has to mandatorily ask for recommendations from TRAI with respect to the need and timing of new service providers and the terms and conditions of the licence to be granted to the service provider.
  • TRAI also has the power to notify in the official gazette the rates at which telecommunication services are being provided in and outside India.
  • The TRAI Act was amended in 2000, establishing a Telecom Disputes Settlement and Appellate Tribunal (TDSAT).

 

 


 

Machine Learning

  • It is a branch of Artificial Intelligence (AI)and computer science which focuses on the use of data and algorithms to imitate the way that humans learn, gradually improving its accuracy.
  • It gives computers the capability to learn without being explicitly programmed.
  • It enables computers to learn automatically from past data.
  • Machine learning uses various algorithms for building mathematical models and making predictions using historical data or information.
  • Features of Machine learning:
    • It is a data driven technology. Large amount of data generated by organizations on daily bases. So, by notable relationships in data, organizations makes better decisions.
    • Machine can learn itself from past data and automatically improve.
    • From the given dataset it detects various patterns on data.
    • It is similar to data mining because it is also deals with the huge amount of data.