The extensive margin of trade growth refers to the expansion of exports through either the introduction of new products (product extensive margin) or entering new export markets (geographical extensive margin) (Reis & Farole 2012). This approach is widely supported in trade literature because of its potential to foster diversification, resilience and innovation in developing economies.

Hummels and Klenow (2005) were among the first to quantify the role of the extensive margin in export growth. The authors found that 60% of 126 countries’ large economies’ export expansion was attributable to new product or market entries between 1980 and 1997. This diversification reduces dependency on a limited range of exports and insulates economies from demand shocks or price volatility. More recently, Young (2024) found that in Nigeria, between 1960 and 2021, the extensive margin delivered larger and more sustained growth benefits compared to the intensive margin.

Exporting new products encourages research, development and innovation, as highlighted by Bernard et al. (2007). These dynamics are essential for improving competitiveness and adapting to international standards. According to Elumalai and Kumar (2025), India’s agricultural exports through the extensive margin helped stabilise trade profits, reduce the vulnerability to price shocks, and support sustainable agricultural development. Additionally, Bernard et al. (2007) found that entry into new markets creates employment opportunities in production, logistics and sales and stated that firms must scale up to meet new demand. This, in turn, supports inclusive growth.

Despite its potential, literature highlights high barriers to entry, such as limited information on new markets, high marketing and regulatory compliance costs, and uncertainty regarding consumer preferences (Matthee et al. 2015). These constraints are particularly relevant in developing contexts, where countries may lack the capacity to absorb the risks associated with market expansion (Matthee et al. 2015). Although South African trade policy often prioritises the extensive margin, studies by Steenkamp (2019), Matthee, Idsardi and Krugell (2016) and Vogel (2024) caution that successful diversification in South Africa requires significant investment in infrastructure, skills and market knowledge. Matthee et al. (2015) also found that more than three-quarters of South Africa’s export growth momentum was within the intensive margin between 2002 and 2012.

The intensive margin of trade growth involves increasing the volume of existing exports to already established trading partners. This margin is increasingly recognised for its low-risk, cost-efficient and high-impact characteristics.

Several empirical studies confirm the intensive margin as the dominant contributor to export growth in developing countries. Amiti and Freund (2010) observed that the bulk of China’s rapid export expansion (between 1992 and 2005) occurred along the intensive margin. Additionally, Fedyunina, Simachev and Drapkin (2023) found that Russia’s intensive margin was a significant contributor to economic growth under intensified sanctions between the years 2015 and 2021.

Benefits of intensive margin export growth include economies of scale and process efficiencies, improved trade relationships, better export survival and depth in trade, as well as gaining export experience to build export strength. Unlike the extensive margin, increasing volumes to known markets avoids high fixed costs related to new market entry (Hong & Luparello 2024). Exporters can optimise production and distribution channels, benefiting from economies of scale and process efficiencies (Hong & Luparello 2024). Strengthening existing trade relationships supports diplomatic and economic stability (Besedeš, Moreno-Cruz & Nitsch 2024). Familiarity with customer preferences, legal environments and logistics reduces uncertainty and enhances trust (Albornoz et al. 2012; Besedeš et al. 2024).

Besedeš and Prusa (2011) analysed disaggregated bilateral industrial exports from 46 countries (between 1975 and 2003) and found that trade survival – the ability to maintain export relationships – is essential for long-term export growth. They argued that weak export performance stems from the high failure rate of new trade relationships (70% of new export relationships fail within 2 years). They identify ‘depth’ in trade (increased volumes) as a stronger driver of export expansion than merely forming new trade ties. Carrère and Strauss-Kahn (2017) show that firms with prior export experience – particularly in nearby or accessible markets – are better able to sustain exports to developed economies.

Given the benefits of both export growth margins, a dual approach to export growth is advocated: building export strength through existing channels (intensive margin) while also venturing into new ones (extensive margin).

As a developing country, South Africa may find it more challenging to change structurally, diversify its exports, and find a successful, balanced export mix that requires large investment and infrastructure development (Matthee et al. 2015; Vogel 2024). The country may find it easier to focus on its strengths and, while pursuing export diversification, not forget about its intensive margin (Matthee et al. 2016). Vogel (2024) supports this argument by indicating that developing countries are faced with obstacles such as limited resources, expertise and global connections that may hinder the development of new product–country ties that will withstand market competition. Therefore, they concluded that a developing country’s intensive margin exports should keep growing (Vogel 2024).

The literature does not present the extensive and intensive margins as mutually exclusive but rather as complementary strategies. The extensive margin is vital for long-term structural transformation, innovation and market diversification, while the intensive margin provides immediate gains, stability and resilience – particularly critical for developing economies like South Africa.

Policymakers are thus encouraged to balance both strategies. While the extensive margin ensures future readiness (diversification) and competitiveness, the intensive margin represents momentum in the short to medium term for sustained export-led growth.

This study sets out to identify and prioritise underutilised export opportunities within South Africa’s intensive margin. Having provided an overview of the literature on the export growth margins within the context of South Africa’s specific challenges, an overview of international market selection literature follows.

