Does freer trade really lead to productivity growth? Evidence from Sub-Saharan Africa

Manufacturing is intensive in the use of reproducible factors and exhibits greater technological dynamism than primary production. As such its growth is central to long-run development in low-income countries. Sub-Saharan African countries are latecomers to industrialization, and barriers to manufacturing growth, including those that limit trade, have been slow to come down. What factors contribute most to increases in output and productivity growth in their manufacturing sectors? Recent trade-IO theory suggests that trade liberalization should raise average total factor productivity (TFP) among manufacturing firms (Melitz 2003), but these predictions are conditional on maintained assumptions about the nature of industries, factor markets and trade patterns that may be less suitable in a developing-country setting. Manufacturing industries and firms are heterogeneous, so this analysis demands disaggregated data. We use firm-level data from the World Bank’s Regional Program on Enterprise Development (RPED) covering Ghana, Kenya, Nigeria, and Tanzania, 1991-2003. Among other things, the data distinguish exports by destination (Africa and the rest of the world), which is important due to the spread of intra-Africa regional trade agreements (RTAs). Econometric results confirm well-known relationships, for example a positive association between export intensity and TFP. However, we also find the destination of exports to be important. Export firms are more productive but have experienced declining TFP growth, and this has occurred at different rates depending on the country and the export market addressed. We show that these differentials are consistent with predictions from a modified statement of the Melitz model. The TFP results add a new dimension to controversies over the development implications of trade liberalization and the promotion of intra-Africa RTAs.