«Policy Research Paper Industrial Policies for the Structural Transformation of African Economies: Options and Best Practices No. 2 Policy Research ...»
Considering the period 2000-2006, in 8 of 35 African countries5, manufacturing exports represented 10 per cent or more of GDP. Only Botswana and Swaziland have reached manufacturing export to GDP ratios which are equal to or higher than values recorded for the East Asia and Pacific region which are around 30 per cent. Moreover, manufacturing export is highly concentrated in few countries (South Africa, Algeria, Libya, Tunisia, Morocco and Egypt). Sub Saharan Africa is the region of the developing world with the highest dependence on primary product exports, especially oil.
Africa’s world market share in light manufacturing has been low and declining (0.8 per cent in 2006 against 0.9 per cent in 2000. Most African countries do not export in any significant way, nor produce simple, locally consumed products, which do not require major investments or skills to produce. Table 2 compares the shares of merchandise exports and imports as a percentage of the world total for African countries, with respect to other regions of the world. It shows an increase in Africa’s export growth during the period 1995-2005, mainly driven by the increase in the export of primary products, especially oil. However, both North African and SSA economies are very marginal to world trade in manufacturing products as shown in Table 3 It is nonetheless worth noting the new dynamism on export of manufactured goods recording an average increase of 9.4% between 1995 and 2005.
Figure 3 offers some additional criteria to evaluate the situation of the industrial sector in Africa.
Primary products now account for nearly three quarters of total exports to both developed and developing countries. Resource-based manufactured goods account for 12–15 per cent while other manufactures (low, medium and high technology) represent a small part of exports to both groups (about 10 per cent of exports to developing countries and 14 per cent of exports to developed countries).6 It is interesting to note that low, medium and high technology manufactures only represented 3.1 per cent of exports to China and 3.9 per cent of exports to India in 2008.
Figure 3: Composition of Africa’s exports to main export partners, 2008 Source: UNCTAD (2010) In the last 15 years the difference in the export vector directed to other developing and developed countries has drastically reduced since the composition of Africa exports to other developing countries has shifted towards primary products (see Figure 4). On the contrary, intra-African trade is characterized by a more balanced composition (see Figure 3). At the same time, Africa is increasingly importing manufactures from non-African developing countries while the share of primary products is decreasing. These broad patterns resemble that of imports from developed countries.
Figure 4: Structure of Africa’s exports to non-African developing countries, 1995–2008 Source: UNCTAD (2010) Table 4 which compares manufacturing export performance of African countries with respect to other developing countries, in terms of volume and price values, shows that over the period 1995–2006, manufacturing export values in Africa increased by 12 per cent, more than the world average as well as all other developing countries. This result is however not homogenous among African countries. The largest increases occurred in post-conflict and oil-exporting countries such as Chad, Equatorial Guinea, Mozambique and Sierra Leone. A number of African countries saw very little growth in export values over the same period - mainly countries that experienced political unrest in the period, such as the Central African Republic, Eritrea and Liberia.
Table 4: Average yearly increases in merchandise export values, volumes and prices, 1995–2006 (Percentage)
Source: UNCTAD calculations based on UNCTAD 2008a.
The analysis of the change in export volumes and prices gives some interesting indication on the characteristics of African manufacturing with respect to other regions. Export volumes increased between 1995 and 20067, but less than both the world average and the developing countries average. Interestingly, this increase is much lower than the increase in the value of exports.
Also, this increase in the export unit price is over four times higher than the world average and nearly three times higher than the developing-country average.8 This indicates that much of the increase in export values in Africa was due to rising prices rather than to increased export volumes. The fact that prices are largely out of the control of African countries suggest that the limited export volume growth indicates weak export dynamics and creates serious doubts about the long-term performance of African manufacturing export.
The very low level of diversification in export products is another weakness of the African economies. Indeed, the export concentration index for Africa has remarkably increased in the last 15 years. The current level of export concentration is much higher in Africa than in other regions in the world whether developed or developing. In comparison with both Latin America and East Asia, the export concentration index in Africa is more than double (Figure 5).
Reasons for the increases in export volumes are country dependent. For a detailed analysis of country’s cases see UNCTAD (2008).
8 As expected, oil-exporting countries recorded the largest increases while non-oil exporting countries suffered reductions in the unit price of their exports.
Figure 5: Export concentration index, 2006
Source: UNCTAD calculations based on UNCTAD 2008a.
Hammouda et al. (2006) show that, while African economies have always exhibited very low levels of diversification, in recent years, the concentration of production has even increased in some African countries. This is especially the case of oil-exporting countries of the Central and Western African region. Figure 6 compares the diversification level of five sub-regions using the Normalized Hirschman Index.9 In 1980, the most diversified sub-regions were COMESA and ECOWAS. The least diversified was CEMAC with SADC and North Africa in the middle. By 2002, the diversification gains at the sub-regional level had changed. SADC was the most diversified region followed by COMESA and North Africa while CEMAC has remained the least diversified subregion.10
Figure 6: Comparison of African sub-regions diversification levels:
Normalized Hirschman Index Source: Hammouda et al. (2006).
