Rothovius, Timo Markus Rikhard (2002). Empirical Evidence on Earnings and Analysts’ Earnings Forecasts. Acta Wasaensia No. 110, 171 p.
If the allocation of savings into investment opportunities is done well, economy can exploit new business ideas to spur innovation and create jobs and wealth in society. Thus it is in the interest of society as well as investors to determine the future earnings of a firm, because the underlying source of value for common stock is earnings. Clearly, earnings are important to stockholders because earnings provide the cash flows necessary for paying dividends in the future. Dividend discount models can also be framed by recasting dividends in terms of earnings and book values. In short, the efficiency of the market is directly linked to the quality of (expectations about future) earnings of the firms. Analysts play an important role as information intermediaries in assessing these expectations.
The thesis investigates earnings and earnings forecasts. The purpose is threefold. First, the information content of different earnings releases is investigated in a market (Finland) where taxation plays an important role in determining released earnings. Second, accuracy of analysts’ earnings forecasts in such a market is compared to a big market with information driven earnings releases and well-established analyzing industry (UK). Third, the dilemma in the financial literature whether analysts’ accuracy has been improving or not over the years, whether there is positive or negative bias, and whether there is over- or under-reaction, if any, to previous earnings changes, is studied, and the differences between negative and positive earnings change firms in this respect are presented.
This study is of interest to investors, analysts, corporations and legislators as well as researchers. The main findings are the following. In Finland, the Annual Account release date is important in assessing share values in regard of the earnings information. This fact should be taken into account in future studies of earnings reactions in similar markets. Next, it was found that the forecast accuracy of the analysts in UK is, or seems to be, much better than the accuracy of the analysts in Finland, but this difference can be explained by the more demanding forecasting environment in the smaller market. Finally, in the third part, the results show that problems with earnings forecasting are concentrated in negative earnings change firms, and the empirical results of whether analysts are better than time series models and whether there is negative or positive bias and over- or under-reaction, are dependent on the number of negative vs. positive earnings change firms in the research period. This is one important explanation for different results in various empirical papers in this field, and it should be accounted for in empirical research.
Timo Rothovius, Department of Accounting and Finance, Graduate School of Finance and Financial Accounting., University of Vaasa, Wolffintie 34, FIN-65101 Vaasa, Finland.
Key words: earnings, analyst, forecast, efficiency, risk