报告.docx

1、报告 南 京 理 工 大 学毕业设计(论文)外文资料翻译学院（系）： 经济管理学院 专 业： 金融学 姓 名： 李晶 学 号： 0907810308 (用外文写)外文出处： JOURNAL OF FINANCIAL AND QUANTITATIVE ANALYSIS Vol. 46, No. 6, Dec. 2011, pp. 19171946 附 件： 1.外文资料翻译译文；2.外文原文。 指导教师评语： 签名： 年 月 日注：请将该封面与附件装订成册。附件1：外文资料翻译译文 投资者能够看穿盈利报告中的错误码？摘要本研究探讨投资者是否能看穿错误陈述的盈利，并且预测盈利重述。通过对最终在年报

2、中重述盈利企业的研究，我们发现投资者在最初盈利公告发布时会被公告收益中的错误信息所误导。投资者对那些接下来会被重述的盈利的反应是积极的，并且他们会把错误陈述的收益与真实收益的价值看成是相同的。我们同时也发现投资者能够在重述公告发布之前的几个月内预测盈利重述并且开始估计股票会价格下降。因此，一个信息重述所带的全部冲击是重述效应的三倍。事实上，我们的研究也表明投资者会惩罚做出信息重述的企业，因为在企业最初夸大收益时，股东所享有的股票收益要远远大于在重述公告做出时被反转回来的。引言最近涉及到夸大收益的公司丑闻已经触犯了证券交易委员会的若干规定以及2002年萨班斯奥克斯利法案的相关条款，该条款旨在提高

3、公司财务报告的真实性。而使用这些规则的前提是投资者并不能够通过标准的财务报告看穿会计错误与利润操纵。但是当投资者运用其他私人和公共的信息来评判公司财务报告的有效性时，新的规定中的许多条款实施起来也许会昂贵而无必要。令人感到惊奇的是，以前的研究很少花时间去检验投资者是否被错误的会计陈述所蒙蔽。相比之下，焦点较多集中在重述公告发布之前和之后的一段时间内价值效应和股票交易行为。但是投资者会被误导多久和多长呢？当每一个国家都在评价2008年金融危机的起因和计划监管改革以预防未来的危机时，在监管大企业经营是否健康方面，财务信息准确与透明的重要性是不言而喻的。举个例子，比如说那些在雷曼兄弟破产之前并没有被

4、审计师和监管者注意到的会计违规引起了人们对“投资者在多大程度上被错误的财务信息所误导”的问题的极大关注。本篇论文的目的在于对“投资者识别或看穿财务报告中错误信息的能力”提供一个全面的分析。我们的研究不同于以前信息重述文献，因为它把焦点集中在市场对于原始的错误收益公告的反应以及在错误期重述企业的公司价值。我们把错误期定义为错误陈述期的一号到重述公告发布的那天。据我们所知，以前的研究都没有检验市场对随后会被重述的原始公告的反应。只有一个研究检验了在错误期的价值，而那个研究距离现在已较远并且重述信息的样本也很小。当财务公告不存在错误时，与利润公告相关的超额收益对于解释异常收益起着非常重要的作用。我们

5、把492个重述样本的异常收益分解为两部分：真实的超额收益和由于利润公告错误而产生的超额收益。如果投资者被错误的信息误导并且把错误与正确的信息同等看待，那么任何与利润公告相关的超额收益在分解为两部分中的任何一部分中都会被同等看待。事实上，通过对最终在年报中重述的公司研究，我们发现在错误期，投资者对于错误部分的超额收益持积极态度。同时投资者把超额收益中正常的部分与由于错误信息而产生的超额收益部分同等看待。我们的证据进一步显示投资者在错误期的开始比末尾更容易被错误信息误导。那些进行利润重述的公司在重述前的几个月里会经历下降的负的超额收益，这个结果对于年报和季报的超额收益以及多种测算超额收益的方法都成

6、立。重述公告之前的买卖超额收益的数量是公告效应的两倍。事实上，这些负的异常收益已经大大超过了当公司最初夸大收益时由公司所享有的超额利润。此外，在以后会重述的公司在重述公告发布以后出现负的异常表现。因此，重述所带来的全部负面效应比公告效应要大得多。总的来说，我们的发现表明市场会以大于由发布错误公告而产生的误估价值的折扣来惩罚那些重述企业。然而令我们惊奇的是，我们的研究发现那些最初低估了利润随后进行重述的公司也在重述前一个月里和重述公告发布时经历了负的超额收益，这表明当公司在公告中将利润由低调高时，信息重述也是一个坏消息。证券监管者或许会把我们的结果解释为投资者可以从得到改善的财务信息的行为中获利

