1、 外文翻译 原文 The Long-Term Post Acquisition Performance of Greek Acquiring Firms Material Source: http:/ Author: Dimitris Kyriazis 1 Introduction and Literature Review The vast majority of empirical studies concerning the biggest and more active markets of the US and the UK have established that the tar
2、get companies shareholders reap on average large and significant abnormal returns.At the same time, the acquiring firms shareholders experience on average small negative or close to zero insignificant abnormal returns around the announcement period, which turn into highly negative and significant in
3、 a long-term post-completion period. So, the issue, which remains one of the unresolved puzzles in corporate finance, is why some firms should decide to bid for the acquisition of other firms, if this has been considered by the market to be on average a wealth decreasing activity for the interest of
4、 their shareholders. Over time, in order to answer this question, several theories have been developed and various hypotheses have been tested. We can classify these theories and hypotheses into two broad categories, those which are consistent with the market efficiency hypothesis and those which di
5、rectly or indirectly cast doubts on the theory of market efficiency. Therefore, the primary objective of this study is to examine the long-term (1, 2 and 3year period) post-bid performance of Greek acquirers and to compare these results with those reported in the previous study of Kyriazis and Diaco
6、giannis (2008) for the short-term announcement period by using the same sample of takeover bids for the period 1993-2006. Based on the methodology of the 3-factor model of Fama and French (1993) for the estimation of abnormal returns, the present study finds that the Greek acquiring companies make l
7、arge negative and statistically significant abnormal returns of about 2% per month for a period of 12, 24 and 36 months after the bid announcement or completion date. Consequently, there is indeed a reversion trend in abnormal returns of the acquirers in the postbid period, which cannot be attribute
8、d either to the application of the “correct” asset pricing model (the 3-factor model instead of the market index model), as Fama (1998) argued, but rather to the overestimation of expected synergies during the announcement period which implies market mispricing and inefficiency. At the same time, th
9、is study investigates the impact of a few other factors on the acquirers gains, such as the medium of payment (cash/stock), the status (listed/unlisted) of the target firms, and the characteristics of acquirers based on the size and the Book-to-Market ratio, and compares them with the findings of pr
10、evious international empirical research studies, as well as the work of Kyriazis and Diacogiannis (2008) for the Greek takeover market. 2 Data-Sample To perform our analysis, we used initially the sample of 100 completed takeover bids, which was collected for the needs of the previous study by Kyria
11、zis and Diacogiannis (2008).This initial sample contained takeover activity data over the period of 1993-2006 years from the Thomson One Banker Deals database (formerly Platinum SDC). This period of time includes two booms in takeover activity, namely the sub-periods of 1998-2000, 2004-2006 and one
12、period of declining merger activity, for the years 2001-3. From an initial population of 700 bids, we ended up with this reduced sample of 100 bids, due to the following filters applied: -only completed domestic deals (both bidders and targets were Greek domiciled firms at the time of the bid) -ther
