The correlation of financial leverage and other financial measures to systematic risk and to stock return with consideration of different industry classifications: The case of frequently-traded non-financial firms in the Philippine Stock Exchange

• Main Authors: Salamanca, Patricia Anne A., Tan, Nicolas Michael T., Tan, Stephanie Grace T.
• Format: text
• Language: English
• Published: Animo Repository 2015
• Subjects: Financial leverage > Philippines; Financial risk > Philippines
• Online Access: https://animorepository.dlsu.edu.ph/etd_bachelors/7694
• Institution: De La Salle University

After running an unbalanced pooled cross-sectional data regression of a sample of 30 Jordanian industrial firms listed in the Amman Stock Exchange (ASE) in the years 2001-2011, and beta coefficient as a proxy for systematic risk, it was guaranteed that the four different definitions of financial leverage are definite factors in estimating systematic risk (Ramadan, 2012). On the contrary, financial intermediaries who have a higher level of financial leverage made them better-off as it minimizes their systematic risks (DeAngelo &Stulz, 2013). However, certain research gaps have been found: (1) there has been no similar study made on the Philippine setting (2) Industrial firms may not necessarily define other industry classifications (3) Lastly, the risk and reward principle may not always hold in answering the concerns of financial investors. To close these gaps, this research paper aims (1) to verify whether financial leverage, to a definite extent, always means risk for all actively-trading non-financial firms in the Philippine Stock Exchange (2) to evaluate the differences in beta appetite among different non-financial industry classifications (3) and as an extension, to know if risk-reward trade-off should always be expected between financial leverage and stock return. The data used is a panel dataset composed of annual beta, annual stock returns and annual firm-specific data from years 2007 to 2013. Daily PSE indices and daily firm-specific stock prices were used to compute for the annual betas using the market model regression anchored in the capital asset pricing model (CAPM). Only non-financial firms with at least 182 trading days in one year were included in the sample in order to prevent lag effects. To facilitate this research, five different models were considered -- Generalized Least Squares (GLS), Driscoll-Kraay Method, Ordinary Least Squares (OLS), Fixed Effects Model (FEM) or Random Effects Model (REM). If heteroscedasticity is present, GLS is used but if serial correlation is present, then Driscoll-Kraay Method is used. If both are not present, then the model selection process using Walds Test, Breusch-Pagan Test, and Hausman Test is used. The Least Square Dummy Variable - Model 3 (LSDV 3) emerged as the best model in the overall regression for all measures of financial leverage, as well as stock return, because it was able to fix the effects for both time and space variations. For financial leverage, most of the dummy variables generated were statistically significant, implying that the panel data used was highly heterogeneous. From here, Ramadan (2012) and Al-Qaisi's (2011) claims were refuted in that with respect to actively-trading non-financial firms in the PSE, not all measures of financial leverage are significantly correlated with systematic risk regardless of industry classification. However, the statistically significant dummy variables paved the way for a unique per sector analysis. The opposite can be said for stock return, as the panel data was highly homogenous, and the insurmountable variation in the data could not significantly explain the contribution of stock return to systematic risk. The per sector analysis of the four measures of financial leverage, as well as the secondary independent variables proved to be value-adding in relation to systematic risk. The results mirrored existing policies and concepts, and broke down barriers between risk and reward, giving reason for investors and firms not to settle for information asymmetry, but to be knowledgeable about industry specific business models, so that policymakers will be keen to avoid implementing blanket policies. Stock return and the secondary independent variables did not make much of an impact to systematic risk, cementing the validity of the random walk theory. The combination of complete data and astute analysts can never outperform the market.