International Market Selection (IMS) refers to the systematic process of identifying, evaluating and prioritising foreign markets that offer the best opportunities for a country or firm to export goods and services. Effective IMS is crucial for developing countries, where resources for export promotion are often limited and must be directed towards markets with the highest potential return (Papadopoulos & Denis 1988).

Selecting the right markets not only improves export performance but also enhances economic growth, industrialisation and trade resilience by allocating efforts and investments towards strategically viable destinations.

Research underscores the consequences of inadequate or ad hoc market selection, such as wasted financial resources in pursuing unviable markets (Rahman 2003); entry into high-risk or saturated markets (Papadopoulos & Denis 1988); failure to achieve sustainable trade relationships (Besedeš & Prusa 2011).

Therefore, scholars and practitioners advocate for structured IMS frameworks that combine multiple dimensions of market potential with quantitative methods for evaluation and prioritisation (Cavusgil 1997; Kumar, Stam & Joachimsthaler 1994). These frameworks typically assess markets based on several dimensions, which are widely supported in the literature. The most common and essential criteria include: (1) import market size; (2) import market growth; (3) market accessibility; and (4) export competitiveness.

Import market size represents current demand for a given product in the target country, with larger markets considered more attractive because of volume potential (Aucamp 2020; Cameron et al. 2021; Cuyvers et al. 1995; Kefi 2021; Kumar et al. 1994).

Import market growth tracks changes in import demand over time – short-term and long-term – and indicates whether a market is dynamic and expanding. A growing market may compensate for smaller current size, making it strategic for future entry (Aucamp 2020; Cameron et al. 2021; Cavusgil 1997; Papadopoulos, Chen & Thomans 2002).

Market accessibility evaluates ease of entry, considering factors like tariff and non-tariff barriers (NTMs), logistics and trade infrastructure and the competitive landscape in the target market. Accessibility is essential for estimating costs, risks and time to entry (Kefi 2021; Rahman 2003).

Finally, export competitiveness indicates whether the exporting country has a competitive edge in producing and supplying a specific product (Cameron et al. 2021; Kefi 2021).

From the empirical literature, recent reports identified export opportunities in different sectors for various regions. In sub-Saharan Africa, there is a rise in consumer goods and agrifood (specifically packaged food, poultry, dairy and personal care products) (DHL 2025). Processed food, transport products and plastic products have export potential in South and East Africa (TIPS Industry Study 2024). In the Caribbean and Latin American regions, nearshoring opportunities in transport products, textiles, pharmaceuticals and renewable energy are expanding rapidly (IDB Report 2022; S&P Global 2025).

Lazaros and Dadakas (2024) found potential for the European Union to expand machinery and transportation exports to countries within the Intergovernmental Authority on Development (IGAD) region of Africa, as well as to countries with whom they have existing trade agreements in the American continent. Moreover, their results revealed trade potential for transportation products in the Middle East (Lazaros & Dadakas 2024). Konstantakopoulou and Tsionas (2024) implemented the Decision Support Model (DSM) to identify realistic export opportunities for the Southern Euro Area countries. Their results showed export opportunities for manufactured goods in all Southern Euro Area countries, machinery and transport products in Italy, Spain and Portugal, and food and live animals in Greece. Cuyvers et al. (2017) also identified realistic export opportunities for Thailand. Their results identified exporting sectors by using the DSM. The authors found that exports in food products, transport products, chemicals and plastics, electronics, rubber and rubber products, machinery, and textiles to Thailand’s neighbouring Association of Southeast Asian Nations (ASEAN) + 3 countries are in growing demand (Cuyvers et al. 2017). Viviers et al. (2014) used the DSM to identify new export opportunities for South Africa. The authors’ results showed that South African exports of minerals, agro-processing food products, machinery, electrical, chemicals and plastics, textiles, and transport products showed export potential (Viviers et al. 2014). More recently, Ferreira and Steenkamp (2025) explored how sub-Saharan African (SSA) countries can improve intra-regional trade by identifying sector-based export opportunities, drawing lessons from Southeast Asia’s integration experience. The authors used the DSM model and found that exports in agro-processing and food products, pharmaceuticals, textiles, chemicals and plastics, construction materials, and lights (consumer) manufacturing products are of high demand in SSA intra-African trade (Ferreira 2020; Ferreira & Steenkamp 2025).

International market selection is a strategic imperative for enhancing export-led growth (Chi, Chang & Chien 2025; Çimen 2025; Cuyvers et al. 2017; Fan, Bei & Hu 2024; Ferreira & Steenkamp 2025; Konstantakopoulou & Tsionas 2024; Naudé & Cameron 2025; Viviers et al. 2014). The literature consistently supports the use of structured, multi-criteria models.