7、。尽管投资者会在重述公告前几个月内开始预测重述公告，但是他们仍然被最初的错误公告而误导。这个平均的错误期对全样本来说是两年而对年报重述来说是两年半。因此，投资者在相当长的一段时间里是依靠错误的信息来交易的。除了检验在错误期内的评估，我们还研究了在其他两个时期内重述企业的收益，一是在重述公告期的一号之前的三年（错误前时期），二是在重述公告发布以后的三年（重述后时期）。我们发现几乎没有证据表明在错误前时期存在异常收益，但是有一些证据表明那些在重述公告中将净利润调低的企业在重述后时期存在负的超额收益。进一步，我们发现在核心账户中出现错误的公司，例如：收益与成本，会比其他重述公司在最后一个四分之一的错

8、误期中表现更差，这表明投资者会惩罚在核心账户中存在错误的公司。有欺骗行为的重述公司比没有欺骗行为的重述公司在第一个四分之一的错误期中有更好的表现，这表明这种具有欺骗行为的利润公告很难被觉察。我们的论文也有助于研究对存在虚假财务报告企业的惩罚。许多论文都发现信息重述会对公司的股权与债券成本带来负面影响。Kravet 和 Shevlin认为重述会增加信息风险，这会导致股本更高的机会成本。同样地，Hribar 和Jenkins的研究结果表明股权的隐含成本会在重述后增加。Graham, Li, 和 Qiu指出重述也会给公司的债务成本带来负面影响。Chen, Cheng, 和Lo发现重述企业在融资上会受

9、到限制，同时重述后这些企业都不太可能获得外部融资。Wilson发现收益的信息满意度会随着重述而暂时下降。除此之外，企业在重述后劳动程度和投资都会下降。最后，Karpoff, Lee, 和 Martin他们发现有证据表明：由于财务方面的错误陈述而被证券交易委员会罚款的企业要付出很高的法律和名誉成本，这些企业中有许多是因为变更了财务陈述。论文的其余部分安排如下：第二部分包括重述的时间链、研究的问题和统计方法。第三部分主要阐述了数据。第四部分阐述了实验结果和内涵意义。第五部分总结了论文。重述时间链条、研究问题与研究方法重述的时间链条和收益期的定义以前的许多研究提供了在错误期的不同阶段有关公司价值的部

10、分证据。我们这样来定义错误期，它是指：错误陈述期的一号到重述公告发布的那天。之前的研究，除了Kinney 和 McDaniel的，都没有定义错误期或者分析在那个时期的收益。大多数以前的研究都使用混合时间窗口，要么是在重述期左右，要么是在错误期开始左右，最终得到的是错误前时期、错误期和重述后时期的混合收益。Kinney 和 McDaniel检验了错误期的情况，并且发现了在有限样本的情况下出现了负的超额收益。但是他们并没有把错误期细分为更小的时期，与此同时，不同于其他研究的是由于他们有限的样本，他们无法找到在重述公告时显著性负的超额收益，这些表明那些有限的样本并不能代表更多近期重述公告的情况。图表

11、一揭示了重述时间链条。我们首先来精确地定义错误期。它开始于包含有利润错误的财务报告期的一号，结束于重述公告发布的那天，即当公司纠正了它的利润错误，揭示企业真正价值的时候。金融理论告诉我们投资者通常会用他们预期的未来现金流来估计股票的价值。然而，投资者经常依靠企业发布的利润报告来帮助自己估计未来的现金流。Graham, Harvey, 和 Rajgopal证实了：大多数的公司把营业利润而不是现金流看做是吸引外部投资者的重要优势。利润也被用来估计市盈率，而市盈率是一种广泛的分析股票价值的方法。因此，包含有错误利润信息的财务报告会导致投资者高估或低估股票的价值。市场对包含有错误利润公告的短期反应投资

12、者并不只是依靠财务信息报告来估计企业的盈利状况和未来的发展前景，他们同时也运用其他的信息资源。各个产业和政府的统计数据，竞争者的盈利报告以及生产者和消费者的信息都可以帮助投资者在财务报告发布之前对企业的盈利状况作出准确的估计。尽管投资者不可能完全忽视企业发布的包含有错误的财务公告，但是如果企业发布的利润与通过其他信息途径所估计的企业利润相距甚远的话，那么企业发布的利润的可信度对于投资者来说是要大打折扣的。Beaver (1968)，Landsman 和 Maydew (2002)，Ball 和Kothari (1991)，Chari，Jagannathan, 和 Ofer (1988), 以及

13、Cohen, Dey, Lys, 和Sunder (2007)都证实：在财务报告发布之前透露给投资者的新闻公告并没有完全征服投资者。但是，投资者在多大程度上依赖于企业的财务报告这仍是一个悬而未决的问题。附件2：外文原文（复印件）Do Investors See through Mistakes in Reported Earnings?AbstractThis study investigates whether investors see through materially misstated earnings, and whether they anticipate earnings r

