Trade liberalisation in developing countries: The importance of roads

Blog Firms

Research on Ethiopia shows that trade liberalisation can boost the productivity of firms in developing countries, but only if they have access to good roads.

Gains from trade liberalisation are not uniform within countries

Many developing countries have liberalised trade in the hope that greater international exposure will improve the performance of local firms.  Lower import tariffs can provide firms with access to cheaper, better quality, and a wider variety of intermediate inputs (inputs used to produce a final good) by reducing their domestic price, thereby boosting firm productivity, improving product quality, and increasing product scope.

However, gains from trade liberalisation are far from uniform within countries and may be dampened by intra-national barriers like weak domestic infrastructure.  Though there is substantial evidence that better transport infrastructure can spur firm activity, facilitate exports and increase employment, there is little research on the complementarity between better transport infrastructure and trade liberalisation effects on firm performance.

Good roads matter

In this study, we show that productivity gains for firms from a reduction in the tariff on inputs are larger with better road infrastructure locally and en route to the port.  It is not obvious that better roads should amplify productivity gains for firms from a reduction in the input tariff.  On the one hand, better road infrastructure lowers the cost for intermediaries of transporting goods from ports to destination regions within a country.  Lower intra-national transport costs lead to better transmission of tariff reductions to reductions in the price of intermediate inputs for firms located in regions with better road infrastructure.  This can lead to greater productivity gains from better access to intermediate inputs.

On the other hand, good roads can invigorate the local economy, spurring competition among intermediaries while also encouraging them to capitalise on better economic conditions to extract a higher price for their products.  This last effect of good roads can actually lead to weaker transmission of tariff reductions to reductions in the price of intermediate inputs for firms, mitigating their productivity gains from better access to intermediate inputs.

Firms, trade, and infrastructure in Ethiopia

We aim to resolve the ambiguity in the role of road infrastructure in determining productivity gains from input tariff liberalisation for firms in our empirical analysis.  We focus on Ethiopian manufacturing firms over the period 1998 through 2009.  Ethiopia is an ideal setting for our research.  Tariffs were reduced progressively starting in the early 1990s continuing into our sample period as part of a trade liberalisation agenda initiated externally. Over the same period, it also embarked on extensive improvements to road infrastructure via the Road Sector Development Programme, aimed at improving connectivity throughout the country. This provides us with rich variation in the quality of road infrastructure spatially and over time.

We use census data to calculate firm total factor productivity (the overall productivity of factors used in production).  We measure the input tariff that firms face by calculating the average tariff on the inputs they use in production, weighted by each input’s share in total inputs. To construct measures of road quality in the local region and en route to the port, we analyse government reports detailing the building of new roads and upgrading of existing ones.  We calculate the total travel distance that can be covered in one hour’s time from the town that the firm is located in and the total travel distance to Galafi, the last town in Ethiopia on the way to the port of Djibouti, which handles more than 90% of Ethiopia’s trade.  Figure shows the total travel distance (km) from towns in our sample in 1996 and the percentage change over our sample period (until 2009).  Towns with poor infrastructure in 1996 saw significant improvements over time.

Figure 1: (a) Total travel distance (km) in 1996, (b) % increase from 1996 to 2009




We examine the relationship between the input tariff and firm productivity differentially for firms in regions with access to varying quality of road infrastructure and connectivity to the port.  We find that firms located in regions with better road infrastructure and better connectivity to the port see larger increases in productivity with a reduction in the input tariff.  In fact, there is no evidence that firms located in regions with poor road infrastructure see any productivity gains from a reduction in the input tariff.  This happens for about 33% of cases in our data.  Importantly, we also find that the role of road infrastructure is stronger once we account for better economic conditions in the local region that allow for higher rents (profit) for intermediaries with market power.


Two main implications stand out from our work:

  1. Good road infrastructure (both locally and connecting regions to economic hubs) can facilitate gains for firms in developing countries from better access to intermediate inputs that comes with trade liberalisation.
  2. Greater market power among intermediaries may erode gains from better road infrastructure as trade is liberalised. More competition among intermediaries can ensure better transmission of tariff reductions to reductions in local product prices.