آیا بازار برای خرید مجدد سهم بازار آزاد، تحت واکنش زیاد یا کم قرار می گیرد ؟ چشم انداز انگلیس

Using UK open market repurchases, we reject the market underreaction hypothesis and the market overreaction hypothesis proposed by Ikenberry et al. (1995) and Peyer and Vermaelen (2009), respectively. The evidence suggests that the UK market reacts slowly to actual repurchases made by value firms. UK repurchases on average do not suffer from share undervaluation prior to the announcement. Value firms perform just as well as glamour firms during the authorisation period but outperform glamour firms significantly 2 years following the announcement. It turns out that value firms repurchase over 6% more shares than glamour firms during the authorisation period.

Share performance of US open market share repurchases has been extensively studied and two main explanations are offered to explain the seemingly die-hard existence of abnormal returns following the announcement. Ikenberry et al. (1995) propose the market underreaction hypothesis which suggests that the market treats repurchase announcements with scepticism, leading prices to adjust slowly over time (4 years). Some years late, Peyer and Vermaelen (2009) find evidence consistent with the overreaction hypothesis which suggests that open market share repurchases are a response to a market overreaction to bad news prior to the repurchase. Long run abnormal returns are in particular persistent among value firms (high book-to-market ratios) and are strongly related to pre-announcement share undervaluation. Will this phenomenon exist in the UK where characteristics of and regulations governing open market repurchases differ significantly from the US? Despite the similarities in law and language between US and UK markets, UK open market share repurchases have significantly different characteristics and are governed by different law and reporting regulations. According to the listing rules of the London Stock Exchange, listed firms must publish actual repurchases on the regulatory news services (RNS) which are viewable to the market and investors the next business day. In addition, the UK company law (Companies Act 1985) required firms to report actual repurchases (the amount and quantity of share repurchases) in Annual Reports. Based on the repurchase information provided by Annual Reports and the RNS, the market and investors can easily gauge the commitment level of managers towards announced repurchase programmes. Unlike US repurchases, UK repurchases have an authorisational duration between two consecutive Annual General Meetings (usually 12 months). Thus, long run share performance is subject to less influence from other corporate events or economic situations which have taken place following the announcement. As a result, UK open market share repurchases present an excellent opportunity to understand the market reaction to the announcement outside of the US markets. The aims of this paper are not to challenge the market efficiency theory or to investigate the motivations of repurchases, but to answer the following questions. First, are open market repurchases announced by UK firms on average preceded by share undervaluation at the announcement? Second, is the initial market reaction of UK open market repurchases a response by the market to correct its overreaction to bad news prior to the announcement? Third, are open market share repurchases attracting a positive and significant return following the announcement? In particular, how does book-to-market affect long run abnormal returns? Finally, would the impact of book-to-market on long run abnormal returns suggest market underreaction to repurchase news or overreaction to bad news? Relying on a sample of 468 repurchases announced between January 1999 and December 2004, we examine share performance surrounding and following the announcement with consideration of characteristics of firms and repurchases. We first consider short-term returns surrounding the announcement. The average market return of sample firms prior to the announcement is statically positive and reveals limited evidence of share undervaluation. The average market return surrounding the announcement is positive and statistically significant, consistent with the literature. Further, when the sample is segmented by size quartiles and book-to-market quintiles, small firms and glamour firms (low book-to-market ratios) appear to earn higher 5-day initial market returns than big firms and value firms. However, announcement abnormal returns of small firms can hardly compensate the share undervaluation these firms suffer before the announcement. Firms in other size quartiles and book-to-market quintiles show no sign of pre-announcement share undervaluation. Finally, the regression analysis shows that the initial market return is unrelated to characteristics of firms and repurchases such as size, book-to-market and the percentage of repurchase. We find no evidence to suggest that the initial market reaction is a response to share undervaluation prior to the announcement. We then investigate the long run share performance. Long run abnormal returns can be influenced heavily by different measurement methods (Barber and Lyon, 1997, Lyon et al., 1999, Kothari and Jerold, 1997 and Mitchell and Stafford, 2000). Here, two different methods are used to determine whether our sample firms on the whole exhibit abnormal share performance 1 and 2 years following the announcement. It is evident that our sample firms outperform matched size and book-to-market portfolios in the long run using any of these two measurement methods. One year buy-and-hold abnormal returns of value firms are only slightly higher than these of glamour firms. Further analysis reveals that the relationship between book-to-market and 1-year buy-and-hold abnormal returns disappears after controlling size, percentage of shares sought, actual repurchases and other factors. On the other hand, book-to-market is a dominant factor in determining 2-year buy-and-hold abnormal returns. The average 2-year buy-and-hold abnormal return of value firms is 15.86% (p = 0.004) while