Returns to scale in small and large U.S. manufacturing establishments

Abstract

The objective of this study is to assess the possibility of differences in the production technologies between small and large establishments in the U.S. manufacturing sector. We particularly focus on estimating returns to scale and then make inferences regarding the efficiency of small businesses relative to large businesses. We undertake this research for two reasons. First, standard industrial organization theory suggests that industrial long-run average cost curves are U-shaped or L-shaped. That is, over a certain range of output, small production units can expand their sizes to produce at declining average costs (increasing returns to scale). At a certain size, average costs flatten out (constant returns to scale). Beyond that size, average costs will increase at an increasing rate (decreasing returns to scale) as the production units continue to expand. Accordingly, this theory suggests that small production units can only exhaust economies of scale by expanding their sizes to some optimal level. This implies that small businesses are subject to inefficiency and eventually will fail, if they do not expand. Yet, a growing body of evidence indicates that