بازار سهام UK در سراسر سابق تاریخ تقسیم

Using UK stock market data this study unveils positive abnormal returns on and around the ex-split date. These excess returns are partially predictable using the publicly available information prior to the ex-split date. There is also a persistent increase in the post-split volatility of these stocks with the results being robust to the choice of the volatility proxy. Post-split volatility is found to be positively related to trading activity. Contrary to the US findings, volatility dynamics following the stock split are better captured by changes in the daily trading volume rather than by the number of trades.

Some investors may view stock splits as advantageous, but there is little evidence that they are benefited in any meaningful way. In many cases, stock splits are seen as a sign of management optimism and investors appear to share this view. Based on market efficiency these favorable managerial signals should be immediately incorporated in the stock prices. Nonetheless, expectations based on theoretical frameworks are often contradicted by empirical evidence. The question of interest is, therefore, ‘how do markets react when it comes to stock splits?’ A number of studies have attempted to explain this event and assess both the short- and long-run performance of stock returns/volatility following the corporate announcements. Differential behavior between split and non-split stocks has been documented indicating that this non-economic action does affect the shareholders’ wealth. Research in the area unveils a rise in risk-adjusted returns following the announcement and ex-split dates (Grinblatt et al., 1984, Ikenberry et al., 1996 and Desai and Jain, 1997) as well as an increase in equity return volatility (Ohlson and Penman, 1985, Dubofsky, 1991 and Koski, 1998). If such an overreaction persists for a period of time, then concerns are raised regarding the speed of price adjustment to information flow, and potentially invalidate (Fama, 1998) efficient market approach. In parallel, work on corporate events suggests that markets appear to under-react to news (Daniel et al., 1998). On the theoretical front, four main hypotheses are put forward to explain market reactions to stock splits. The first one, the signaling hypothesis, postulates that such actions aim to reduce information asymmetries between shareholders and management regarding the firm's financial prospects1. The second approach, the optimal price range hypothesis, states that managers believe that when shares trade within a certain price range, their decision to split the stocks will enhance their liquidity. This ‘optimal’ price is typically set at the historical average price of the firm's equity, or of the market/industry as a whole. The third approach, the liquidity hypothesis, can be seen as a hybrid form of the signaling and the optimal price range hypotheses, and posits that post-split liquidity enhancement benefits from both the lower price range and positive signals conveyed on the split announcement. The former is supported by Copeland (1979) who states that there is an optimal price range wherein stocks are most liquid. The fourth approach, the tick size hypothesis, suggests that stock splits increase the tick size relative to the stock price thereby boosting the profitability of market making (Schultz, 2000 and Angel, 1997). On the empirical front, Lamoureux and Poon (1987), Schultz (2000) and Angel et al. (2004) find an increase in institutional ownership following a split, while Ikenberry et al. (1996) and Dennis and Strickland (2003) refute these findings. Examining the equity performance of firms executing stock splits, Fama et al. (1969) and Byun and Rozeff (2003) found insignificant abnormal returns, which contradicts the results of Desai and Jain (1997) and Wu and Chan (1997). Muscarella and Vetsuypens (1996) and Mohanty and Moon (2007) report higher liquidity following the stock splits, while others report the opposite (Copeland, 1979, Murray, 1985 and Lamoureux and Poon, 1987). Hwang et al. (2005) observe that an unexpected stock split yields abnormal return within the first 3 months, but the effect disappears over longer horizons for both expected and unexpected splits. The post-split volatility increase has also received extensive attention in the empirical literature, but there is no consensus as yet on the reasons behind this phenomenon. Some researchers have highlighted the role of market microstructure effects, such as bid-ask spread and price discreteness, which introduce noise in volatility measurement (Ohlson and Penman, 1985 and Ball, 1988). Another explanation relies on the impact of splits on trading activity, which places the rise in the daily number of (small) trades as the main driver behind the post-split volatility increase (Jones et al., 1994 and Kamara and Koski, 2001). The literature has put forward a number of reasonable explanations for both the rise in the risk-adjusted returns and their volatility following the split event, but the majority of studies have focused on the US market. Thus the objectives of the present work are as follows. First, this is the first study, to our knowledge, considering the equity behavior around stock splits in the UK market. Focusing on the UK market is crucial since (a) there is a bi-directional trading and investment relationship between the UK and the US, (b) the UK is a financial centre interacting continuously with European, Asian and the American markets, and (c) the extent to which the findings for the US firms hold in another major market can be assessed. Second, the study looks at the excess returns and equity volatility surrounding the ex-date rather than the announcement date. Anecdotal evidence of ex-date abnormal returns has been the impetus of this research. The occurrence of ex-date excess returns is rather surprising, as this date is known in advance and lacks the material information2 to justify any abnormal market reaction. Yet ex-date abnormal returns have been documented by a few US studies (Dravid, 1987 and Julio and Deng, 2006) with mixed explanations. The third goal of the current research is to provide a robust analysis of the factors affecting the UK equity performance for both returns and volatility following the split event (ex-date). The ability to identify these factors is of great importance to market participants, fund managers, and academic scholars as they have a notable impact on portfolio selection, asset allocation, financial pricing, and corporate strategy. Finally, turning to the methodological framework, a GARCH-based approach (Corhay and Tourani Rad, 1996) is employed as the platform for scrutinizing excess returns. In order to examine the volatility behavior, two different volatility proxies, both adjusted for microstructure biases (Kaul and Nimalenndran, 1990), are employed. Since volatility estimates embrace both systematic and unsystematic exposure, an adjustment is made to ensure that analyzed proxies mirror the idiosyncratic component of the firm's return volatility. This adjustment has not been previously applied to the analysis of volatility behavior around the event dates. In what follows, the paper presents the hypotheses tested in Section 2. The data and methodology employed are described in Section 3. Section 4 discusses the empirical findings, and Section 5 concludes the paper.

Stock splits have received great attention in the finance literature. Various studies have attempted to explain and examine the existence of the excess risk-adjusted returns and the increase in returns ‘volatility following the split. While most papers concentrate on the stock market behavior around the announcement date, the reality of increased risk-adjusted returns as well as returns’ volatility around the effective date has not been conclusively analyzed. The current work delves into these issues using a sample of stock splits in the UK market. The results suggest that the UK firms experience positive abnormal returns on and around the ex-split date. These abnormal returns may be explained on the basis of information publically available prior to the split ex-date. The model proposed in this paper is able to explain approximately 24% of the market abnormal reaction patterns registered on the ex-date. The analysis of the returns volatility behavior shows that even after controlling for microstructure/market biases, there is a statistically significant increase in the volatility figures following the event date. The latter is unaffected by the choice of the volatility proxies employed. Regression analysis suggests that there is a strong and positive relationship between the measures of trading activity and the returns’ volatility over the pre- and post- split horizons. However, when analyzing the changes in volatility behavior, following the split event, the post-split volatility changes appear to be better captured by the changes in the daily trading volume rather than the daily change in the number of trades. This observation provides evidence leading to a disagreement with the findings reported for the US market where the number of trades is found to be the key measure determining the volatility behavior.