While a number of studies have demonstrated the importance of factor utilization in economic analysis, the impact of operating hours and/or hours of work in the production function remains largely unknown, particularly in terms of the capital operating time. Using French data on industrial firms for the period 1989 to 2001, we estimate a Cobb-Douglas production function that considers the stocks and both the capital operating time and the working time. We draw on the framework defined by Blundell and Bond (2000), assuming that serially correlated shocks allow a dynamic representation of the production function, and we use the system generalized method of moments for estimation. Splitting capital operating time into shiftwork patterns and working time, our results show that shiftwork patterns matter for the estimation of a production function while working time is less informative. Specifically, our estimates yield identical output elasticities for shiftwork and capital: thus, doubling the shifts is equivalent to doubling the stock of capital. In addition, we cannot reject the hypothesis of constant returns to scale and the Cobb-Douglas specification is accepted when taking into account the capital operating time and/or the working time. Otherwise, a Translog production function is more appropriate.
