Mis-Valuation of cyclical companies

Aim of this master thesis is to understand if business cyclicality decreases the accuracy in corporate valuation. To assess the existence of any kind of potential misvaluation, a multiple approach analysis is performed for two samples of companies: cyclical and non-cyclical companies. The following multiples: current EV/EBIT; current EV/EBITDA; current and forward P/E have been computed for a sample of 174 companies listed on the London Stock Exchange during the interval 1995 - 2007. In the first place I have compared the estimated value with the observed value on the market for both samples, the difference is called Valuation Error (VE) and could be considered a proxy for valuation accuracy. If VE is significant, one can conclude there is a potential misvaluation. I found VE to be higher for companies categorized as cyclical. As a second step, I performed a OLS analysis in order to understand which are the potential variables affecting valuation errors. Finally, a Logit analysis confirms the results from the OLS model. The dummy variable indicates cyclical companies have a positive effect on the magnitude of VE. Volatility of companies‘ earnings increases VE, while the larger the company being valued, the smaller is the VE (more accurate valuations).