13、e was a full change of corporate control in the post-bid period (the bidder increased his acquisition stake to more than 50%) -successive acquirers for a period of 1 year before the first bid were excluded, in order to have a clean estimation period of returns for the estimation of the abnormal retu
14、rns by the market model. The number of cases was further reduced, when we had to identify the mode of payment (pure cash/ pure stock) and collect information about the daily stock prices of the merging firms and the DataStream Total Greek Market Index, dividend payments and ex-div dates (to compute
15、total shareholders returns, as well as the monthly values of book and market equity. This sample of 100 bids was reduced to 86 for the purposes of the current study, since we needed the same dataset as above for the first, second and third year after the completion period of the takeover event, excl
16、uding at the same time any successive bid made by the initial acquirers over this 3year post-acquisition period. From this final sample of 86 bids, 42 cases concerned listed targets and 44 unlisted targets. The 76% of bids made for the listed target firms were pure stock offers (32 cases), while the
17、 rest 24% were pure cash offers (10 cases). The 86% of bids made for the private target firms were pure cash offers (38 cases), while the rest 14% were pure stock offers (6 cases). So, we can clearly see a preference for cash offers by the acquirers targeting private firms, while exactly the opposit
18、e occurs by bidders targeting listed firms, by having a preference for stock offers. It is likely, that in the cases of privately held targets the shareholders of these firms prefer to share information with the bidders, who consequently face less information uncertainty/asymmetry about the true val
19、ue of those targets, which encourages them to make cash offers. To apply the 3-factor model of Fama Rmt - Rft = the excess return of the market index over the risk-free rate (equity risk premium); SMBt = the difference between the return of a portfolio of small capitalization stocks and a correspond
20、ing portfolio of large capitalization stocks; HMLt = the difference between the return of a portfolio of high book-to market stocks (value stocks) and a corresponding portfolio of low book-to-market stocks (growth stocks); ait = the abnormal stock return (intercept term) per month of the event firms
21、 over the period measured; 1i, 2i, 3i = the sensitivity coefficients of the model; it =a stochastic term which is assumed to have a zero mean and a constant variance. In order to form the SMB and HML portfolios, it is necessary first to classify the acquiring firms of our sample by size (market valu
22、e) and then by BM ratios following the standard method of Fama & French (1993). Thus, for the SMB portfolio formation, we first take the whole sample of stocks which compose the ASE Index, rank them in to small and big size stocks based on their median value of size (market capitalization) and then
23、we classify our sample stocks in the appropriate category (small/big). For the HML portfolio formation we repeat the same ranking procedure by ascending order and then based on this ranking we classified the ASE stocks in the low BM group if they fall in the first 30% of the total number of companie
24、s, and stocks in the high BM group if they lie within the last 30% of the total number of companies.The group of companies which lies in between these groups is classified as the category of the medium BM stocks. The acquiring sample firms are then classified depending on their values in the groups
25、of high and low BM stocks accordingly. Having completed the classification of the acquiring forms in groups of small and big size as well as low, medium and high MB ratios, we need to create the six value weighted portfolios from the intersection of the groups formed by the size and BM factors. The
