This paper examines the relative value relevance between IFRS net income and loan-loss reserve adjusted net income for Korean banks, which was mandated for banking industry regulation purpose as a result of IFRS adoption in 2011. IFRS requires firms to use the incurred loss model for recognizing bad debt expenses. The incurred loss model is considered to recognize less loan loss allowance as compared to the expected credit loss model. As a result, the Korean Financial Supervisory Commission requires banks to record additional loan-loss reserves in the wake of IFRS adoption in fear of recognizing insufficient loan loss. The additional loan-loss reserve is a non-GAAP approach which is imposed for regulation purpose. But it is considered necessary to comply with the leverage ratio set by the Basel Committee on Banking Supervision. The non-GAAP additional loan-loss reserve affects two elements of financial statements. First, it is subtracted from the capital measure, the numerator in the leverage ratio. It is subtracted from the IFRS net income to arrive at the loan-loss reserve adjusted income. The additional loan-loss reserve is allowed to be disclosed in the face of or in the notes to the financial statements. From an analysis of the relative value relevance between the non-GAAP adjusted income and the IFRS net income for 156 firm-years in the period of 2011 to 2014, we document that the non-GAAP income has higher value relevance than the IFRS net income. We also document that the disclosure pattern also affect value relevance differently. More specifically, those banks which report the income adjustments on the income statements have higher value relevance than those banks which report them in the notes to the financial statements. Furthermore, using a sub-sample of firm-years which report the adjustments on the income statements, we document that the value relevance of the adjusted income is higher than that of the IFRS net income. We contribute to the literature by empirically showing that the non-GAAP additional loan-loss reserve has additional information content and can be justified for regulation purpose. Our result also provides empirical supports for the validity of the expected credit loss model in IFRS 9 which supersedes the incurred loss model.