Economic growth is one of the most and still discussed issues in economics literature with diverse views on its determinants. The literature identifies several factors that contribute to economic growth such as policies, institutions and geography (Lee and Kim 2009). In those factors, mostly in international trade, policies are often represented by trade openness and integrations with the so-called “Washington Consensus” (Williamson 1994, 1996, 1990). However, the literature argues that good policy prescriptions fail due to poor institutional environments such as insecure property rights and a weak rule of law (Acemoglu, Johnson, and Robinson 2001, 2002). Specifically, traditional factors would not bring about effects on economic performance in an absence of stable and trustworthy institutional environment to sustain the economy (Easterly, 2005). Other authors, alike, have a divergent view on the robustness of institution variable. Glaeser et al. (2004) argued that it is economic growth which brings in good institutions, such as democracy, citing the case of formerly authoritarian states like South Korea. Lee and Kim (2009) find that low- and lower-middle-income countries are the only countries where institutions are significant.
The other stream of research addresses economic growth question by focusing on the ‘growth spurts and collapse’ in short periods of time which is a more prevalent phenomenon in many countries in the South. It has been established that most of the third world countries could show growth spurts for a certain period of time, less than a decade, then fail to sustain that growth over a longer period (Jones and Olken, 2005; Hausman et al, 2005; Rodrik, 2006). This underlines the importance of building a more sustainable growth rather than focusing only on initial growth. A closely related phenomenon is the so-called ‘middle-income trap’, which indicates the problem of declining growth in middle-income countries (Eichengreen et al, 2012, 2013; Lee, 2013; World Bank, 2010).
Another well-studied area of economic growth factors is connected to economic integration and openness, or trade variables in the literature. The real effects of openness and variables that would best represent the international integration are still under debates (Dollar 1992; Ben-David 1993; Sachs and Warner 1995; Edwards 1989; Vamvakidis 1999; Harrison 1996). Researchers like Frankel and Romer (1999), Rodriguez and Rodrik (2001) and Yanikkaya (2003) find that economic growth and trade openness are positively linked. While some studies find that trade openness, as a factor for economic growth, is not robust (Rodriguez and Rodrik 2001; Vamvakidis 2002; Lee and Kim 2009). Similar disagreements are observed for the FDI4 variable between pro-FDI and skeptical-FDI groups (Hermes and Lensink, 2003; Carkovic and Levine, 2002; Adams, 2009). Another controversy is around the export diversification where some studies find this idea is significant for economic growth in the South while others find export specialisation to have significant effects on growth (Ramanayake and Lee, 2015). The study by Ramanayake and Lee (2015) tested the robustness of export specialisation variable as a growth determinant using different estimation techniques and finds that the variable was a robust determinant of growth in developing countries.
Export diversification has been a dominant element in the discussion of growth dynamics in developing countries since the 1950s. There has been a mix of different theoretical and empirical propositions since then regarding its growth effects.
4 FDI is not included as a variable of interest in this study because it does not necessarily influence growth through export.
Export Diversification: What Does It Really Mean?
Export diversification can generally be defined as the changing of country’s export composition and structure. The process can be achieved by changing existing export commodities pattern or through expanding innovation and technology on them. Dennis and Shepherd (2007) describe export diversification as broadening the variety of products that a country is exporting. Actually, export diversification can take mainly two-dimensional forms namely, horizontal and vertical (Ali et al., 1991; Herzer and Nowak-Lehnmann, 2006).
Generally, a horizontal diversification of exports is simply an increase in the number of primary products mix which usually takes place within the same export sector. Referring to studies by Herzer and Nowak-Lehnmann (2006) and Samen (2010), adding new products on existing export basket within the same sector helps reduce the effects of fluctuation of global commodity prices and alleviate adverse economic risks. This brings forth the stability in export-oriented sector earnings and independence of export-oriented growth from a certain sector (Al-Marhubi, 2000). To achieve economic growth by a way of horizontal diversification, a country should either increase its share of products in the market to increase export earnings or introduce new products which can fetches good prices in the world market (Ali et al., 1991). Herzer and Nowak-Lehnmann (2006) argues that a horizontal diversification of exports generates positive externalities to other sectors of the economy brought about by the dynamic learning activities in export-oriented sectors acquired through exposure to foreign firms and international competitions.
