Predicting Accuracy of Valuation Multiples Using Value Drivers: Evidence from Indian Listed Firms

Abstract

The objective of this research study is twofold: 1) to evaluate the prediction accuracy of four valuation multiples across three sectors for Indian listed firms and 2) to identify the fundamental drivers for these multiples. The valuation multiples identified for this study are: price to earnings (P/E), price to book value (P/BV), price to sales (P/S) and enterprise value to earnings before interest, depreciation, tax and amortization (EV/EBIDTA) and the sectors taken are steel, banking and automobile. Multiple regression methodology is followed with the valuation multiple as dependent variable and the value drivers as independent variables, to get predicted multiples on 470 firm observations. By regressing the multiples on fundamental variables, the best suited multiple for each sector and the key drivers of the multiple are obtained. The empirical findings based on root mean square error (RMSE) and Theil coefficient reveal that least prediction errors are observed in P/S and EV/EBIDTA for the automobile sector, EV/EBIDTA for the steel sector and P/BV for the banking sector. It is also observed that the significant variables that explain these multiples are beta, return on equity (ROE), return on capital employed (ROC), dividend payout ratio (D/P) and net profit margins (NPM). These findings are in line with the derivation of fundamental drivers for each multiple as explained in Gordon model. Damodaran: 2007 [1] . The present work contributes to emerging market literature on equity valuations and attempts to compare valuations based on market approach using value drivers. A comparison of forecasts with actuals helps in recommendations to buy/sell/accumulate/hold for equity investors and is also pertinent for market participants and financial regulators.