14、estatements. For firms that restate at least one annual report, we find that investors are misled by mistakes in reported earnings at the time of initial earnings announcements. Investors react positively to the component of the favorable earnings surprise that will subsequently be restated, and the

15、y attach the same valuation to it as to the true earnings surprise.We also find that investors anticipate the subsequent downward restatements and start marking stock prices down several months before a restatement announcement, so that the full impact of a restatement is about three times as large

16、as the restatement announcement effect. Indeed, we show that investors punish restating firms because the stock price gains that shareholders enjoy when firms initially announce overstated earnings are more than reversed by the time of the restatement announcement.I. IntroductionRecent corporate sca

17、ndals involving overstated earnings have motivated several Securities and Exchange Commission (SEC) rules and Sarbanes-Oxley Act of 2002 provisions aimed at boosting the integrity of financial reports (Palmrose, Richardson, and Scholz (2004). The premise of these rules is that investors typically do

18、 not see through the standard financial reporting to identify accounting mistakes and earnings manipulations. But if investors use other private and public information to validate reported earnings, several of the new rules may be costly and unnecessary.Surprisingly, earlier studies spend little tim

19、e examining whether investors are fooled by erroneous financial statements. Instead, most focus on the value effects and stock trading behavior just before and after restatement announcements.1 But how often and for how long are investors misled? The importance of accurate and transparent financial

20、statements in monitoring the health of large firms is center stage as the country assesses the causes of the 2008 financial crisis and plans regulatory reforms to prevent future crises. For example, accounting irregularities, such as those that went unnoticed by auditors and regulators prior to Lehm

21、ans demise, drew greater attention to the question of how much investors are misled by erroneous financial information (De la Merced (2010).The goal of this paper is to provide a comprehensive analysis of investors ability to see through mistakes in financial statements. Our study differs from prior

22、 literature on restatements in that it focuses on market reaction to the original announcement of misstated earnings and the valuation of restating firms in the error period, which extends from the 1st misstated period to the day of restatement announcement.2 To the best of our knowledge, no prior s

23、tudy examines the market reaction to initial announcements of incorrect earnings that are subsequently restated. Only one previous study examines valuation in the error period for an older and smaller sample of restatements (Kinney and McDaniel (1989).In the absence of mistakes in financial statemen

24、ts, abnormal returns associated with earnings announcements are an increasing function of earnings surprises. For our sample of 492 restatements, we decompose the earnings surprise into 2 components: the true earnings surprise and the surprise due to the error. If investors are misled by erroneous e

25、arnings and treat the correct and error components the same, then any associated abnormal returns will be proportionately the same for each component.Indeed, for the subsample of firms that restated at least one annual report, we find that, during the error period, investors react positively to the

26、error component of the earnings surprise and attach the same valuation to the error component of the earnings surprise as to the true part of the earnings surprise. Furthermore, our evidence suggests that investors are more misled by mistakes made at the beginning rather than at the end of the error

27、 period.Firms that restate earnings downward experience negative abnormal returns several months before they restate. This result holds for both annual and quarterly restatements (restate only 10-Qs) and for various methods of calculating abnormal returns. The magnitude of abnormal buy-and-hold retu

28、rns prior to restatement announcements is more than twice the announcement effect.4 Indeed, these negative abnormal returns more than offset the initial positive abnormal returns enjoyed by firms when they initially report overstated earnings. Furthermore, firms with downward quarterly restatements

29、exhibit negative abnormal performance after the restatement announcement. Hence, the total negative impact of a restatement is much larger than the announcement effect. Overall our findings indicate that the market punishes restating firms with a discount that is greater than the misvaluation create

30、d by the misstated earnings.Somewhat surprisingly, we find that firms that understate earnings and thussubsequently restate earnings upward also experience negative abnormal returns 1 month prior to their restatement and at the restatement announcement, suggesting that restatements are bad news even

31、 when firms revise earnings upward.Securities regulators could interpret our results as supporting the view that investors would benefit from improved financial reporting practices. Although investors start to anticipate restatements several months prior to announcement, they are misled at the initi

32、al announcement of erroneous earnings. The average length of the error period is 2 years for the full sample and 2.5 years for annual restatements, so investors trade on erroneous information for quite some time.In addition to examining valuation in the error period, we also study returns of restating firms during 2 other periods: i) the 3-year period prior to the 1st restated report (pre-error period); ii) and the 3-year period after the restatement announcement (post-restatement period).We find little evidence of abnormal performance in the pre-error period and fin