26、first 3 portfolios (S/L, S/M, S/H) are constructed by the combination of small size and low, medium and high BM stocks, while the other 3 portfolios (B/L, B/M, B/H) are constructed by the combination of big size and low, medium and high BM stocks. In this way, two mimicking portfolios were created,
27、by which the first portfolio reflects the market value factor (SMB) and the second portfolio represents the BM factor (HML). More specifically, the SMB portfolio will be equal to the difference between the average monthly return of the small size stocks and the big size stocks and the HML portfolio
28、will be equal to the difference between the average monthly return of the high BM ratio stocks and the low BM ratio stocks. These can be expressed as: SMB = 1/3(S/L+S/M+S/H) 1/3(B/L+B/M+B/H) (2) HML = 1/2 (S/H+B/H) 1/2 (S/L+B/L) (3) 4 Conclusions In this study we have attempted to examine the long-t
29、erm post bid performance of Greek acquiring firms by using the methodology based on the 3-factor model of Fama and French (1993). First, we found that the Greek acquiring firms experience on average negative abnormal returns of about 2% per month for a period of one, two and three years after the bi
30、d announcement and completion dates. Our results are more or less in agreement with the majority of the empirical studies reported for other markets abroad, though the negative abnormal returns appear to be much higher. Second, we provided some evidence, although the differences were not statistical
31、ly significant, that the Greek acquirers in the long-term post-acquisition period seem to lose more when the target is a listed firm, a finding which is in line with the majority of other published studies for US (Fuller et al., 2002) and UK markets (Draper and Paudyal, 2006). Comparing these findin
32、gs with the corresponding results for the short-term period reported by Kyriazis and Diacogiannis (2008), we can realize that the significant and positive abnormal returns of about 5% which acquirers had reaped during the announcement period turn to a significant and negative average abnormal return
33、 of about 2% in the 3 years after this date. This reversion of abnormal returns implies the possibility of mispricing in the Greek takeover market due to the overvaluation of the anticipated synergies of the merger event. With respect to the impact of the mode of payment on the acquirers gains in th
34、e long-run post-acquisition period, our results are somehow different from those found by Kyriazis and Diacogiannis (2008) for the short-term announcement period, because they are sensitive to the status (listed/unlisted) of the target firms. Thus, the acquirers of listed targets have less negative
35、average monthly abnormal returns in cash offers than in stock offers, while the acquirers of unlisted targets experience slightly less negative returns in stock offers. The first finding is similar to that observed for the short-term announcement performance by the study of Kyriazis and Diacogiannis
36、 (2008) and consistent with the majority of the previous empirical evidence (e.g. Fuller et al., 2002 and Draper and Paudyal, 2006). However, the second finding about the performance of the acquirers of unlisted targets between cash and stock offers moves to a different direction in favour of stock
37、offers. Although, the differences in the abnormal returns according to the medium of payment are not statistically significant, our result in the current study implies a large reversion in the performance of acquirers of private targets in cash offers from the short-term (announcement) period to the