it is 6.74% for glamour firms (p = 0.154). Firm size does not have significant influence on long run abnormal returns in spite of the fact that small firms suffer from significant share undervaluation prior to the announcement. The results suggest that long run abnormal returns can hardly be explained by the market overreaction hypothesis or the market underreaction hypothesis. We also notice that actual repurchases during the authorisation period are not influenced by share price movement. However, the analysis of actual repurchases shows that large firms repurchase over 4% more shares than small firms while value firms actually repurchase 6% more shares than glamour firms during the authorisation period. Evidently, share undervaluation is not the motivation for high actual repurchase activity among large and value firms. Moreover, long run abnormal returns are not affected by the percentage of share sought at the announcement. These results suggest that long run abnormal returns among value firms are likely to be the market's delayed reaction to actual repurchases. We find that share performance of UK repurchases is similar in many ways to that of US repurchases. For example, UK repurchases generate positive initial market returns in the short-term and have outperformed their matched size and book-to-market portfolios in the long-term. In particular, long run abnormal returns are significantly high among UK value firms. On the other hand, we notice several differences in share performance between US and UK repurchases. In the short-term, we are unable to find the evidence to suggest that UK repurchase on average suffer from share undervaluation prior to the announcement. There is no evidence to indicate that the market reacts excessively to bad news prior to the announcement. In the long run, abnormal returns among repurchases are only related to book-to-market. It seems that actual repurchases conducted by value firms promote significantly high abnormal returns 2 years following the announcement. More critically, we are the first to examine the statistical impact of different announcement dates on share performance. Previous papers recognise that UK repurchases do not have a US style of announcement dates, but the impact of different definitions of announcement dates on share performance has never been examined (Rau and Vermaelen, 2002 and Oswald and Young, 2004). Our results reveal that different definitions of announcement dates do not have significant influence on share performance. Our final result is surprising. We interpret the relationship between book-to-market and abnormal returns as an indication of market underreaction to actual repurchases by value firms which are different from the results of Ikenberry et al. (1995) and Peyer and Vermaelen (2009). The findings of this work provide an alternative explanation to the findings reported by prior studies using American, Canadian and Hong Kong data. Previous papers largely explain long run abnormal returns among value firms as evidence to support the share undervaluation hypothesis or the market delayed reaction to the announcement. The main problem in using open market share repurchases to signal share undervaluation is that open market share repurchases are not firm commitments and that managers have discrete power to implement or discard the repurchase programme and the market and shareholders are fully aware of the flexible nature of repurchase programmes. If managers feel so strongly about share undervaluation, there are plenty of other costly payout signals such as dividend increases and fixed price tender offers which can be adopted by managers to make an immediate and strong statement regarding perceived share undervaluation. It is also at odds to relate long run abnormal returns to a corporate event, a share repurchase programme, which was announced 3 or 4 years earlier, not to mention the fact that there must be plenty of interim and final results announcements made during the 3 or 4 year period following repurchase announcements. These papers also seem to ignore the possibility that long run abnormal returns are the market reaction to actual repurchases made by these firms. Our findings link long run abnormal returns earned by value firms with actual repurchases. In addition, Von Eije and Megginson (2008) reveal that UK firms are the largest and most prolific share repurchasers in Europe and share repurchases in the UK have been triggered by privatizations during the 1980s. Open market share repurchases were first legalised in the UK in 1981 while most European countries prohibited share repurchases until 1998 (Lasfer, 2001). Even though many European countries have civil law systems compared with the common law system in the UK and Ireland, the findings of UK papers would be helpful to law makers in European countries. The RNS news published on the London Stock Exchange are open to the market and shareholders which would help them make sell or non-sell decisions when companies offer to buy back their shares. If shareholders in the long run would benefit from holding shares in firms which purchased all or part of shares as promised, shareholders and the market would place a great emphasis on actual repurchase reporting which can be viewed by the public. Recently, it has become relatively common for US firms to report their actual repurchases in their IQ-Qs and 10-Ks, though the exact timing remains unreported (Zhang, 2005). It is clear evidence that a good practice in the UK can spread to other countries. Finally, since a lot of English speaking countries and regions around the world have been significantly influenced by English corporate law and stock trading regulations, the findings of this paper would motivate researchers to look into the relationship between actual share repurchases and share performance of repurchase announcements in these countries. This study is organised as follows. Section 2 reviews the literature on share repurchases. Section 3 describes the data, share performance measurement, significance tests and variables. Section 4 presents empirical results and Section 5 concludes.

