In a survey conducted for U.S and Korean chief financial officers about their financial reporting behavior, managers consider accounting earnings as on of the most important performance measure, and are willing to sacrifice economic value for meeting or beating earnings benchmarks (Graham et al. 2005; Chun et al. 2012). Given the increasing research and the related concerns about managers' earnings management, a natural question arises as to the perception of financial analysts, who are the primary users of financial accounting information. Therefore, we seek to assess the perception of analysts of such managers' financial reporting behavior using a combination of a survey instrument and empirical study. To do this, we first conduct a survey that asks 185 Korean analysts of 15 brokerage firms to describe their views related to the firm's financial reporting behavior, specifically earnings management using discretionary accruals and real actions to meet earnings targets. Then, we compare the result of our survey with the one of Chun et al.(2012)'s survey to investigate the gaps in opinion between analysts and CFOs on the firms' earnings management to meet financial reporting goals. Lastly, we empirically examine whether earnings management via accruals and real decisions directly affects quality of financial reporting that analysts assess by OLS regression. We employ discretionary accruals as our proxy for accrual-based earnings management. Following Kothari et al.(2005), we include return on assets(ROA) as a regressor in the estimation model to control for the effect of performance on discretionary accruals. We also measure the extent of real earnings management by the sum of abnormal cash flow from operations, discretionary expenditures, and production costs using Roychowdhury (2006)’s models. We use disclosure quality as a proxy of financial reporting quality data from our survey by asking analysts to select the firms with high and low disclosure quality. The main results are as follows. First, our survey results indicate that analysts are mostly satisfied when the firms report the exactly the same earnings as their forecast numbers since forecast accuracy impacts on analysts' career concerns and external reputation. This contrasts with the results in a Chun et al.(2012)'s survey that CFOs prefer to report slightly beaten earings to signal higher future growth prospects to the external investors. Second, analysts believe that meeting earnings benchmarks relates to firm's better growth prospects, management credibility and stable operating performance in the future. Also, analysts thinks that meeting analysts forecasts helps to maintain good relationship between analysts and managements, even though CFOs do not agree with relationship incentives. Third, analysts reply that the majority of firms are engaging in earnings management to meet or beat market expectations. In contrast, only a few of CFOs admit that they would sacrifice economic value to achieve earnings targets in Chun et al.(2012)'s survey. Also analysts state that they reflect the implication of earnings management into their earnings forecast. Fourth, analysts think that managers use a large set of earnings management methods to achieve a earnings targets, which involve not only pure accounting discretion such as changing accounting principles or estimation, but also real decisions such as decreasing discretionary spending, delaying starting a new projects, accelerating sales, selling of fixed assets, etc. However, CFOs reply that they would not take any accounting actions to meet earnings benchmarks, nor would they take real actions except for decreasing discretionary expenses. Lastly, the results of empirical test using OLS regression suggest that disclosure quality perceived by analyst is higher for firms with low level of accrual and real earnings management. This study contributes to the literature in several ways. This is the first study to ask directly analysts' perceptions and views on managers' earnings management behavior by survey instrument. Especially, we observe that analysts and CFOs have totally different point of views as to the firms' earnings management to achieve earnings targets. This gap in opinion on earnings management between information makers and users has not been addressed in prior literature. Lastly, we document that earnings management have adverse impact on reliability of financial reporting and impresses analysts as poorer financial reporting quality. Especially we use disclosure quality collected from survey data as a dependent variable in the regression. This has not been tried in prior literature that only focused on the relation between earnings management and analysts' earnings forecast characteristics. In addition, this qualitative measure, directly assessed by analysts' own experience and insight, enable us to explore new implication of impact of earnings management on analysts information environment.