This study investigates financial analysts’ consolidated earnings forecasts under Korean International Financial Reporting Standards (K-IFRS). Specifically, we examine whether financial analysts forecast consolidated earnings based on unbiased estimates of the persistence of parent and subsidiary earnings. We also analyze financial analysts’ incentives to provide consolidated earnings forecasts with supplementary separate earnings forecasts for the parent companies. Since the mandatory adoption of K-IFRS in 2011, Korean listed firms with subsidiaries should prepare and disclose consolidated financial statements as the primary financial statements. Financial analysts also provide earnings forecasts for consolidation entities under K-IFRS. Prior to the introduction of K-IFRS, accounting regulations in Korea have adopted individual financial statements as the primary financial statements, financial information users including financial analysts and investors are familiar with individual financial statements. Therefore, financial information users have difficulties in understanding consolidated financial statements under K-IFRS. However, even though there are significant changes in the financial reporting system from individual to consolidated statements due to the adoption of K-IFRS, there is little evidence whether accounting information users fully understand consolidated financial statements. In this study, we focus on financial analysts’ consolidated earnings forecasts and investigate two research questions. First, we examine whether financial analysts forecast consolidated earnings efficiently by analyzing analysts’ perceptions of the persistence of parent and subsidiary earnings. Specifically, we empirically test whether analysts reflect unbiased estimates of the persistence of parent and subsidiary earnings in forecasting consolidated earnings. Second, we focus on financial analysts’ supplementary separate earnings forecasts and investigate how supplementary separate earnings forecasts affect consolidated earnings forecasts efficiency. The sample consists of 2,408 forecasts-firm-year observations on Korean listed firms which have subsidiaries and forecasts of consolidated operating income for the period from 2011 to 2013. Based on Herrmann et al.(2007), parent earnings are measured by using operating income in parent’s separate income statement, and subsidiary earnings are measured by using the difference between consolidated operating income and parent’s separate operating income. To test the degree of forecast efficiency in the use of accounting information, we use the first consolidated earnings forecast issued after the announcement of prior year earnings within the individual analysts. We find the positive relation between the forecast error of consolidated earnings in the current year and consolidated and parent earnings in the prior year. This result indicates that financial analysts do not use a firm-provided consolidated earnings information efficiently and overestimate the persistence of consolidated and parent earnings in forecasting future consolidated earnings. We also find that financial analysts who issue both consolidated earnings forecasts for consolidation entity and separate earnings forecasts for parent company provide more efficient consolidated earnings forecasts than financial analysts who issue consolidated earnings forecasts only. If financial analysts issue both consolidated and separate earnings forecasts, information users can evaluate not only analysts’ ability to predict future performance of consolidation entity, but also analysts’ ability to predict the source of consolidated earnings. Therefore, these results imply that financial analysts are more likely to supplement consolidated earnings forecasts with separate earnings forecasts for parent company when they are better informed or have the superior predictive ability. This study contributes to the extant literature as follows. First, while significant changes occur in an accounting environment from individual to consolidated financial statements based disclosure system after the adoption of K-IFRS, there is little evidence whether financial information users fully understand consolidated financial statements. We provide empirical evidence that financial analysts do not correctly estimate the persistence of consolidated and parent earnings in forecasting consolidated earnings. The results that financial analysts as sophisticated users of financial information do not fully understand consolidated financial statements suggest the need for improving the quality of financial disclosure for consolidation entities. Second, we extend the literature on financial analysts’ supplementary forecasts by investigating their incentives to provide supplementary separate earnings forecasts on top of the consolidated earnings forecasts. Especially, considering the circumstances that investors have trouble in understanding financial information due to three different types financial statements under K-IFRS, our results might provide useful information to investors by demonstrating that financial analysts are more likely to issue supplementary separate earnings forecasts when they are better informed.