Essays on analyst behavior

This dissertation is comprised of two essays on analyst behavior. In the first essay, I examine changes in analysts' earnings forecasts after Regulation FD and the Global Analyst Research Settlement both in the United States and 40 other countries. I find that analysts around the world generally make overly optimistic earnings forecasts. However, in the U.S., forecast bias declined after Regulation FD and then further after the Global Analyst Research Settlement. Outside the U.S, forecast bias declined only after the Global Settlement, but was not affected by Regulation FD. Forecast bias declined more in countries with civil law, less protection of minority shareholders, and less transparent accounting disclosures. Furthermore, the presence of the twelve banks involved in the Global Settlement positively correlates with the reduction in forecast bias abroad.;In the second essay, I study long-term earnings growth (LTG) forecasts around seasoned equity offerings. I document that LTG forecasts are generally higher for firms with fewer analysts than for firms with more analysts covering the stock. In contrast to prior studies, I find no difference between LTG forecasts made by affiliated analysts who work for investment banks that have relationships with equity issuing firms and LTG forecasts made by unaffiliated analysts. Further analysis shows that aggregation bias influences the prior studies' conclusion that affiliated analysts make more optimistic forecasts. Both affiliated and unaffiliated analyst forecasts are more optimistic for firms with fewer analysts. In addition, the number of unaffiliated analysts relative to the number of affiliated analysts covering the same firm increases with the firm's analyst coverage. As a result, the mean forecast bias of unaffiliated analysts turns out to be lower than the mean forecast bias of affiliated analysts. However, this observed difference in the mean forecast bias is driven by analyst coverage rather than analyst affiliation.