Working Paper: CEPR ID: DP6271
Authors: Berthold Herrendorf; Akos Valentinyi
Abstract: Many applications in economics use multi-sector versions of the growth model with Cobb--Douglas production functions at the sector level. In this paper, we measure the U.S. income shares of capital and labour for five sectors that encompass the typical sectors used in the literature. We also split the capital shares of these five sectors into the sector income shares of land and of structures and equipment. We find that the factor income shares differ widely across sectors. For example the capital share in agriculture is about twice that in construction. Moreover, the land shares in agriculture and in services are sizeable whereas the land shares in all other sectors are small. Our findings suggest that the general practice of using the economy-wide factor income shares also at the sector level is not a good practice.
Keywords: industry-by-commodity; total requirement matrix; input-output tables; sector factor shares
JEL Codes: O41; O47
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| sector type (agriculture vs. construction) (L79) | capital share (D33) |
| sector classification (land-intensive vs. labor-intensive) (R30) | capital and land shares (D33) |
| economy-wide capital shares (D33) | inaccurate estimations of sector-level shares (E25) |