Media Briefings

Productivity Differences Between Rich And Poor Countries: New Evidence From Trade Data

  • Published Date: September 2011

Differences in productivity between rich and poor countries are systematically larger in sectors that require skilled labour and extensive research and development (R&D) – so sectors such as Scientific Instruments rather than Apparel or Textiles.

That is one of the findings of research by Harald Fadinger and Pablo Fleiss, which uses trade data to measure productivity across 24 manufacturing sectors in over 60 countries in both the developed and developing world.

Their study, published in the September 2011 issue of the Economic Journal, also finds that economies with a larger pool of educated workers adopt new technologies faster – and this effect is stronger in human capital-intensive industries. Having a more developed financial system also leads to higher productivity levels and growth rates within sectors.

Policy-makers in developing countries are often concerned that their industries are inefficient and not competitive in world markets. Thus, they would like to know in which sectors their countries’ technologies are particularly lagging behind the ones used in industrialised economies. Moreover, they would like to know which factors make some of their industries particularly inefficient.

A particular challenge for measuring industry-level differences in economic efficiency – so-called ‘total factor productivity’ (TFP) – between rich and poor countries is that standard economic methods require comparable information on output and inputs at the sector level, as well as output price data.

But there are very large gaps in these data for virtually all developing countries because data collection standards in most of those countries are very poor. As a result, very little is known about the magnitudes and patterns of industry-level efficiency differences outside the industrialised world.

This new study introduces and applies a new methodology for estimating TFP at the industry level, which relies on information contained in bilateral trade. This makes it possible to provide a comparable set of productivities for 24 manufacturing sectors in more than 60 countries at all stages of development.

The idea behind the method is to exploit the information contained in bilateral sectoral export values, appropriately adjusted for differences in production volume and input costs, across export markets. These data are readily available for most countries because importers, which are mostly industrialised economies, collect them.

As an example, consider how to infer Ghana's TFP relative to the United States in the textile sector. The researchers first measure the fraction of Ghana's textile production relative to US textile production that is exported to each market. If Ghana exports more to an average market than the United States, adjusting for relative input costs and relative bilateral trade costs, this indicates a higher level of productivity in Ghana’s textile sector.

Having estimated TFPs at the industry level, the researchers show that TFP differences between rich and poor countries are systematically larger in sectors that are intensive in the use of skilled labour and research and development. Specifically, productivity gaps are far more pronounced in sectors such as Scientific Instruments, Electrical and Non-electrical Machinery than in sectors such as Apparel, Textiles or Furniture.

Moreover, the study finds that cross-country TFP differences in manufacturing sectors average about the same substantial orders of magnitude as has already been found by other studies at the aggregate economy level.

The researchers use their estimates to test various development theories that make predictions on industry-level differences in efficiency between rich and poor countries.

The results show that technology spillovers are important in explaining cross-country sectoral TFP differences: countries that use more backward technologies experience larger efficiency gains by upgrading to more advanced ones than countries that are already using rather sophisticated technologies.

The study also finds that economies with a larger pool of educated workers adopt new technologies faster, and this effect is stronger in human capital-intensive industries.

Finally, it shows that having a more developed financial system affects productivity levels and growth rates by leading to a more efficient allocation of credit within a given sector.

ENDS

Notes for editors: ‘Trade and Sectoral Productivity’ by Harald Fadinger and Pablo Fleiss is published in the September 2011 issue of the Economic Journal.

Harald Fadinger is at the University of Vienna. Pablo Fleiss is at the International Labour Organisation.

For further information: contact Romesh Vaitilingam on +44-7768-661095 (email: romesh@vaitilingam.com); Harald Fadinger vai email: harald.fadinger@univie.ac.at; or Pablo Fleiss via email: fleiss@ilo.org