“Product Substitutability and Productivity Dispersion”, 2004-05-01 (; backlinks; similar):
Tremendous differences in producer productivity levels exist, even within narrowly defined industries.
This paper explores the influence of product substitutability in an industry on this disparity. When consumers can easily switch between producers, inefficient (high-cost) producers cannot operate profitably. Thus high-substitutability industries should exhibit less productivity dispersion and have higher average productivity levels.
I demonstrate this mechanism in a simple industry equilibrium model and test it empirically using producer-level data from 443 US manufacturing industries.
I find evidence that substitutability—measured in several ways—is indeed negatively related to within-industry productivity dispersion and positively related to median productivity.
…Perhaps surprisingly, a great amount of productivity variation between plants is observed within what may seem to be narrowly defined (for example, four-digit SIC) industries. Table 1 shows statistics demonstrating this dispersion. Using plant-level data from the 1977 Census of Manufactures, I compute productivity distribution moments for 4-digit manufacturing industries for each of 4 different productivity measures. As can be seen in the first numerical column, the average within-industry interquartile range of logged plant-level labor productivity values is roughly 0.66. This corresponds to a nearly 2-to-1 ratio in value added per labor unit (employee or employee-hour) between the 75th-percentile and 25th-percentile plants in an industry’s productivity distribution. Bear in mind that these differences are observed when restricting attention to the middle half of the distribution; including more of the tails amplifies intra-industry heterogeneity. The average 90–10 and 95–5 percentile productivity ratios within industries are over 4 to 1 and 7 to 1, respectively. Factor intensity variations are not solely responsible for these large differences, either. Intra-industry total factor productivity differences, though smaller, are still sizable. The values in the bottom half of Table 1 indicate average interquartile total factor productivity (TFP) ratios between 1.34 to 1 and 1.56 to 1, depending on the measure. It is important to note that the heterogeneity observed here is a persistent phenomenon. Empirical studies using other (but perhaps less comprehensive) cross sections have found similar within-industry productivity differences.
See Also:
“Why do Some Countries Produce So Much More Output Per Worker than Others?”
“Beyond Computation: Information Technology, Organizational Transformation and Business Performance”
“Computer and Dynamo: The Modern Productivity Paradox In A Not-Too Distant Mirror”
“The Productivity J-Curve: How Intangibles Complement General Purpose Technologies”
“Ten Facts on Declining Business Dynamism and Lessons from Endogenous Growth Theory”
“People and process, suits and innovators: the role of individuals in firm performance”