Capital Structure, Macroeconomic Variables, and Stock Returns: Evidence From Greece

Theoretical finance has always regarded debt as one of the principle sources of financial risk. Rational, risk-averse investors should demand a leverage premium which signals an expected positive relationship between a firm's leverage and stock returns. This paper aims to investigate the impact that the capital structure of a firm has on its stock price performance. Following Sivaprasad and Muradoglu (2007), we adopt different approaches, leverage definitions, return estimations and sets of regression analysis in order to examine the relation between leverage and stocks returns, controlling for certain macroeconomic variables. The sample used consisted of Greek listed no financial companies over the period 2003-2008. The analysis was applied at the full sample and at the industry's sub-sample level. The methodology consists of four sets of regression analysis. First, we attempt to test if leverage is value relevant to an equity investor. We estimate cumulative abnormal returns in excess of market returns over one year and examine if leverage can be used as the basis of a profitable investment strategy. The variables used here are leverage, beta, size, market-to-book ratio, price-earning ratio, interest rates and industry leverage. Second, we examine the relation between leverage and stock returns by estimating returns in excess of the risk free rate. The variables used are leverage, beta, size and market-to-book ratio. Third, we explore the relationship between portfolio leverage and expected returns by adopting the Fama and French methodology. The variables used are leverage, market risk, size and market-to-book ratio. Finally, following FF, we test if leverage is priced as a risk factor by constructing a leverage factor. We form portfolios to mimic the underlying risk factor related to leverage of firms. We follow the procedure in FF(1993). We also undertake estimations to see if returns can be explained by firm leverage even if portfolios constructed to mimic other factors related to market risk, momentum, size and market-to-book to capture variation in returns is in the time series regression.