This paper is motivated by the gap between the widespread usage of the Price- Earnings multiple in valuation practice and the lack of research related to the key assumptions using the Price-Earnings multiple. The purpose of this paper is to examine how the valuation estimate of the Price- Earnings multiple can be maximized, by adjusting the chosen earnings figures and how comparable firms are selected. To examine this, we use the American S&P1500 index over a ten year period (2002-2011) as the basis of our study. We examine if using earnings figures adjusted for extraordinary items result in a better valuation estimate than earnings figures including extraordinary items, if historical figures gives a better valuation estimate than forecasted figures, if comparable firms should be chosen on the basis of similar value drivers or on the basis of industry affiliation (and the finesse of the industry classification). Furthermore we examine whether additional precision in the valuation estimate can be obtained through a combination of alternative methods of selecting comparable companies. Through our empirical analysis of S&P1500 over the ten year period, we answer the above stated questions. We conclude that earnings figures adjusted for extraordinary items results in lower error compared to non-adjusted figures. We conclude that forecasted earnings figures result in more precise valuation estimates than historical figures. Especially the two and three year forecasted figures result in significantly more precise valuation estimates compared to reported figures. Furthermore we conclude that selecting comparable firms through an industry classification system provides practitioners with the best method of selecting comparable firms, and that comparable firms should be chosen from the narrowest industry definition. In conclusion we find that no further precision in the valuation estimate can be obtained through a combination of alternative methods of selecting comparable companies.
|Educations||MSc in Finance and Accounting, (Graduate Programme) Final Thesis|
|Number of pages||127|