Examining Initiation of Dividend Payments
Dividend is a big thing in Finance. Much of shareholders’ return comes in the form of dividend payments. Yet, firms are not obligated to pay dividends. Being a choice of the firm, it is important to understand the implications of the firm’s decision on whether to payor not to pay dividends.
In this assignment we will examine instances where firms start paying dividends. We base our analysis on a sample of firms that announce the initiation of dividend payments. Eachdatapoint consists of a firm and the date it announced that it would start paying dividends (and the condition that the firm had not paid any dividends in the 10 years prior to that announcement). For each data point of our study—that is, for each firm and the date when it announced that it would start paying dividends—we also include another firm that did not announce initiation of dividends on that date. Data on dividend initiations and matched firms appear in the dataset “d_div.sas7bdat” (available on Canvas).
Figure 1 illustrates the sample by showing the first few observations in the dataset. For each observation, the column INIT identifies whether it refers to a firm that initiates dividend (INIT=1) or does not initiate dividend (INIT=0). Take the first record. It says that the firm with PERMNO=13936 (“PHILLIPS VAN HEUSEN CORP”) declared in DCLRDT=April 8th, 1970 that it was going to start paying dividends. The next row, with INIT=0, has the identification of a matched firm. More specifically, the firm with PERMNO=28265 (“UNITED FINANCIAL CORP CA”) had not paid any dividends in the 10-year period prior to April 8th, 1970, and kept on not paying dividend in that date.
_n_ pair init permno comnam dclrdt
1 1 1 13936 PHILLIPS VAN HEUSEN CORP 8-Apr-70
2 1 0 28265 UNITED FINANCIAL CORP CA 8-Apr-70
3 2 1 27713 PERKIN ELMER CORP 19-Nov-70
4 2 0 44644 AUTOMATIC DATA PROCESSING INC 19-Nov-70
5 3 1 27115 GREAT WESTERN FINANCIAL CORP 10-Apr-72
6 3 0 31835 FEDERALS INC 10-Apr-72
… … … … … …
Figure 1. Example of data on dividend announcements
Why are we examining dividends? The baseline theory in Finance is that dividends do not matter: whether you return your money to shareholders via dividends or capital gains is irrelevant. On the other hand, people know that it is bad news when a firm decides to stop paying dividends (see Figures 4.4 and 4.5, module 4)—which suggests that dividends have valuation implications for the firm. We complement the analysis in module 4 by looking at situations when firm start (rather than quit) paying dividends.
There are at least three relevant questions that can be examined with respect to dividend initiations:
Do dividends have valuation implications? If so, markets should react to the firm’s decision to start paying dividends. If dividends convey good news about the firm,market reactions should be positive.
What are the determinants of market reactions to the firm’s announcement that dividends will start being paid. In part I, one examines if markets react to the announcements, while here you examine in a regression framework what are the determinants of the market reactions to such announcements.
III.What characteristics seem to drive the firm’s decision to start paying dividends?
Part I: Valuation Implications
We start our investigation by answering question (I):
Do dividends have valuation implications? If so, markets should react to the firm’s decision to start paying dividends. If dividends convey good news about the firm, market reactions should be positive.
You will address this question with an event study—employing the technique covered in module 4. For the sample of firms that announced the initiation of dividend payments, you will examine and show the pattern of abnormal and cumulative abnormal returns over the 11-day window around the announcement day.
For the definition of abnormal return, use the CAPM-adjusted return model—that is, define abnormal return using the following expression (from section 12.3):
𝐴𝑏𝑛𝑜𝑟𝑚𝑎𝑙 𝑅𝑒𝑡𝑢𝑟𝑛௧ = 𝑅௧ − (𝛼 + 𝛽 𝑅௧ )
where Rit is the firm’s stock return at time t (the variable RET in the CRSP dataset “dsf.sas7bdat”, located in “/wrds/crsp/sasdata/a_stock”), Rmt is the market return at time t (the variable VWRETD in the CRSP dataset “dsix.sas7bdat”, located in “/wrds/crsp/sasdata/a_indexes”), and αi and βi are parameters from the CAPM model. The parameters αi and βi should be estimated based on data from the window [DCLRDT 270,DCLRDT–21]. That is, for an announcement with the values (PERMNO, DCLRDT),one collects data on that firm’s return and the market return using the window [DCLRDT–270,DCLRDT–21] (that is, from 270 calendar days before the announcement to 21 calendar days before the announcement), then regress the firm’s return on the market return, and αi and βi will the intercept and the coefficient on the market return, respectively.
Important: notice that the abnormal return here is different from the abnormal return used in module 4. In module 4 we employed the market-adjusted return model while here we use the CAPM-adjusted model.
Please examine formally whether markets react positively to dividend initiations. Prepare and show a table showing average abnormal returns, t-stats and p-value for 11-day window around announcements. Also create and show a graph with the pattern of average cumulative abnormal returns over the same 11-day window. Anchor your inferences on formal hypotheses testing. Also, examine whether markets respond efficiently to news in dividend initiations (for this you can assume that some dividend announcements happen after the close of the market—that is, market reactions to such announcements could happen up to one day after the event date).
(Hint: notice that the examination here is based on 551 events—the observations in the dataset for which INIT=1.)
One criticism of the event study is that it might be capturing something else that happens around the announcement day. Let’s say, perhaps some other news that happen to be released together with announcements of dividend initiations affect many firms out there— including the ones announcing dividend initiations.
The use of abnormal returns aims at addressing this possibility. Nevertheless, another way to address this criticism is to employ a placebo test. For each data point of our event study—that is, for each firm and its announcement that it would start paying dividends—one can draw another firm—a matched observation—that did not announce initiation of dividends and observe what happens with this matched firm in the day the original firm issued its announcement. For example, if the matched firm is collected such that it belongs to the same industry as the firm announcing dividend initiation and it happens that there is some good news for the industry at the date of the dividend initiation (DCLRDT), then it might be the case that the effect we capture in the event study above comes from the industry good news rather than the dividend announcement.
Hence, let us run an event study on the returns of the matched firms around the date of the original firms’ announcement of initiation of dividends. More specifically, repeat the event study except that now you examine the firms with INIT=0 instead of the firms with INIT=1.
Please generate and show a new set of outputs for this new event. Examine whether there is any market reaction to the matched firms around the announcement dates.
(Hint: again the examination is based on 551 events—but now the relevant observations are the ones in the dividend dataset for which INIT=0.)
EasyDue™ 支持PayPal, AliPay, WechatPay, Taobao等各种付款方式!
E-mail: email@example.com 微信:easydue