Measuring Swedish Investor Sentiment and its Effect on Stock Market Response to Earnings News

Sammanfattning: The purpose of this thesis is to examine whether the Swedish stock market reacts differently to earnings announcements during periods of high investor sentiment than during periods of low investor sentiment. We define investor sentiment as investor beliefs about future cash flows or discount rates not supported by prevailing economic and financial fundamentals. Research in the field of behavioral finance has found that when investor sentiment is high, investors place excessively optimistic valuations on future cash flows and discount rates, whereas when investor sentiment is low, investors are excessively pessimistic. Therefore we hypothesize that the stock market should react more strongly to good news during periods of high sentiment, and more strongly to bad news during periods of low sentiment. Using six proxy variables for investor sentiment previously employed in investor sentiment research, we construct a composite sentiment index controlled for macroeconomic indicators. We then use standard earnings response methodology, and test whether the earnings response coefficient is significantly different for positive and negative earnings announcements during periods of high and low sentiment. We also test whether firm size can explain the effect of investor sentiment on the cross-sectional returns surrounding the earnings announcement. Further, we investigate whether accounting earnings' information content co-varies with degrees of investor sentiment.The major finding in this study is that the earnings response coefficient for good news is significantly larger during periods of high sentiment than during periods of low sentiment, and that the earnings response coefficient for bad news is significantly larger during periods of low sentiment than during periods of high sentiment. We find that the information content of earnings follow this pattern, with a larger information content inherent in good news during high sentiment and bad news during low sentiment. These findings are consistent with previous research on, and central theories of, investor sentiment. Further, we are unable to show that firm size is a relevant factor in determining the outlet of investor sentiment on the cross-sectional stock returns.