A DISSERTATION SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR AWARD OF Msc COMMERCE-FINANCE AND ACCOUNTING IN THE SCHOOL OF GRADUATE STUDIES AND RESEARCH AT KCA UNIVERSITY AUGUST, 2016The Efficient market hypothesis EMH concept states that a market is efficient if security prices immediately and fully adjust to reflect all available information. One of the information that affects share price reaction is earning announcement. Thus the study analysed the effect of semi-annual earnings announcement on share price responsiveness of financial firms listed at NSE. This was specifically analysed through the following specific objectives: To determine the effect of pre-earning announcement on the share price performance of financial firms listed at NSE; to determine the effect of post earnings announcement on share price performance on financial services firms, to determine the effect of earning announcement date on financial services firms listed at NSE. The study adopted an event study methodology. The study target population comprised of 17 companies in both insurance and banking industry. Census sampling was used to select all the companies in insurance and banking industry. Data was collected through secondary sources from Capital Markets Authority and Nairobi Securities Exchange. Data was analysed by STATA through the use of t-tests with the results presented through graphs and tables. The study utilized a 15 day event window, i.e. 7 days before the event date and 7 days after the event date. The event date itself was represented by 0. The researcher used event study methodology to test the responsiveness of prices to earnings information releases for a sample of 13 companies in the 20-share index. Results indicate that there were significant abnormal average returns in day -7, and -4. The average abnormal returns in day -7 and -4 were negative (-0.64837 and -0.99533) whereas there were positive average abnormal returns after event day except for day 6. This indicates existence of average abnormal negative returns before the event day and positive average abnormal returns after the event day. However, results regarding CAARs indicated that there were no significant CAARs in the entire event window thus indicating that abnormal negative returns were cancelled by the abnormal positive returns. From the results it is recommended that companies should be compelled to release timely and accurate information to enable investors to make accurate decisions in both annual and semi-annual announcements.