Working Paper: NBER ID: w11931
Authors: Jeremy Atack; Fred Bateman; Robert Margo
Abstract: We use establishment level data from the 1850-80 censuses of manufacturing to study the correlates of the use of steam power and the impact of steam power on labor productivity growth in nineteenth century American manufacturing. A key result is that establishment size mattered: large establishments, as measured by employment, were much more likely to use steam power than smaller establishments. Controlling for firm size, location, industry, and other establishment characteristics, steam powered establishments had higher labor productivity than establishments using hand or animal power, or water power. We also find that the impact of steam on labor productivity was increasing in establishment size. The diffusion of steam power was an important factor behind the growth of labor productivity, accounting for 22 to 41 percent of that growth between 1850 and 1880, depending on establishment size.
Keywords: steam power; labor productivity; manufacturing
JEL Codes: N61
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Larger establishments (L68) | adoption of steam power (L94) |
| adoption of steam power (L94) | higher labor productivity (J24) |
| steam power (L94) | labor productivity growth (O49) |
| steam power (L94) | productivity advantage over water power (L94) |