On the other hand, a vertical diversification of exports occurs when a country’s export structure shift from primary products to secondary or tertiary sectors, or manufactured products. The process employs a use of existing and new advanced merchandises by undertaking value-addition such as processing and marketing (Poverty and Development Division, United Nation, June 2004). A manufacturing production process of this nature creates spill-over effects in the form of externalities on knowledge and new technologies, relative to a production of primary exports which does not generate such spillovers (Matthee and Naudé, 2008). Al-Marhubi (2000) and Herzer and Nowak-Lehnmann (2006) stress that such spill-over benefits going to other sectors generate and improve capabilities of other industries to compete in the world market. These improvements bring the stability of export earnings as prices of manufactured exports are less fluctuating compared prices of primary exports (Ali et al., 1991). According to Ali et al., 1991, growth via vertical export diversification comes either by introducing and expanding value-added activities or select new products based on their value-added potentials. Hausmann et al. (2007) concluded that export structure matters in a country that has a higher productive capacity with a diversified export structure and it performs better in the world export market.
Both horizontal and vertical export diversifications can produce positive results for a country’s economic growth, but their performances have different dependencies on the technology, marketing and skills. Vertical diversification requires more advanced technology, sophisticated policies, skills and initial capital investment relative to the horizontal diversification. As noted above, vertical diversification may result to more dynamic externalities than that of horizontal diversification.
Theoretical Review: Export Diversification and Growth
A theoretical argument for a connection between export diversification and economic growth was originally advanced in the 1950s by Raul Prebisch and Hans Singer. The idea mainly focused on growth hindrance by the export of primary goods. The argument by Prebisch (1950) and Singer (1950) is that a strong export concentration of developing countries on primary goods impedes growth, declines the terms of trade and escalates the instability of income. This theoretical proposition is known as the “Prebisch-Singer Hypothesis”. To prevent instability of income, a country needs to diversify its export composition, the effect identified as the “portfolio effect”. Moreover, developing countries need to compete in the international market with other countries that export similar goods. Consequently, the rise of prices in one country renders products of that country less competitive in the international market since their products will easily be substituted by products of their rivals. This mechanism works mainly due to a low-income elasticity of the international demand for primary products. The demand for manufactured goods increases more rapid than the demand for primary products, making the terms of trade for exporters of primary commodities to weaken in the long run. Dogruel and Tekce (2011) noted that there is a low growth spillover to another sectors from the production of primary products in the economy due to the impact of low-skills and poor technology in the primary sector. Under this framework, “diversification” has become a common goal of economic policies in the less developed countries (Brainard and Cooper, 1968; Dogruel and Tekce, 2011).
The second mechanism by which export diversification might positively affect economic growth is through the dynamic knowledge spillover effects. Knowledge spillovers range from new techniques of production, new management, or marketing practices from exporting industries to possibly assist other industries through imitations and adoptions (De Piñeres & Ferrantino, 2000). Al-Marhubi (2000) noted that an improved production technique which is associated with export diversification in one of exporting sectors is likely to assist other industries through knowledge spillovers. The knowledge externalities include productivity improvements resulting from increased international competitiveness, more efficient management styles, better organisation forms, labour training, and knowledge in terms of technology and international marketing (Herzer & Nowak-Lehnmann; 2006). The success of this strategy brings desirable effects on resource allocation and increases firms’ profits. While resource reallocation raises the income level, the dynamic profit from export diversification plays an important role in increasing the rate of income growth (Hamed et al, 2014). This follows from the result of increased use of factory’s capacity utilisations, achieving economies of scale, and job creations. Growth is stimulated through exporting labour-intensive products which trigger a multiplier effects to increase the demand for intermediate inputs, consumer goods and leads to the rise of total factor productivity (TFP). Moreover, the life cycle models literature (e.g Vernon, 1966; Krugman, 1979; Grossman and Helpman, 1991) argues that the diversity of export products innovated by the North is imitated by the South and exploited by taking the advantage of cheap labour the South has.
