«Policy Research Paper Industrial Policies for the Structural Transformation of African Economies: Options and Best Practices No. 2 Policy Research ...»
3) Population level (large vs. small: country is defined as large (small) if its population is larger (smaller) than the African mean population) The starting point of the typology is Ndulu et al (2008) which categorize SSA countries based on two criteria: a) endowments (resource–rich countries56 vs. non-resource rich); and b) location (coastal or landlocked, based on the idea that coastal countries have higher opportunities than landlocked economies). Interestingly, they argue that the geographical aspect is not important when the country is resource-rich because the endowment element outweighs the geographical element. It follows that their categorization yields three groups: resource-rich (does not matter if coastal or landlocked), resource-scarce landlocked and, resource-scarce coastal countries.
In the categorization here proposed, the two dimensions discussed in Ndulu et al. (2008) are adopted and – as in their paper - the geographical location dimension is excluded when categorizing resource-rich economies. The novelty of the proposed categorization is the inclusion of an additional dimension to Endowments and Geographical location.
The new dimension is the Population level. The population dimension of a country gives some indications on how much – at least in theory – the industrialization process can be driven by domestic demand. The larger the population the more likely it is that domestic demand could be an engine for industrialization. Obviously, the population level is only a necessary condition since per-capita purchasing power would also be crucial in determining the dimension of domestic demand. Notwithstanding, the population is the pre-condition for any domestic driven development. Small states, independent of their economic development can hardly aim for a development strategy focused on the domestic market.57 Table 9 categorizes all African countries58 according to the three dimensions. 59 55 Other categorizations have been proposed by Ndulu et al (2008) (AERC typology), Diao et al, (2008) (the IFPRI typology), World Bank (2008) (World Development Report typology) and Thorbecke (2009).
56 To be classified as resource-rich at least ten per cent of the country’s GDP has to come from primary commodity value added).
57 As an alternative measure to total population, one could consider to use per-capita income. But the very high inequality that characterizes African countries could make the conclusion biased. Finally, one could overcome this problem using a dispersion measure but the lack of reliable data make this possibility not feasible at the moment.
58 The list includes a number of countries which are currently in a situation of high political and economic instability (i.e. Somalia and Cote d’Ivoire). Those were not excluded because, when the minimal conditions for the functioning of the economy will be met again, the best industrial policy would still be the one indicated in the proposed categorization.
59 Note that one may also consider demand related elements in the taxonomy. Among these particularly, relevance could be given to the Diaspora communities in prospective target markets which could surely be a lever that can be pursued to create market pull. This later factor has played a significant role in the growth, development and industrial development of China, India, Korea and even Israel. African Diaspora could indeed be a catalyst to industrial development.
Table 9: A categorization of African countries
Source: Author’s elaboration. Population data are from US Census Bureau and refer to year 2009. Note: Resource-rich (Resource poor) if more (less) than 10% of the country’s GDP comes from the primary commodity value added. Large (Small) if population is larger (smaller) than the mean population of African countries.
The next step would be to outline which is the industrial policy that best fits each category in Table 9. This will form the basis for identifying the country’s best industrial policy mix. This could constitute a new direction of research. One possible approach to achieve this result is the methodology proposed based by Lo (2010) (see box 1).
Box 1: Methodology to identify the best policy mix for a country
1. Consider three discriminating Criteria for the Countries:
(i) Resources (Rich (RR)/Poor (RP)), (ii) Population (Large (LP), Small (SP)), (iii) Geography ( Coastal (C), Landlocked (L) )
Then eight possibilities (eight groups) exist:
Group A: RR/LP/C; Group B: RR/LP/L; Group C: RR/SP/C; Group D: RR/SP/L;
Group E: RP/LP/C; Group F: RP/LP/L; Group G: RP/SP/C; Group H: RP/SP/L
2. Take a broad sample of African, American, Asian and European Countries that were industrially under-developed in 1960 and put each of them in the relevant group (among the 8 identified above) depending on its situation.
3. Compare the industrial performances of the countries in the sample in 1960, in 1980, in 1995, and in 2005. Show the wide and progressive variability of results in each group and for the whole sample.
Identify the top and the low performers globally and in each group.
4. Design a Synthetic Quality of Industrial Policy Index (QIPI) aggregating Three Sub-indexes:
(i) Sub-Index 1: Quality of Sectoral and Competition Policies (ii) Sub-Index 2: Quality of Education and Innovation Policies (iii) Sub-Index 3: Quality of Trade Policies For each Sub-Index, appropriate variables will be identified. The weight of the variables will vary among the groups, some of them being critical or very important for the Group due to the conditions it faces (eg.: Trans-borders roads or rail facilities for landlocked countries).
5. Give a Score to each country for the variables, sub-indexes and for the QIPI, using Data Analysis techniques (Multiple Factor Analysis): in 1980; 1995 and 2005. Verify that there is continuous correlation between scores for QIPI and industrial performances of the countries.
6. Have a ranking of the countries in the sample and identify which variables are the most important in each group to increase its QIPI (discriminating variables).