The causes for the low level of African manufacturing production and export are numerous but three seem to be the most important. First, Africa lacks the technological capabilities needed to set a successful process of industrialization in motion. Second, the continent lacks the financial resources necessary to finance the development of a manufacturing sector. Moreover, the political instability that characterizes most African countries adds to these difficulties additional costs that further reduce the incentive to invest in manufacturing. Third, African countries have a comparative advantage in the production and export of primary commodities given the abundance of natural resources, combined with the continent’s low human capital (Wood and Mayer, 2001) The increasing concentration of Africa’s exports on primary commodities casts some doubts on the potential for future growth in the region. Indeed, there is considerable evidence that the type of product that a country exports matters in terms of long-run economic performance (Hausmann et al., 2007; Lall et al., 2006). Manufactures, especially medium- and high-technology manufactures have forward and backward linkages with other sectors which may generate positive externalities for the whole economy. In contrast, primary products have production structures that are capital intensive and often poorly linked to the rest of the economy. Moreover primary products’ prices are set at the world level and are usually more volatile than those of manufactured products.
Another concern is the increase in the imports of manufactures from emerging countries (in particular China and India) and its impact on local manufacturing production in Africa. In most of the cases, domestic producers suffer this competition and are obliged to leave the market. In some fewer cases however, competition has increased the ability of domestic firms to compete (i.e. the Ethiopian shoes sector). Finally, for some other firms, the rise of China and India has offered some new opportunities. Indeed, the fact that these two countries are rapidly moving up the global value chain and leaving space for other developing countries to produce some of the low technology manufactured goods is an important opportunity for African countries. In order to take advantage of these new opportunities for export market expansion in the manufacturing sector, governments need to intervene by improving access to credit as well as addressing the problem of poor infrastructure.
All these elements support the argument in favor of the need of industrial policy to foster industrialization in Africa. Moreover, the comparison with the experience of emerging countries from East Asia show that industrialization has been the engine of growth. At the same time, the failure of past industrialization strategies in Africa calls for a renewed developmental industrial policy.
An emerging element in the African economic landscape is the industrial cluster.11 The fact that the cluster’s main feature is the geographic proximity between firms makes it to be particularly adapted to the African context characterized by poor infrastructure and weak information systems. Clusters are important because they are believed to play a significant role in the promotion and development of small enterprises. In general, the benefits of clustering are indicated as gains in collective efficiency. Indeed, clusters: 1) make market access easier; 2) are characterized by labor pooling; 3) facilitate technological spillovers; and 4) create the environment conductive to joint actions. Moreover, industrial clusters enhance enterprise performance by reducing transaction costs that would otherwise have been present when marketing through traders. It is clear that the benefits of clusters would be particularly valuable to African SMEs, given the difficult economic environment in which they operate.
Clusters are increasingly attracting attention from policy makers too, judging from the number of conferences organized on the topic and the growing research interests on the subject. A number of publications have analyzed the experience of clusters in Africa, describing their history, development and characteristics (see Mwamila et al., 2004; Oyelaran-Oyeyinka and McCormick, 2007; and Zeng, 2008).
McCormick (1999) provides a detailed analysis of six clusters in three African countries (Kenya, Ghana and South Africa). The cases considered show that African clusters, far from being homogeneous, vary in both internal structure and level of industrialization. The McCormick study interprets these three types as the subsequent steps any cluster should make to fully develop.
These are the groundwork cluster, industrializing cluster and the complex industrial clusters.
Groundwork clusters lay the groundwork for industrialization by improving producers’ access to markets and offering an environment in which joint action can begin. Industrialising clusters are “industrializing”, in the sense that they have begun the process of specialization and differentiation that leads to greater efficiency and ultimately to industrialization while complex industrial clusters have diversified their size structure and inter-firm linkages in such a way that they have been able to tap wider national and global markets.
Research on African countries has identified clusters, particularly of SMEs, as a source of competitiveness.12 There are a number of other aspects related to clusters that have been studied.
For instance, the analysis presented in Diyamett (2009) shows that only cooperation can make clusters work. The in-depth analysis presented in Zeng (2008) testifies that clusters in Africa are elements of vital economic activity. Indeed, belonging to these clusters has allowed enterprises to overcome many binding constraints in the areas of capital, skills, and technology. In some cases, 11 There different definition for cluster. For a review and a comparison of the alternatives see Navdi and Schmitz (1999).
12 Oyelaran-Oyeyinka (2004a) discusses the determinants of inter-firm and inter-organizational collaboration among SMEs in Nigeria, Kenya and Zimbabwe. Wood and Kaplan (2005) examine innovation and the role of firm networks in the South African wine industry; Lorentzen et al. (2007) analyses in detail the Durban auto cluster while Gibbon (2001) describes the cotton and the fresh fruit SME clusters in Tanzania.
it has allowed them to access global knowledge fostering production value chains, and achieving efficiency gains. The sectors identified range from natural resource-based activities, such as
fishing, high-tech industries, auto parts and computer manufacturing. The clusters analyzed are:
• The Suame Manufacturing Cluster in Ghana (also in McCormick, 1999);
• The Kamukunji Metalwork Cluster (also in McCormick, 1999) and the Lake Naivasha Cut Flower Cluster in Kenya;
• The Nnewi Automotive Components Cluster and the Otigba Computer Village Cluster in Nigeria;
• The Mwenge Handicrafts Cluster and the Keko Furniture Cluster in Tanzania;
• The Lake Victoria Fishing Cluster in Uganda;
• The Textile and Clothing Cluster in Mauritius; and • The Wine Cluster and the Western Cape Textile and Clothing Cluster (also in McCormick, 1999) in South Africa.
These studies emphasize the important role that a cluster could play in the African context.
For instance, Zeng (2008) concludes that clusters significantly contribute to Africa’s economic growth and to job creation. McCormick (1999) concludes that clusters are contributing to the industrialization process in Africa. Nonetheless, in today’s increasingly knowledge-intensive and competitive global economy, these clusters also face serious challenges. The big question is: how to deal with them? This will be discussed in the last section of the paper.