38、 long-term post-acquisition period, suggesting that the overpayment from the side of the Greek acquiring firms is more likely to happen in the case of cash bids targeting privately held firms. Finally, we also provided some evidence, yet statistically insignificant, that in the first year after the
39、bid the small size acquirers and the growth (glamour) acquirers underperform the large size acquirers and the value acquirers respectively, while this trend is reversed for the next two years. This finding which is against the results offered by other empirical studies may be sample specific and can
40、 be attributed to the limitations imposed by the small size of our sample in the construction of portfolios according to the market value and the book-to-market ratio. 译文 希腊收购公司长期岗位承购表现 资料来源 : http:/ 作者： 季米特里亚 1介绍和研究综述 绝大多数关于美国和英国最大和更加有效的市场的实证研究 ,已经建立了目标公司大股东的平均收益和重要的异常收益。 与 此同时， 根据收购公司 股东的经验，平均小于 零
41、 或接近零的异常报酬显著围绕着公告期，这将在长期竣工后的 一段 时期变得非常消极和显著。因此， 为什么 有些企业决定为其他企业的承购出价 ，但 公司财务依然是其中一个未解决的难题。为了回答这个问题，几种理论被开发了 ，并且 有了 各种各样的假设检验。 我们可以分为 二个宽广的类别 ： 理论和假说 ， 与市场效率假说是一致的，直接或间接令人对市场效率的理论产生怀疑。 因此，本次研究的主要目的是探讨长期（第一，二，三年期间）希腊受让人投标后的绩效，与早期报道的基里亚 齐 斯和 迪亚戈吉亚尼斯 2008 关于在19932006 期间通过用收购行动这个典范来阐述短期绩效的研究作比较。 就在法玛三因子模
42、型和 法兰奇 的异常报酬率估计（ 1993）方法的基础上，目前的研究发现，每月招标公告后或完成日期的 12,24 和 36 个月 ， 希腊收购公司负异常显 著 ，共约 2的回 报率。结果的确有在承购者的反常回归的逆向趋向在岗位出价期间，不可能归因于“正确的”财产定价模型 (而不是市场指数模型的三因子模型的 )应用。而 法玛（ 1998）认为，在公告期间，到了预期的协同效应，这意味着市场定价错误和低效率高估。同时，本研究探讨对收购的涨幅的其他一些因素的影 响，如付款（现金 /股票），公司的目标状态（上市 /非上市），与收购的特点。 与早期国际实证研究的调查结果相比较，基里亚 齐 斯和 迪亚戈吉亚
43、尼斯 2008 关于收购希腊市场的文献中，也证实了这一点。 2数据采样 为了执行我们的分析，我们用了 100 个完整 的收购竞价中的首个事例来作为我们的研究对象。这个事例是由基里亚 齐 斯和 迪亚戈吉亚尼斯 2008 的写的早期研究的需求。 这个最初的样品包含了来自一位银行家汤姆森数据库（原名白金议会） 的 1993-2006 年 期间 的 接管活动数据，。这段时间包括两个繁荣的收购活动，即分 1998-2000 年 ， 2004-2006 年期间。 通过选择比较 ，从 最初的 700个 中 ，我们最终了获得 100 个样品： 只完成了国内交易（包括投标人，目标是在希腊注册的公司投标时）； 有
44、一个在后的投标期间企业控制全部变更（投标人增加了他收购的股权超过 50）； 在首次投标被排除前，受让人在连续一年的期间里担任，为了确保市场模式对异常收益的回馈有个明确的预测期 。 为了进一步减少 案 例的数量， 我们必须辨认付款方式 (纯 现金 、纯 股票 )， 并收 集有关合并 公司 的 股价和每日总希腊市场指数的数据流，股息付款及前分区信息日期（计算总股东回报，以及市场公平的月度值）。 从 100 个减少到 86个 投标样品 。因为我们需要研究同样数据的收购事件完成期的第一，第二和第三年以上 ，在同一时间不包括任何连续投标 ， 由在这 个三年后收购期的最初收购。从这 86 个投标，最终 得
45、到 42 例样本中列出的目标和 44 个有关非上市的目标。对所列出的目标公司出价 76是做纯股票信息（ 32 例），其余 24的纯现金收购建议（ 10 例）。该公司为私营目标 ， 86是纯现金收购建议（ 38 例），其余 14是纯股票信息（ 6 例）。所以，我们可以清楚地看到了现金收购 是 私营公司的收购目标选择，同时出现了完全相反的目标 ，由 上市公司的投标人所偏爱的股票提供。在私人持有的目标的情况下，这些公司的股东更愿意分享信息，不至于导致 谁的信息较少因而面临的不确定性或非对称性，鼓励他们 以现金提供真正的价值。 在并购后的一段期间，应用 法玛和 法兰奇 的第三 因素 模式。这个对于 1
46、 年、2 年、 3 年后及其投保公告日后 19932008 期间，收购公司所收集的每月的股价很有必要。 在 整个市场指数的雅典证券交易所（ ASE） ，同一数据集经过同一个时期提取 的价格和市场资本化与 产品上市（骨髓）所 构成比例的公司市场指数。为了与法国法玛 的方法一致 ，我们要在每个年度 6 月底（ t）和帐面价值在上一财政年度 12 月底（ t - 1 年）的市场价值 进行分析。 为了计算零风险率， 我们抽取了希腊政府 3 个月中的短期国库券中的年度利率，把它分成 12 等分，从而得出的月收益率。 3方法论 要检查收购企业长期业绩，我们遵循的法玛三因素模型和 法兰奇 标准的做法（ 19
47、93） 。 我们估计在整个收购的 86 个样本的异常报酬，然后我们使用子样本，以区分上市及非上市公司的收购目标，以及 提供 现金收购和收购的股份 。我们还区分价值和成长的收购，以及大、 小（以市值计算）的收购。三因子模型可表示如下： Rit Rft =it +1i (Rmt - Rft ) +2i SMBt +3i HMLt +it (1) 其中： Rit Rft =事件的公司股票 投资组合的超额回报超过无风险利率； Rmt - Rft =对在无风险利率（股权风险溢价）市场指数超额收益； SMBt =小市值股票和大盘股相应的投资组合投资组合回报之间的差异； HMLt =投资组合回报高的帐面价值
48、对市场的股票（价值型股票）和低 上市股票（成长型股票）相应的投资组合之间的差异的； ait =反常股票收益 (截 止 期限 )每个事件的月变牢固在测量期间； 1i, 2i, 3i =该模型的灵敏度系数； it =随机术语，假定有一个零均值和恒定方差。 为了形成中小企业和会展管理公司的投资组合，有必要 首先按大小（市场价值）进行分类，我们的样本企业的收购和骨髓比率 跟随 法国法玛 的 标准方法（ 1993 年）。因此，对于中小企业的投资组合的形成，我们先看看整个的股票组成样本的 ASE 指数，他们 根据 小级别和大尺寸对他们的规模（市值）的股票中值 进行分类 ，然后我们在分类的样本股 中选择 适
49、当的类别（小 /大）。 对于会展管理公司的投资组合，我们通过升序来保持同样的等级程序，在此基础上，如果它们在所有的公司中跌了前 30%，。我们把美国证券交易所归入到低的管理组，如果它们在所以公司中的最后的 30%的范围之内，这些股票被归入高层管理组。 其中的集 团化企业，在这些集团之间被归类为介质的骨髓股票。样本公司的收购， 根据他们的价值观分为高、 中 、 低的相应骨髓股票。 在完成了收购形式的小型和大型的规模以及低、中、 高市净率群体的分类后，我们需要创造价值，从中 所形成的规模和骨髓因素交集的群体 进行 加权组合。前 3 份股份单 (S/L， S/M， S/H)由小型和低、 中等和高骨髓股票的组合修建，而其他 3 份股份单 (B/L， B/M， B/H)由大小和低、 中等和高骨髓股票的组合修建。这样，创建了两个模拟投资组合，其中第一个投资组合反映了市场价值因素（ SMB）， 第二个组
