Measuring Total Factor Productivity Using the Enterprise Surveys : A Methodological Note
Total factor productivity is a key element of economic growth and an important performance metric for policy makers. This note describes the methodology for measuring firm-level total factor productivity using the World Bank's Enterprise Surve...
| Main Authors: | , , , |
|---|---|
| Language: | English |
| Published: |
World Bank, Washington, DC
2020
|
| Subjects: | |
| Online Access: | http://documents.worldbank.org/curated/en/306091607457083831/Measuring-Total-Factor-Productivity-Using-the-Enterprise-Surveys-A-Methodological-Note http://hdl.handle.net/10986/34913 |
| Summary: | Total factor productivity is a key
element of economic growth and an important performance
metric for policy makers. This note describes the
methodology for measuring firm-level total factor
productivity using the World Bank's Enterprise Surveys
cross-country data. It also presents some estimates
recovered from the production function. Two versions of the
production function are estimated: one Cobb-Douglas, the
other a more flexible translog specification. Both
estimations are at the two-digit industry level pooling all
the Enterprise Surveys data across economies. Evidence is
found against using a Cobb-Douglas specification, which is
more parsimonious, and in favor of using the flexible
translog specification. The resulting firm-level estimates
are all published in the Enterprise Surveys database with a
unique firm identifier to link to the rest of the Enterprise
Surveys data; because the estimates are reliant on new data,
they are updated periodically as new Enterprise Surveys data
become available. The results show that: (i) median firms
operate close to constant returns to scale; (ii)
gross-output and value-added production functions provide
similar ranking of sectors in terms of output elasticities,
capital intensity, and returns to scale; (iii) there is
large, firm-level heterogeneity in output elasticities; and
(iv) gross-output-based total factor productivity measures
are less dispersed than the value-added ones. |
|---|