Measuring returns to scale in nineteenth-century French industry

Were nineteenth-century French firms too small? Previous research has found either constant returns to scale throughout (Nye, 1987) or increasing returns to scale for most industries (Sicsic, 1994). This research sheds some light on these contradictory results by proposing and estimating a series of models that consider previously ignored econometric issues. My analysis of the 1839-1847 census suggests that there were increasing returns to scale for some industries and constant returns for others, while the analysis of the 1861-1865 census suggests constant returns to scale throughout. Assessing the economic implications of the estimated returns to scale, I argue that there were significant unexploited scale economies in the first half of the nineteenth century. In the second half of the nineteenth century, though, France had little to gain from scale economies. (C) 2004 Elsevier Inc. All rights reserved.