However, the whole concept of export diversification appears to challenge the classical trade theories predictions, particularly the Ricardian theory of comparative advantage which predicts countries to specialise (Salvatore, 1998, Matthee and Naudé, 2008). Ricardo argues that countries gain in the international market by specialising in the production of products in which they have a comparative advantage and thereby increase total productivity. So to achieve economic growth by means of Ricardo’s conception, a country should promote a sector in which it has a comparative advantage. A similar view is shared with other classical theories which are based on perfect competition, comparative advantage and constant returns to scale. The theories are based on Adam Smith’s idea of division of labour and specialisation for economic growth and development, and the Heckscher-Ohlin-Samuelson (HOS) model of international trade, which predicts that countries would specialise in producing goods in which they have a comparative advantage based on their factor endowments. Contemporary literature, however, finds that the acceleration of global trade in the latter half of the 20th century has revealed a pattern of trade which is vastly contrary and countries appear to diversify their production and exports as they grow (Krugman, 1980: Hesse, 2006). Helpman and Krugman (1985) argue that greater economies of scale caused by increased exports can lead to growth in the level of productivity. Uncertainty, however, still remains in many scholars on impacts of specialisation on long run growth despite all the relationships identified between trade and productivity. For example, Sachs and Warner (1997) identified the negative influence of comparative advantage in raw materials on economic growth.
Lately, diversification and specialisation issues have been examined as part of the endogenous consequence of a country’s stage of development (Acemoglu and Zilibotti, 1997); Imbs and Wacziarg, 2003). The model used is constructed based on country’s production because the level of production affects the level of exports. Ramacharan (2006) identifies that a one standard deviation rise in diversification is accompanied with about a 0.81 standard deviation growth in the level of credit to the private sector. From that finding, diversifying a structure of sectors in an economy will favour developments of the financial sector which further allows a country to involve in a more specialised mode of export, provided that a financial market provides an insurance cover against risks (Chang, 1991). In view of the thesis above, it is more plausible to argue that export composition structure may go by phases, from low diversified to more diversified, then followed by a phase of a small extent of diversification and more specialisation, as financial sector expansion develops (Saint-Paul, 1992). This would further imply that diversification of production structure for domestic economy requires export diversification first and later export specialisation.
Other researchers also pointed out theoretical reasons which suggest export diversification leads to a higher per capita income growth in the long-run. In order to stabilise the country’s exports in the long-run, many developing countries opt for the policy of export diversification and liberalisation. As a result of the above-stated volatility, risk-averse firms would not invest in a country where its macroeconomic environment are unstable and can be unfavourable to the long-term economic growth. In that respect, many countries liberalise their trade. Michaely (1958) conducted the study on export and import concentration using the GINI coefficient for a dataset of 44 countries and 150 Standard International Trade Classification of commodities and found that countries with a more diversified export structure are more developed in terms of income per capita. He also noted that countries with a higher diversification were more industrialised in terms of primary commodity share in the total exports. Therefore, export diversification can be useful for long-term policy targets to stabilise export earnings (Ghosh and Ostry, 1994; Bleaney and Greenaway, 2001). This proposition is also supported by the structural economic model which infers that for a country to achieve a sustainable growth, it should move from primary export to manufactured exports (Syrquin, 1989).
Another theoretical explanation was advanced by Marianne Matthee and Wim Naudé who argue that a move towards more diversification leads to a spatial inequality. The observation was made from trade liberalisation effects which cause small businesses to suffer and further leads to a fall in GDP and consequently retards economic growth. Once firms have the capacity to export more in the international market due to liberalisation domestic firms become less dependent on the domestic market resulting in a decrease of agglomeration forces (Matthee and Naudé, 2008). Moreover, not all developing countries benefit from more exports since their country location can also be an important factor for their low export tendency. However, export diversification still contributes to country’s growth (Herzer and Nowak-Lehnmann, 2006).
At this juncture, the emerging theoretical idea is that a heavy dependence on a narrow range of export products renders the country’s exports unstable in case of a negative demand shock for her products. Export diversification makes the country’s export more stable and less vulnerable to demand shocks. According to Matthee and Naudé (2008), however, the stability from diversification comes at a price of effective resource allocation which is associated with specialisation benefits.
