기존의 연구들은 발생액이 현금흐름에 비해 이익 지속성이 떨어짐에도 투자자들은 이러한 차이를 구분하지 못한다는 사실을 보여 주고 있다. 이처럼 자본시장이 발생액의 이익 지속성에 대해서 과대평가하는 사실을 발생액 이상 현상이라 부르고 있다. 선행연구에서는 이와 같은 발생액 이상 현상의 가장 주된 원인으로 경영자의 이익 조정 행위를 꼽고 있다. 본 연구는 Mishkin test를 활용하여 발생액 이상 현상이 기회주의적 이익 조정에 기인하는지 검증하는 것을 목적으로 한다. 이익을 조정한 기업과 조정하지 않은 기업을 구분하여 이익 조정 기업에서 발생액 이상 현상이 더 크게 나타난다면, 자본시장이 기업의 이익 조정 행위를 가치 평가에 제대로 반영하지 못하여 발생액 이상 현상이 나타난다는 사실을 실증하는 것이다. 반대로 이익을 조정한 기업에서 발생액 이상 현상이 작게 나타난다면, 시장은 기업의 이익 조정을 효율적으로 인지하고 있으며 기능적 고착화가 발생액 이상 현상의 원인이 아님을 보여주는 것이다. 실증 분석 결과는 다음과 같다. 첫째, 전체 기업 집단에서 발생액 이상 현상은 선행연구와 일관되게 나타나고 있다. 둘째, 이익을 조정한 기업은 이익을 조정하지 않은 기업에 비해 이익 지속성이 낮게 나타났으며, 그 차이는 발생액이나 재량적발생액 항목에서 유의하다. 셋째, 발생액 이상 현상은 이익을 조정하지 않은 기업에서만 발견할 수 있고, 오히려 이익을 조정한 기업에서는 발생액 이상 현상을 관찰할 수 없었다. 본 연구는 발생액 이상 현상의 원인으로 지목되는 경영자의 기회주의적 이익 조정 여부를 명시적으로 구분하여 분석하였고, 그 결과 선행연구의 주장과는 달리 발생액 이상 현상이 이익 조정의 산물이 아닐 수 있음을 보여 주었다. 따라서 발생액 이상 현상의 주된 원인으로 인정 받고 있는 이익 조정에 대한 시장 비효율성의 주장에 보다 신중한 접근이 필요하다는 시사점을 제공한다.
Researchers have opposite perspectives for discretionary accrual. Some support its positive aspect in that it can deliver managers’ private information for future performance, having a signalling effect. On the contrary, others doubt it since it reflect managers’ opportunistic earnings management. It has been the most popular metric measuring earnings management in a lot of earnings management study. Thus, discretionary accrual is still controversial among researchers. Specifically, some have a positive eye on discretionary accrual due to its additional explanation on earnings persistence, while others a negative eye because of its relatively lower explanation power than cash flow or non-discretionary accrual. In addition, some justify information content of discretionary accrual in that it explains stock returns significantly, while others belittle since the capital market overvalues it due to unawareness of earnings management. These differential perspectives are caused by earnings management. Chan et al. (2006) argued that discretionary accrual can be a leading indicator of future business if only manager does not manage earnings. Therefore, the objective of this study is to examine whether there is a difference between earnings managed firms and non-earnings managed firms after classifying earnings managed firms. Earnings management has not been clearly considered in prior studies, because they treated all firms into one portfolio. But in order to synthesize different opinions that seem to be contradictory, we attempt to divide all firms into two groups, earnings managed firms and non-earnings managed firms. After that, we investigate whether earnings persistence and market reaction are different in each group. If difference is found, we can show that discretionary accrual carries managers’ private information in non-earnings managed firms, while it indicates managers’ opportunistic activity in earnings managed firms. These synthetic study is carried out in two aspect, earnings persistence and market reaction. First, discretionary accrual with earnings managed firms would be able to explain future earnings relatively less, while it can account for future earnings better in non-earnings managed firms. Second, we try to investigate how the capital market evaluates each earnings persistence in two groups. Prior empirical studies have shown that market participants do not catch that the accrual is less accountable for earnings persistence than cash flow. Accruals anomaly is such a phenomenon that the stock market overvalues accrual related to earnings persistence. Studies examining accruals anomaly attributed it to market inefficiency for earnings management and functional fixation is widely regarded as a hidden driver of market mispricing. Similarly, if market cannot recognize earnings management, accruals anomaly in earnings managed firms would be bigger than in non-earnings managed firms. But if market is efficient enough, it can identify earnings managed firms and there are no accruals anomaly or it would be much smaller in earnings managed firms. This can enable us to test market inefficiency for earnings management which has been ascribed to accruals anomaly. Empirical results indicate three findings. First, accruals anomaly for total accrual and discretionary accrual is consistent with prior studies. Market overvalues earnings persistence of total accrual, non-discretionary accrual, and discretionary accrual. But Korean stock market differs from US in that undervaluing for cash flow is not significant. Second, non-earnings managed firms show more strong earnings persistence. And this is significant for both total accrual and discretionary accrual. Third, accruals anomaly exists only for non-earnings managed firms. That is, non-earnings managed firms drive market mispricing and the market has an insight into earnings management. We can synthesize opposite perspectives of prior studies through these results. First, the higher earnings persistence in non-earnings managed firms implies that its discretionary accrual has managers’ private information. However, the lower earnings persistence in earnings managed firms indicates that managers abuse it to exaggerate their performance. Second, discretionary accrual in earnings managed firms cannot account for stock returns, while that in non-earnings managed firms is associated significantly. This means that market can recognize the meaning of discretionary accrual whether it is the result of earnings management. Third, a lot of researchers have conjectured that market imperfection or functional fixation is responsible for accruals anomaly. But these results are inconsistent with prior studies in that the accruals anomaly is not caused by earnings management. Therefore, we suggest that researchers should be more careful in asserting that market inefficiency for earnings management is the main cause for accruals anomaly.
