This study examines the markets reaction to announcements of dividend increases. In particular, it considers the factors that affect the magnitude of abnormal returns during the days that surround announcements of dividend increases. The objective is to find whether the market reaction to dividend increases has weakened with the passage of time and whether market conditions affect the reaction. Eventually, this study is expected to reveal whether dividends continue to be important to investors. <p>This research is motivated by the findings of Fama and French (2001). They suggest that since 1978 firms have had a declining propensity to pay dividends. They propose that dividends are declining as a result of the ease by which investors can make homemade dividends through selling small portions of their holdings. They argue that recent market developments, particularly the introduction of negotiated commissions and discount brokers, have made homemade dividends easier and less costly. Their results may suggest that investors are now less interested to receive dividends than in the past. One objective of this study is to examine whether investors preferences regarding dividend payments have changed over time. This is accomplished by measuring the abnormal returns following announcements of dividend increases. Benartzi, Michaely, and Thaler (1997) suggest that the reaction of the market to dividend increases is an acceptable method of determining the value of dividends to investors. <p>In addition, this study explores the theoretical factors that may affect dividend valuation. Previous studies, such as Allen, Bernardo and Welch (2000), suggest that the existence of debt holders and institutional investors reduce the potential for agency costs as these stakeholders monitor managers. In contrast, Jensen (1986) suggests that high cash flows make it easier for managers to spend on perquisites and empire building. Thus, the potential for agency costs increases. Therefore, paying dividends when cash flows are high reduces the likelihood of agency costs. At the same time, Benartzi, Michaely and Thaler (1997) suggest that increasing dividends following higher cash flows signals managements expectation that future performance warrants a dividend increase. Consequently, the agency and signaling theories suggest that investors may react positively to dividend increases when cash flows are high. <p>Several observations are obtained from this study. First, investor reaction to dividend increases seems to have weakened over time. Second, the reaction is different when the increase is announced in a bear market rather than in a bull market. Third, the market reaction to dividend increases is less in firms that are more liquid. This finding may be interpreted as evidence that dividends are valued less in more liquid firms because it is easier for the investors of these firms to make homemade dividends. Fourth, the magnitude of the reaction is directly related to the increase in trading volume following the announcement. <p>Surprisingly, the evidence disputes the predictions of the agency cost theory of dividends. This theory states that dividends are valued because they decrease the amount of cash available to management, which in turn decreases the potential for waste. Given this theory, it is expected that firms with high debt loads already have agency costs decreased so the market reaction to their dividend increases would be less than other firms while firms with high free cash flows would have a greater market reaction to their dividend increases because of the large potential for waste on managements part. Instead, the results suggest that firms with high debt loads experience positive market reaction following dividend increases while firms with large free cash flows experience negative reactions. It seems that the signaling theory of dividends is contributing heavily to this result.<p>Future research should be directed to investigate the possibility that share repurchases may be replacing dividends as a way to redistribute surplus cash to shareholders. In addition, future studies may focus on the signaling theory of dividends as useful tool to explain the dividend policies of corporations.

This study examines the markets reaction to announcements of dividend increases. In particular, it considers the factors that affect the magnitude of abnormal returns during the days that surround announcements of dividend increases. The objective is to find whether the market reaction to dividend increases has weakened with the passage of time and whether market conditions affect the reaction. Eventually, this study is expected to reveal whether dividends continue to be important to investors. <p>This research is motivated by the findings of Fama and French (2001). They suggest that since 1978 firms have had a declining propensity to pay dividends. They propose that dividends are declining as a result of the ease by which investors can make homemade dividends through selling small portions of their holdings. They argue that recent market developments, particularly the introduction of negotiated commissions and discount brokers, have made homemade dividends easier and less costly. Their results may suggest that investors are now less interested to receive dividends than in the past. One objective of this study is to examine whether investors preferences regarding dividend payments have changed over time. This is accomplished by measuring the abnormal returns following announcements of dividend increases. Benartzi, Michaely, and Thaler (1997) suggest that the reaction of the market to dividend increases is an acceptable method of determining the value of dividends to investors. <p>In addition, this study explores the theoretical factors that may affect dividend valuation. Previous studies, such as Allen, Bernardo and Welch (2000), suggest that the existence of debt holders and institutional investors reduce the potential for agency costs as these stakeholders monitor managers. In contrast, Jensen (1986) suggests that high cash flows make it easier for managers to spend on perquisites and empire building. Thus, the potential for agency costs increases. Therefore, paying dividends when cash flows are high reduces the likelihood of agency costs. At the same time, Benartzi, Michaely and Thaler (1997) suggest that increasing dividends following higher cash flows signals managements expectation that future performance warrants a dividend increase. Consequently, the agency and signaling theories suggest that investors may react positively to dividend increases when cash flows are high. <p>Several observations are obtained from this study. First, investor reaction to dividend increases seems to have weakened over time. Second, the reaction is different when the increase is announced in a bear market rather than in a bull market. Third, the market reaction to dividend increases is less in firms that are more liquid. This finding may be interpreted as evidence that dividends are valued less in more liquid firms because it is easier for the investors of these firms to make homemade dividends. Fourth, the magnitude of the reaction is directly related to the increase in trading volume following the announcement. <p>Surprisingly, the evidence disputes the predictions of the agency cost theory of dividends. This theory states that dividends are valued because they decrease the amount of cash available to management, which in turn decreases the potential for waste. Given this theory, it is expected that firms with high debt loads already have agency costs decreased so the market reaction to their dividend increases would be less than other firms while firms with high free cash flows would have a greater market reaction to their dividend increases because of the large potential for waste on managements part. Instead, the results suggest that firms with high debt loads experience positive market reaction following dividend increases while firms with large free cash flows experience negative reactions. It seems that the signaling theory of dividends is contributing heavily to this result.<p>Future research should be directed to investigate the possibility that share repurchases may be replacing dividends as a way to redistribute surplus cash to shareholders. In addition, future studies may focus on the signaling theory of dividends as useful tool to explain the dividend policies of corporations.