Resource misallocation is a potential problem emanating from concentration in exports of a certain product. A boom from a natural resource discovery could take away resources from a manufacturing sector and most likely it leads to a real exchange rate appreciation. This problem has further consequences in an economy, the so-called “the Dutch Disease”, which is a decline in the competitiveness of country’s trade products in the world market. Since natural resource abundant countries earn adequate foreign exchange currencies required for their importations, they usually have little incentive to industrialise (Dogruel and Tekce, 2011). If industrialisation takes place in this case, normally a country specialises in a production of physical goods which are capital intensive rather that in goods which are knowledge-intensive, which brings out adverse impacts on equality in wages and human capital developments (Bonaglia and Fukasaku, 2003). However, instead of viewing the abundance of natural resources as a curse, should be regarded as an opportunity to build capacity in competitive advantages in non-traditional goods (Bonaglia and Fukasaku, 2003).
However, the empirical evidence for the Prebisch-Singer hypothesis shows a negative relationship between the abundance of natural resources and growth. The evidence from Ng (2006) suggests that it is not the abundance of natural resources that hinders growth, but a concentration of exports in primary goods. For instance, some of the natural resource abundant countries like Canada, Australia and the Scandinavian countries started with a concentrated export pattern in primary goods but later they positively diversified their exports (Hesse, 2008). According to Carrère et al. (2007), the evidence in favour of the Prebisch-Singer hypothesis shows that moving away from primary products is the only appropriate strategy, not that diversification is desirable by itself.
The proposed influence of export diversification on growth by the Prebisch-Singer hypothesis faces challenges from other propositions in the literature. One hypothesis is referred to as the Export-led Growth (ELG) hypothesis, which stresses that a growth of exports stimulates the total factor productivity (TFP) growth and by its positive impacts on higher rates of capital formation helps to relax foreign exchange currency constraints by facilitating imports of capital goods (Lee and Huang, 2002). The ELG further explains that a rise in exports implies that there is an increase in the country’s demand for its goods and services. It also indicates an increase of the GDP of a country and/or there is a reallocation of productive resources toward industries that have high efficiencies at the global level (Awokuse, 2008). This proposition suggests a case that it is the growth of exports rather than the diversity of exports which promote growth and export growth should have a higher priority.
Trade openness is another alternative hypothesis linked to determinants of economic growth. This hypothesis is founded on the Smith (1776) ideas, the absolute advantage theory, that when a country opens up to the international trade gains an access to an extensive international market which leads to productivity improvements through a division of labour. A small size of the domestic market in most of developing countries limits economic growth. The major dynamic benefit derived from the international trade is that domestic producers get the access to extensive markets (Thirlwall; 2000). Similar argument support comes from the Ricardo (1817)’s theory of comparative advantage in factor endowments which maintains that the international trade is beneficial and leads to productivity and consumption gains to trade partners. Also, economies opened to international trading broadens their industrial sector domestically at a faster rate by importing needed inputs, technologies and machinery from economies which have advanced technologies in an easier manner than closed economies. The ideas suggest that a higher degree of openness has a greater influence on economic growth.
Therefore, the developed picture is that there is a disagreement between the traditional neoclassical theories with the new trade theories when it comes to effects of the export structure (composition) on economic growth. Most of the modern theoretical studies strongly advocate for a diversified export structure, contrary to classical theories, and they argue that a highly concentrated export structure dominated by few primary goods is vulnerable to external demand shocks which proved to have negative consequences on the stability of macro-economy and growth. Then, it is theoretically plausible to assume that a region which is dominated by exports of primary goods will have a limited growth, Africa in a particular case for instance.
Determinants of Export Diversification
The question of determining factors for export diversification, particularly in developing countries, is comprehensively studied in the literature too. One of the arguments put forward is the infant industries argument which maintains that countries attempt to protect their infant industries by means of both horizontal and vertical diversifications in intra-industry trades (Grossman and Helpman, 1991). The proposition was suggested by Grossman and Helpman (1991) after studying causes of human capital and R&D expenditures on export diversification. Herzer et al (2004) stated that a knowledge spillover brought about by openness is of high significance and is explained by externalities known as a “learning by doing” and predominantly the “learning by exporting” coming from diversification of exports.
