How earnings releases cool off Sentiment-driven expectations

Earnings announcements play a significant role in disciplining market expectations fueled by news sentiment, concluded professors Eric Holzman and Brian Miller from Indiana University and Brady Twedt from Texas A&M University. Their paper “Curbing Enthusiasm: Media Sentiment and the Disciplining Role of Quarterly Earnings Announcements” uses RavenPack Analytics to quantify media sentiment and its correlation with the stock market's reaction to earnings announcements, highlighting their importance in mitigating mispricing.

Background

In recent years, there has been growing recognition of the impact that sentiment driven by media coverage can have on market dynamics and asset prices. Sentiment-driven expectations can be a concern in financial markets for several reasons:

When sentiment becomes excessively positive or negative , it can create market bubbles or panics, leading to asset price distortions and volatility. These distortions can result in mispricing, where the market value of an asset deviates from its intrinsic value, potentially leading to investment losses.

Secondly, sentiment-driven expectations can also contribute to herd behavior and irrational investment decisions .

Finally, when market sentiment becomes disconnected from economic fundamentals , it becomes difficult to assess and manage systemic risks. It can also create challenges in maintaining market stability and integrity.

Paper Methodology

• The authors worked with a comprehensive database of over 6.2 million media articles from over 6400 media outlets , sourced from RavenPack. They retained only articles that were fully relevant to the firm of interest, with a relevance score of 100 and excluding press releases.
• Further, they utilized RavenPack Sentiment Indicators and a dataset comprising 108,666 firm-quarters from 2007 to 2017 to examine media sentiment.
• To measure daily media sentiment, they used RavenPack's Event Sentiment Score (ESS) , adjusted to be centered around neutral sentiment. By summing these adjusted scores for all articles published about a firm on a given day, they obtained a measure of daily firm-specific media sentiment. This daily sentiment was then aggregated over a window of [-60,-3] days to derive the initial measure of quarterly media sentiment.
• The authors also developed novel measures of quarterly media sentiment, accounting for factors like media outlet influence and article novelty.
• Additionally, they created a weighted measure of media sentiment that isolated the impact of sentiment on company stock returns, independent of common risk factors and fundamental news.

Key Findings

Negative Correlation

Between quarterly media sentiment and the stock market's reaction to earnings announcements suggests that earnings announcements play a crucial role in curbing mispricing resulting from media sentiment.

Higher order belief framework

Investors with varying information and horizons may deviate from fundamentals for short-term gains. Media affects average investor trading, causing price overreactions. Earnings announcements correct these overreactions, influencing trading in the opposite direction.

Variations in strength

Correlation strength varies based on factors such as retail and transient institutional investors, fundamental uncertainty, and short-selling constraints. These factors indicate increased mispricing linked to media sentiment.