Economic transformation occurs when resources are shifted from low-productivity to high-productivity activities. This happens, for example, when resources are shifted from low-productivity firms to high-productivity firms within a sector. We use productivity measures at the firm level to measure this process.
Generally the scope for such shifts is greater in developing countries than in developed countries, because there is less pressure and competition (including because of increased trade protection). In developing countries and hence fewer penalties for being less productive.
We can examine total factor productivity at firm level using WB enterprise survey (WBES) data. The files include two charts per country using:
- Dispersion in productivity across firms by sector. This shows two types of density distribution using histogram (bar) and kernel (line) charts. The higher the peak around zero, the more firms are centred around average productivity. The lower the peak and the wider the distribution, the more dispersion there is between firms in a sector, which indicates opportunities for improving productivity within a sector.
- Distribution in productivity by country – comparing kernel and normal distributions. For many countries, the kernel density peaks higher and before the normal distribution. This means that a large number of firms have a productivity that is just below the average level productivity. With a long tail to the right meaning very few firms have very high levels of productivity (although this could also pick up sector effects).
There are a number of pros and cons in using firm-level data to analyse economic transformation. Whilst the WBES data cover a wide range of countries, the number of firms covered in each country is often low and high selective.
|Area of transformation||Data sources||Pros||Cons||Uses|
|Productivity at firm-level||World Bank enterprise survey data||Number of countries, comparability across countries||Number of firms covered in each country can be low||Total factor productivity level, dispersion across sectors|
|Wages by occupation||ILO – OWW database||Historical data on wages||Data availability in recent years; measurement issues and variability across time|