7. Draw conclusions and lessons for African Countries, depending on the Group where they are.
Source: Lo (2010)
Each country’s characteristic (endowment, geographical location, population) indentified in Table 9 is relevant to industrial policy. For instance, the population level not only influences competition polices but is also relevant for education and trade policies. A small country needs to trade more because it is not efficient to produce domestically for a very small internal market.
The first dimension is Endowments which is related to sectoral policies. Resource-rich economies should devise sectoral policies aimed at favoring their resource sectors. There are different possibilities in this sense (see for instance, Pietrobelli and Rabellotti, 2006) which range from cluster-based initiatives to credit facilitation. One of the main objectives of these policies should be the creation of the financial condition for the development of other sectors. Indeed, a full set of industrial policies could be implemented using the resources coming from the natural resource sector.
The second dimension is Geographical location which is related to trade policies. Given their disadvantaged location, landlocked countries should push for increasing regional trade integration, because they are the ones more likely to gain most from it. At the same time, they should devise policies to foster infrastructure investment and all the set of incentives to attract foreign firms and facilitate domestic export (i.e. through export promotion agencies). Given that they are among the poorest economies in the continent, they most likely would need external technical and economic aid. Coastal countries enjoy a better geographical position in terms of trade opportunities. Given their advantage in terms of transportation cost, their process of trade liberalization could be pursued with some more flexibility, leaving more policy space to governments in determining trade relations. For these countries, policies aimed at creating and improving physical and immaterial infrastructure that support export would be less expensive, most effective and thus more rewarding.
The third dimension is Population level which is related to competition policies. Only countries characterized by a large population could consider the possibility of reducing domestic firm competition with the objective of creating the opportunities to exploit economies of scale and prepare them for foreign competition. In other words, a large market is the necessary condition for implementing a controlled creation of economic rent in the domestic market as was done in the NICs (see Di Maio, 2009). As can be understood, the necessary condition is that the domestic market is sufficiently large. But it also requires a strong monitoring capacity by the government. In the case of small countries, the domestic market cannot be the main objective of domestic firms and that access to foreign markets becomes the priority.
The main advantage of this proposal is that it directly links the economic characteristics of the country to industrial policy60. Based on the success-stories of countries facing similar natural conditions, it shows a simple way to determine how to combine government intervention in the different domains (education, trade, technology, etc.) with a coherent and successful industrial policy.
60 Some important clarification and warning are in order. The objective of the typology is not to rank countries.
Indeed, this would be impossible since the combination of the characteristics does not in itself provide an obvious combination of best country conditions. For instance, the fact that a country is resource poor does not necessarily suggest a negative situation. Consider, for example, the case of Tunisia. The fact that in the taxonomy it results to be a resource-poor country is merely due to the fact that in this country the primary sector is small and industry and tourism represent a larger share of GDP. Thus the table should be carefully read and interpreted.
Conclusion Industrial policy considerations have recently become very topical in the international development agenda. African countries need to re-evaluate their industrial policy since it holds the key to the structural transformation of their economies. The fact that the discussion is again about which are the best industrial policies to be implemented by governments rather than if governments should intervene or not, creates the conditions for improvement of economic management in African countries.
The aim of this Report is to contribute to this discussion and to provide a tool able to inform policy makers and industrialization stakeholders about what should and can be done on the continent to buttress current industrialization initiatives and plans.
A number of important issues have been analysed in this Report. The starting point has been a careful description of the current situation of industry in the continent. The analysis has shown the numerous weakness of a small manufacturing base and an increasing role of primary production and export. This poses a number of doubts about future growth prospects for many African countries. The current characteristics of the economic structure of African countries are, among other elements, the result of the industrial polices implemented since independence. To evaluate their impact, the paper has provided a comprehensive overview of industrial policy in Africa - in terms of history, current characteristics and perspectives. In particular, the similarities and differences with other emerging regions concerning both the Developmental State period and the following structural reform era have been discussed. The central role of the external factor in determining the difficulties of the Developmental State and the consequent advent of the SAPs were emphasised. It has also been argued that the poor performance of the SAPs is one of the reasons for the renewed interest in industrial policy in the continent.
The current situation is also characterized by the increasing importance of world trade and by changing rules of the game. To evaluate how well African countries are facing the new challenges, a survey of the industrial policy instruments adopted by African governments has been provided.
The result shows that governments are employing a vast range of policies, from sectoral policies to education and trade policies. Moreover, new instruments are increasingly being employed to foster industrialization, namely policies to support industrial clusters and firms’ upgrade along the agro-industry value chains. To complement the description of the current situation with some policy recommendation, the paper suggests a way to identify the best industrial policy mix for each African country.
The first step is the definition of a new taxonomy of African countries based on their main economic characteristics. Based on this, the best industrial policy mix for adoption is proposed for each group of countries. It is argued that trade, education and innovation, sectoral and competition policies should be used in different combinations depending on the characteristics of the specific country. In this way, the peculiarity of each country is taken into consideration and the objective of each policy is clearly targeted.
The main contribution of the present report is a novel taxonomy of industrial policy for African countries. The aim is to provide a tool able to indicate which industrial policies governments should implement given the country’s characteristics but also able to give indication about which policy should be followed in a dynamic perspective. Although in need of refinement, this taxonomy is arguably a useful instrument in helping to choose the best industrial policy mix in each African country.