Our replication of a study estimating production function of Russian manufacturing firms in 2012-2018 by Kaukin and Zhemkova (2023) revealed unrealistic capital elasticities.
| industry |
labour |
capital |
| 01. Food, beverages, and tobacco |
0.78 |
0.09 |
| 02. Textiles and textile products |
0.83 |
0.03 |
| 03. Wood and products of wood and cork |
0.74 |
0.05 |
| 04. Pulp, paper, printing and publishing |
0.82 |
0.07 |
| 05. Coke, refined petroleum |
0.73 |
-0.04 |
| 06. Chemicals and chemical products |
0.72 |
0.02 |
| 07. Pharmaceutical products |
0.82 |
0.20 |
| 08. Rubber and plastics |
0.79 |
0.09 |
| 09. Other non-metallic mineral |
0.69 |
0.05 |
| 10. Basic metals |
0.63 |
0.06 |
| 11. Fabricated metals |
0.75 |
0.05 |
| 12. Electrical and optical equipment |
0.83 |
0.12 |
| 13. Electrical equipment |
0.78 |
0.01 |
| 14. Machinery, not elsewhere classified |
0.74 |
0.04 |
| 15. Transport equipment |
0.69 |
0.16 |
| 16. Furniture |
0.84 |
0.07 |
| 17. Manufacturing, not elsewhere classified |
0.81 |
-0.01 |
| 18. Equipment repairs and installation |
0.85 |
0.00 |
We provide 5 explanations for the apparent failure to replicate KZ exactly:
- Data differences between Orbis and the RFSD
- Differences in sample restrictions
- Differences in deflators
- Value added definition
- Proxy variable
Thank you for any contributions.
Our replication of a study estimating production function of Russian manufacturing firms in 2012-2018 by Kaukin and Zhemkova (2023) revealed unrealistic capital elasticities.
We provide 5 explanations for the apparent failure to replicate KZ exactly:
Thank you for any contributions.