Three essays on financial markets and news announcements

Abstract

This thesis focuses on the relation between financial markets and news announcements, comprising three empirical essays. The first essay (Chapter 2) investigates the predictability of realized moments on the cross-section of stock returns around earnings announcements. I construct realized measures (variance, skewness, kurtosis, and relative jump) using high-frequency intraday stock prices. My results show that realized variance, skewness, and relative jumps strongly predict stock returns around earnings announcements, but realized kurtosis does not. These findings are robust to various event windows, after controlling for firm characteristics, and to a range of additional tests. I further show that the predictability of realized measures is not affected by unexpected earnings. The findings also suggest that pre-announcement realized measures absorb part of the information contained in unexpected earnings.The second essay (Chapter 3) examines the stock market reaction to macroeconomic news announcements under different market conditions. I employ several jump/tail risk measures as proxies for extreme economic conditions. I focus on the jump and tail index (JTIX), which is constructed from the out-of-the-money call and put options on the S&P 500 index and is closely associated with the financial market and economic conditions. My results provide supportive evidence that the stock market responds more strongly to bad news than to good news, and this asymmetry effect becomes stronger with increasing JTIX (i.e., in periods of distress). Moreover, my results support the hypothesis that proposes the no-news-is-good-news effect.The third essay (Chapter 4) examines the information content of options prior to recommendation changes. Specifically, I construct the model-free implied variance and decompose it into upside (good) and downside (bad) variances. I show that the pre-event good (bad) implied variance contains distinctive information on stock returns, i.e., good (bad) implied variance is positively (negatively) related to post-recommendation stock returns. This relation extends over several days after recommendation changes and is robust after controlling for well-known firm characteristics and higher-order (i.e., third and fourth) implied moments. The ordered probit model shows that my findings are more consistent with the analyst tipping hypothesis.