The level of development is another suggested determinant of export diversification in literature, normally represented by the GDP per capita of a country. This argument is supported by the demand side and supply side theories that as the GDP per capita rises, a pattern of preference determining consumptions changes (Aghion and Howitt, 1992; Fiorillo, 2001). The change in the elasticities of demand forces the productivity of sectors to change and further changes the composition of an economy. As a result, the production of primary exports will undergo advancements to secondary products or tertiary sectors. The lack of capital and indivisibility of investment projects are cited as limits to exploitations of diversification potentials at a lower level of development (Acemoglu and Zilibotti, 1997).
Potential risks for diversification declines as an income increases and export diversification increases as GDP per capita rises (Imbs and Wacziarg, 2003). But, beyond a certain threshold of the income level, an incentive to diversify declines; as high-income economies have the propensity to be economically and institutionally more stable, and a necessity for diversification declines. Therefore, the growth of the GDP per capita has a positive influence on export diversification in the lower income country, whereas at the higher GDP per capita a further increase in incomes leads to a higher export concentration.
Foreign direct investment (FDI) is mentioned as one of the influencing factors on export diversification. FDI creates a diversification of export either direct to a non-traditional sector or indirect through raising the export of traditional goods which have a smaller share in exports (Gourdon, 2010). However, if FDI is directly poured into an exploitation of natural resources, it is likely to cause an increase in the concentration of exports with natural resources. For instance, Ekholm et al. (2007) examined the impact of FDI on export diversification and found that, under a certain condition, FDI results into export diversification in the South. For the case of a natural resources abundant country, Ekholm et al. (2007) argued that FDI greatly influences vertical diversifications than horizontal export diversification, meaning that most of the transferred knowledge in the process does not change a core of the economy’s export structure.
Trade policy of a country is cited as another determinant of export diversification. Melitz (2003) stated that under protectionist trade policies, a very limited number of firms will be able to participate in the export business since not all firms can cover the fixed cost associated with export activities and it may possibly lead to a more concentration of exports. Likewise, a trade policy that lowers tariffs as a result of trade liberalisation policy improves the country’s access to foreign market capital which in the long-run leads to export diversification as a country adjusts its capacity to serve a more diverse international market.
The relationship between export diversification and economic growth is well studied in the empirical literature with the main focus on developing countries. One of the studies was conducted by Al-Marhubi (2000) which used a cross-country sample with 91 countries for the period eight-year from 1961to1988 and it applied a cross sectional country growth regression. He reported the existence of a negative relationship between export concentration and growth. Other studies were conducted in Chile by de Piñeres and Ferrantino (1997) and Herzer and Nowak-Lehmann (2006) which revealed that the country positively profited from its diversified export sector. A negative relationship between the GDP per capita growth and concentration of exports was detected in the study by Lederman and Maloney (2003) from their regressions analysis of both cross-section and panel data. The findings by De Ferranti et al. (2002) affirmed a positive relationship and he further observed that a 1 percentage increment in the concentration of exports is accompanied with a 0.5 percentage fall in the GDP per capita growth.
Some of the empirical studies (e.g. Hausmann and Rodrik, 2003; Hausmann et al., 2007) have developed a theoretical framework and estimations for analysing a linkage between advantages of export diversification and export in the general economic growth. The uniqueness of their approach is based on the argument that it is not a comparative advantage that drives model of economic growth to invest in new activities but the country’s export diversification as suggested in other literature (Dogruel and Tekce, 2011). Hausmann et al. (2007) developed an indicator EXPY which measures the productivity of a country export basket. The EXPY also provides indications for economic developments. The Results of EXPY from their study shows some traded goods (service products, manufactured goods) in the basket have a greater level of productivity than others (primary goods). This implies that both productivity and economic growth of a country depends on what type of goods are produced (Dogruel and Tekce, 2011). Countries which have high-productive goods would experience a faster growth than countries which produce low- productive products.
2. LITERATURE REVIEW
2.1. Export Diversification: What Does It Really Mean?
2.2. Theoretical Review: Export Diversification and Growth
2.3. Determinants of Export Diversification
2.4. Empirical Review
3. EXPORT DIVERSIFICATION IN AFRICAN ECONOMIES
4.1. Export Diversification Measure
4.2. Empirical Growth Model and Estimation
5. EMPIRICAL RESULTS AND DISCUSSION
6. CONCLUDING REMARKS
GET THE COMPLETE PROJECT