The unique characteristics of UK open market repurchases have provided us an opportunity to examine whether extraordinary share performance surrounding and following the announcement is only a US phenomenon. This paper aims to determine whether share performance is a correction to the market overreaction to bad news prior to the announcement or is the market delay's reaction to repurchase news. Thus, the analysis of this paper is centred on share performance surrounding and following the announcement. We first consider short-term returns surrounding the announcement. UK repurchases on average do not suffer from pre-announcement share undervaluation. In two occasions where small firms and repurchases announced in 1999 suffer from significant share undervaluation, the average announcement returns are hardly large enough to encourage firms to use repurchases as a signal for pre-announcement share declines. It appears that UK repurchases are not preceded by the market overreaction to bad news. Looking at the announcement abnormal returns, there is some evidence to refute share undervaluation. Large repurchase programmes are received less favourably by the market while glamour firms have a higher announcement abnormal return than value firms. The evidence suggests that the announcement abnormal returns are not a response to the market overreaction to bad news before the announcement. We then consider long run share performance following the announcement. Two measurement methods show that UK repurchases on average attract a positive and significant return during the authorisation period and 2 years following the announcement, conditional on size and book-to-market. When analysing the causes of long run abnormal returns, we find that the size impact is not consistent under the two measurement methods. The small number of small firms might have influenced the buy-and-hold long run abnormal returns while the calendar time approach is more suitable to calculate long run abnormal returns of small firms. The calendar time approach reveals that size is not the dominant factor in determining abnormal returns. Small firms underperform large firms 1 year and 2 years following the announcement. The impact of book-to-market under these two measurement methods suggests that value firms outperform glamour firms 2 years following the announcement. On the other hand, value firms do not outperform glamour firms during the authorisation period (1 year following the announcement). The regression results show that size does not influence long run abnormal returns while book-to-market is a dominant factor in 2-year buy-and-hold abnormal returns. Further analysis reveals that such a return pattern can be explained by actual repurchase activity. Small firms which suffer from pre-announcement share undervaluation repurchase fewer shares than large firms which enjoy positive and significant returns before and around the announcement. In contrast, value firms repurchase considerably more shares than glamour firms during the authorisation period. Thus, we conclude that the UK market underreacts to actual repurchases made by value firms. Finally, this paper contributes to the literature by examining share performance of UK repurchases governing by regulations and law which are different from US repurchases. We find some similarities between US and UK repurchases. Like US repurchases, UK repurchases attract positive and significant short-term abnormal returns surrounding the announcement and long run abnormal returns following the announcement. Two-year buy-and-hold abnormal returns are particularly high among value firms. The differences between US and UK repurchases are striking too. Unlike US repurchases, UK repurchases on average do not suffer from share undervaluation prior to the announcement. There is no evidence to indicate that the market reacts overly to bad news prior to the announcement. In the long run, abnormal returns among repurchases are only related to book-to-market. It seems that actual repurchases conducted by value firms promote significant high abnormal returns 2 years following the announcement. This conclusion is different from the market underreaction hypothesis or the market overreaction hypothesis. We also investigate the impact of different announcement dates on short-term and long-term share performance. The results suggest that share performance of UK share repurchases are not statistically influenced by different kinds of announcement dates. The abnormal long run share performance following open market share repurchases seems to be a universal phenomenon and has not died away after it is first discovered by US studies. It is interesting for future research to investigate whether a portfolio of repurchasing firms would outperform benchmark market, size, value and momentum portfolios in the long run.