Cornell University
Library
Cornell UniversityLibrary

eCommons

Help
Log In(current)
  1. Home
  2. Cornell University Graduate School
  3. Cornell Theses and Dissertations
  4. A Cross-Time Study Of U.S. Earnings, Splits, And Dividends Data

A Cross-Time Study Of U.S. Earnings, Splits, And Dividends Data

File(s)
Motelson, Kerry.pdf (1.54 MB)
Permanent Link(s)
https://hdl.handle.net/1813/13769
Collections
Cornell Theses and Dissertations
Author
Motelson, Kerry
Abstract

This paper details the share price reaction to dividend, earnings, and stock split announcements over a 37-year period. It first considers whether there is differential information content in similar corporate news announcements for different types of firms. Second, it investigates whether the value of news information about these firms has declined over time (addressing the question of whether news has become "less newsworthy"). We go on to study the relationship between stock price reactions to corporate news announcements and characteristics of the firms. Operating under the assumption that news announcements have an asymmetrical impact on stock price according to factors like firm size, years of being publicly traded, or industry classification, we categorize firms by whether their corporate news announcements will be more or less valuable to the public. For example, since the public may know more about larger firms, we expect the market to react less strongly (in absolute value) to new information from large firms. We find strong support for this hypothesis. We find little evidence that is consistent with the idea that "news has become less newsworthy" over the past four decades. However, although we do find that the share price reaction to "good" dividend news has become less positive and to "bad" dividend news has become less negative over time, no such related evidence exists for stock splits and earnings announcements. We also find an increase in standard deviation of three day returns around earnings and splits announcements over time, with noteworthy convergence amongst positive, negative and neutral earnings announcements. Additional investigation of entire distributions of returns using kernel density estimators also rejects the "news is no longer newsworthy" idea.

Date Issued
2009-10-09T13:41:09Z
Type
dissertation or thesis

Site Statistics | Help

About eCommons | Policies | Terms of use | Contact Us

copyright © 2002-2026 Cornell University Library | Privacy | Web Accessibility Assistance