Properties of Individual Analysts’ Earnings Forecasts for the Turkish Stock Market

Abstract

We examine the properties of individual analysts' forecasts of annual earnings for Turkish companies. Using a sample obtained from the Institutional Brokers Estimate System (I/B/E/S) database of earnings forecasts, we measure individual analysts' forecast errors and relative optimism, and test for serial correlation in these measures. We also compare the forecast accuracy of analysts working for Turkish brokerage firms to the forecast accuracy of those employed by non-Turkish firms. An important contribution of this paper is that it is the first study to focus on the earnings forecasts of Turkish companies at the individual analyst level.The past decade has seen substantial growth in emerging markets as investors seeking high returns and potential diversification benefits have channeled their funds into assets in these markets. Turkey, which is identified as a big emerging market by the U.S. Department of Treasury (see Business America, 1994), has enjoyed substantial growth in its equity market since the inception of the Istanbul Stock Exchange (ISE) in 1986. Sustained growth in the Turkish equity market and its integration with the world capital markets require the availability of timely and accurate information on company performance. Security analysts' forecasts of company earnings are an integral part of the information set for equity investors. The role of analysts' earnings forecasts as a representative of the market's earnings expectations is well documented in the literature (see Brown, 1978;Fried and Givoly, 1982;Rendleman et al., 1982;and Brown et al., 1987). Also, analyst following is a good indicator of interest in and potential growth of an equity market. Figure 1 shows the number of Turkish companies for which analysts reported earnings forecasts to the Institutional Brokers Estimate System (I/B/E/S) for each year during the 1994-2001 period. As is apparent in Figure 1, there was a significant increase in the number of Turkish companies followed by analysts, which can be interpreted as a sign of increased interest in the Turkish equity market.There is a rich body of research that focuses on the characteristics of individual analysts' earnings forecasts. Tamura (2002) studies the properties of individual analyst forecasts and finds that an analyst's forecast tends to be influenced by her personality. Cooper et al., (2001) rank individual analyst forecasts and conclude that lead analyst forecasts tend to impact the markets more than the follower analyst forecasts do. On the international analyst forecast side, Firth and Gift (1999) compare the precision of consensus analyst forecasts across countries and find that estimates are more precise for eight countries compared to the U.S. consensus estimates. Allen et al., (1997) find differences in forecast errors between Pacific Basin markets and more developed capital markets. Ang and Ma (2001) document that analysts failed to anticipate the Asian financial crisis. De Bondt and Forbes (1999) find evidence for herding behavior among U.K. analysts....