경영자 이익예측정보가 미래 이익조정 및 미래 시장기대치 달성에 미치는 영향

In sum, the objective of this paper is two-fold. First, in this paper, we empirically examine the effect of current management earnings forecast accuracy and bias on next year's earnings management. Second, we empirically investigate how the current management earnings forecasts accuracy and bias influence the probability of meeting or beating market expectation of following year. In order to properly perform the analysis, we hand collect management earnings forecasts for fiscal years 2003-2009 from Financial Supervisory Service's Data Analysis, Retrieval and Transfer System (DART). In terms of management earnings forecasts, we focus on the operating earnings mamagement forecasts based on the number of prior literatures. The samples in order to properly examine the hypothesis 1 and 2 turned out to be 491 firm-years and the samples for the purpose of examining hypothesis 3 and 4 are 330 firm-years. Using these samples, our specific results are given as follows. First, we find that as the error or optimistic bias of current management earnings forecast increase discretionary accruals and performance matched discretionary accruals in the next year increase. In this paper, based on the prior studies, we expect that managers have incentives to manage next year's earnings upward, when market penalties are given for the current inaccurate/optimistically biased management earnings forecast. Therefore, the systematic associations between the accuracy/optimistic bias of current management earnings forecast and next year's earnings management are consistent with our expectations. Second, as the current management earnings forecast is less accurate or shows more optimistic bias, the probability of meeting or beating market earnings expectation of the following year becomes higher. These results indicate that manager's income- increasing earnings management of next year which is due to current management earnings forecast error/optimistic bias, is being conducted to increase the probability of meeting or beating market earnings expectation of following year. That is, manager's future earnings management induced by management earnings forecast error/optimistic bias is being used to enhance the probability of meeting or beating future market earnings expectation. Third, in our robustness analysis, we find that as the error or optimistic bias of current management earnings forecast increase accruals quality in the next year decreases. This result provides more insight in relation to the results of hypothesis 1 and 2. Also we show that the results of hypothesis 3 and 4 do not change even if we change the measurement of market expectation in the next year. This finding corroborates that our result is not driven by specific proxy of market expectation. Lastly, various results we mentioned above are consistent when we control the endogeneity problems. Specifically, we utilize Heckman’s two-stage regression analysis in order to lessen endogeneity issues. The result show that our results are robust after we control the endogeneity effects. This paper, by demonstrating the significant effects of current management earnings forecast accuracy or bias on next year's earnings management, provides a new insight about the market punishment for the inaccurate or biased management earnings forecast might also influence the manager's disclosure behavior related to future mandatory disclosure. Especially, the fact that prior researches mainly focus on the evidence whether market penalties for the current management earings forecast improve the future management earnings forecast, stresses our distinct contribution. Second, the relation between the current management earnings forecast errors/ optimistic bias and the next year's earnings management provides an indirect evidence that the market punishment for the inaccurate or biased management earnings forecast may be limited in improving the quality of future management earnings forecast. More specifically, this is because the fact that market punishment to current management earnings forecast error and optimistic bias induces managers to manage earnings upward implies that next year's management earnings forecast could also be distorted, given that firm's reported earings and earnings forecast might be managed concurrently. Third, outside investors might understand better with respect to the information content in next year earnings when they incorporate the fact that current forecast error and optimistic bias affects earnings management of following year. Fourth, establishing the fact that next's years earnings management due to current manager's forecast error and optimistic bias increases the probability of meeting or beating market expectation of the next year, market participants may better understand regarding the probability of meeting or beating market expectation. Lastly, notwithstanding the a number of contributions we mentioned above, our sample is confined to firms which disclose the management earnings forecasts. Therefore, our sample size is relatively small in order to generalize the results. Specially, the unsignificant results regarding pessimistic bias may be due to smaller sample size.