Contrast effects in reactions to earnings announcements

Loading...
Thumbnail Image

URL

Journal Title

Journal ISSN

Volume Title

School of Business | Master's thesis

Date

2022

Major/Subject

Mcode

Degree programme

Finance

Language

en

Pages

44 + 6

Series

Abstract

Contrast effect is a behavioral error, where a related prior observation has an inverse effect on the perception of the following one. New evidence in financial literature shows that the effect has potential to induce short-term mispricing when investors contrast consecutive pieces of earnings announcement news on one another. The purpose of this paper is to revisit the topic and to study the effect by employing a new set of data consisting of large publicly listed U.S. companies in the 21st century. The results presented in this paper are inconclusive and to an extent contradicting with prior literature. The results of the full-sample OLS regressions show no evidence of a contrast effect between any proxy for a salient earnings surprise and the short term returns of firms announcing their earnings the following day. However, splitting the sample into subsamples based on business cycles shows that, during the longest continuous economic expansion period from 2009 to 2020, the contrast effect is economically and statistically significant, albeit not robust to the inclusion of year-month fixed effects. During this period, the mispricing persists for up to five trading days, and is driven by consecutive announcements by firms in the top NYSE size decile.

Description

Thesis advisor

Suominen, Matti

Keywords

behavioral finance, contrast effect, earnings announcement, mispricing

Other note

Citation