Growth in manufactured exports across African states

The first part of this work has recently been completed by Paul Collier and Tony Venables in ‘Rethinking trade preferences: How Africa can diversify its exports’ (forthcoming, World Economy). They suggest that European and American trade policy reform may be a good place to start. They argue that developed-country trade preferences for Africa could act as a catalyst for manufacturing exports, leading to rapid growth in exports and employment. To do so, preferences need to be designed to be consistent with international trade in fragmented ‘tasks’ (as opposed to complete products) and need to be open to countries with sufficient levels of complementary inputs such as skills and infrastructure. Collier and Venables draw on a rich dataset of all global, bilateral trade flows compiled by the UN to estimate the impact of the U.S. African Growth and Opportunities Act (AGOA) and European Everything but Arms (EBA) initiatives on African garment exports.   They find that AGOA raised Africa garment exports to the U.S. by a factor of seven, while the impact of EBA was insignificant in their main specification. They attribute the differential impact to rules of origin in EBA and its narrower focus on least developed countries, suggesting possible reforms for European policy. 

In a next phase, this work will be taken forward by exploring the correlates of export success, beyond the impact of European and US trade policy. We are currently compiling trade and industrial production data on a few key sectors, such as garments, to see how the location of exporting has shifted over the past two decades, and in particular, how particular sub-categories have become clustered in a few locations. We are attempting to get evidence on the more recent phenomenon of clustering according to ‘tasks’ rather than products.

As Collier and Venables caution, international trade preferences are not a substitute for domestic policy reform.  What, then, are the domestic institutional arrangements that have allowed some countries but not others to capitalize on preferential market access?  The first stage tests the hypothesis that the five aspects of the business environment that are likely to be affected by institutions are important in determining export success. The dependent variable is some measure of success of manufactured exports such as the value of these exports in year t as a share of GDP in some baseline year. Proxies are built for the five functions that institutions might potentially affect. The key prices that are sometimes subject to regulation are exchange rates, wages of unskilled labour, and interest rates. The severity of predation of firms might include measures of government corruption, crime, and union power. The risk of default on contracts may be proxied through firm opinion surveys. The provision of public services pertinent for manufactured exports might include the ports, water and electricity. Firm network services might be proxied by exports at various levels of aggregation. In addition to these proxies for functions that are affected by institutions, a range of control variables would be included. The results might indicate the extent to which variation in these functions, from whatever causes, is associated with variation in export performance.

The second stage begins testing the hypothesis that variation in these five aspects of the environment is affected by variation in institutions. The detail of how this work is designed depends upon the results of the first stage. Ideally, each aspect of the business environment that is important for export performance is related to proxies for the institutions which might affect it.

The third stage explains these institutions in terms of some more fundamental aspects of the society, such as its history, the relative size of interest groups or other institutional determinants. The fourth stage then uses the third stage to identify the causal relationships investigated in the first two stages. This analysis also nests a test of the empirical relevance of the underlying narrative related to the differences in opportunities across states with different underlying geography for export performance.