You are here

On the forecasting power of corporate sales growth determinants

Publication Year 
This article presents a panel regression model to evaluate the forecasting power of the determinants of corporate sales growth cited in the literature. The model is estimated using information on 189 thousand unique firms over 2008-2021. The results point for a negative relation with size, age (firm, employees, and managers), employees and managers’ gender (female) and productivity, and a positive relation with access to external funding, profitability, belonging to an economic group, the shareholder being simultaneously a worker, employees’ education, lagged investment and financing flows, as well as changes in the external environment (industry, local and macroeconomic). The relation with leverage is concave and depends on debt composition. The effect of autocorrelation depends on the activity sector and it is typically positive for larger firms. Our specification outperforms a model where the growth rate of sales is the same for all firms. The variables that contribute the most to the model performance are those related with the external environment, in particular the growth rate of domestic demand and exports. Except for investment-related variables, most other firm-level variables cited in the literature have a negligible forecasting power.
Document link