Firms’ Fundamentals, Macroeconomic Variables and Quarterly Stock Prices in the U.S

This paper modeled the effects of firms' fundamentals such as total assets and long-term debt and of macroeconomic variables such as unemployment and interest rates on quarterly stock prices of over 3000 U.S. firms in 2000-07. The merged CRSP/Compustat database was augmented by macroeconomic variables that are an important source of variation between reporting quarters and analyses were also performed for ten Global Industry Classification Standards (GICS) groups. The comprehensive dynamic models for quarterly stock prices were estimated by maximum likelihood taking into account the heterogeneity across firms. The main findings were that the estimated coefficients of lagged stock prices in simple dynamic random effects models were in the interval 0.90-0.95 though the unit root null hypotheses were rejected at the aggregate and GICS group levels. Second, the comprehensive dynamic models for stock prices showed firms' earnings per share, total assets, long-term debt, dividends per share, and unemployment and interest rates were significant predictors. Moreover, there were significant interactions between firms' long-term debt and interest rates underscoring the costs of borrowing. Finally, dynamic models were estimated for firms in ten GICS groups and results showed the importance of employing sector-specific macroeconomic variables for predicting stock prices