From the literature discussed in Section International market selection methods and market attractiveness criteria, it is clear that many approaches to international market selection exist. Criteria that are commonly used include the size of import demand, growth in import demand, market accessibility and the exporter’s competitiveness. For the purposes of this study, selected criteria of a quantitative country-level international market selection model called the DSM were used. This method was originally designed by Cuyvers et al. (1995) for Belgian export promotion purposes. The criteria used in the model have since been applied to other countries, including Thailand (Cuyvers et al. 2017); Namibia (Spies, Idsardi & Steenkamp 2014; Teweldemedhin & Chiripanhura 2015); Rwanda (Cameron & Viviers 2017); Louisiana, USA (Oluwade 2018); South Africa (Viviers et al. 2014), China (Cameron et al. 2021); the South Euro Area (Konstantakopoulou & Tsionas 2024); selected African countries (Ferreira 2020; Ferreira & Steenkamp 2020, 2025); and the European Union (Naudè & Cameron 2025). This model was selected because it starts with all possible product-and-destination country combinations worldwide and systematically filters through these (Aucamp 2020; Aucamp, Steenkamp & Bezuidenhout 2025). The model picks winners (by means of an elimination process), which aligns with the purpose of Phase 1 (Ferreira 2020).

The original DSM was developed for government export promotion purposes and comprises four filters. To limit the amount of data to be analysed on a detailed product-level, entire countries were eliminated in Filter 1 based on political and/or commercial risk and macroeconomic indicators. Filter 2 considered the size, short-term and long-term growth in import demand on a product–country level for the remaining destination markets (Cuyvers et al. 1995). In Filter 3, the degree of concentration (Filter 3.1) and trade restrictions or barriers (Filter 3.2) were considered to evaluate the market access conditions in each market (Cuyvers et al. 2012). Filters 2 and 3.1 were also applied in the studies of Ferreira and Steenkamp (2020; 2025), Spies et al. (2014), and Viviers et al. (2014). To inform export promotion strategies, Filter 4 classifies the product-country combinations identified with potential in Filters 1 to 3 by mapping the size and growth of import demand against the exporting country’s market share relative to the main competitors in the import market (Cuyvers et al. 2012).

For this study, Filters 2 and 3.1 of the DSM are applied. Because of the developments in information technology since the late 1990s, the early elimination of entire countries in Filter 1 is not necessary, and the analysis can start evaluating import demand on a detailed product-country level (more than 5199 product codes and 227 countries worldwide, totalling almost 1.2 million possibilities to start with). In the original model, Filter 3.2 deals with trade barriers. These are incorporated in Phase 3 of this study (see Phase 3: Prioritisation among the underutilised, intensive margin export opportunities). Rather than eliminating product-country combinations based on trade barriers, the approach in this study follows Ferreira and Steenkamp (2020) to keep the opportunity but rank it lower if high trade barriers exist. Filter 4 involves a categorisation specifically for export promotion purposes and does not attempt a ranking of opportunities. This study, however, aims to prioritise export opportunities in Phase 3.

In Filter 2, short-term import growth, long-term import growth and import market size are reviewed (Cuyvers et al. 1995). Import growth is the rate at which a country’s imports of a particular product group are increasing. The value of imports of a specific product group by a particular country is referred to as the import market size. The short-term import growth is considered the most recent year’s growth rate in imports, whereas the long-term growth is calculated as the compound average annual percentage growth in imports over 5 years. The third variable, import market size, is calculated by the ratio of imports of country i for product group j and the total imports of all countries that entered this filter of product group j (Cuyvers et al. 1995, Cuyvers 2004). The selection criteria of this filter are based on global averages for import size and growth per product (Cuyvers et al. 2012).

When defining the cut-off values for short- and long-term growth in import demand, a scaling factor (s_j) was calculated to take into consideration South Africa’s level of specialisation in exporting product j (Cuyvers 2004:260), as follows (see Equation 1):

where Revealed Comparative Advantage (RCA_Saj) is the revealed comparative advantage of South Africa in exporting product j (Reis & Farole 2012), in Equation 2:

with X_SAj and X_wj being South Africa’s and the world’s total exports of product j; and X_SA,tot and X_w,tot being total exports of South Africa and the world.

The cut-off values for the growth variables were then expressed as: g_ij ≥ G_j

with g_i,j, the importing country i’s short- or long-term import growth rate of product j; and:

and g_w,j, the growth rate of aggregate world imports of product j.

In terms of import size, the cut-off values are defined as S_j (Cuyvers 2004:260): M_ij ≥ S_j

where M_ij is the country i’s total import of product j; and:

with M_w,j denoting aggregate world imports of product j.

South Africa’s competitive advantage in exporting the product is therefore taken into consideration when defining the cut-off values for the growth and size of import demand. This implies that, if South Africa was not yet specialised in exporting the product (0 ≤ RCA_SAj < 1), the destination country’s import growth rate (g_i,j) should be higher than the global average for that product. The growth rate in import demand is allowed to be marginally lower than the world average growth rate, where South Africa had a comparative advantage (RCA_SAj ≥ 1) in exporting the product (Cuyvers et al. 2012:62–63). Similarly, for the size of import demand, the cut-off value is 2% of total world imports of the product if South Africa is specialised in exporting the product, and a bit stricter, at larger than 2% and up to 3% of world imports, if South Africa does not have a comparative advantage in exporting the product (Cuyvers et al. 2012).

For final selection in Filter 2, import growth and size are combined. All markets that qualify in terms of size are selected. Those with short- and long-term growth above the cut-off values are also selected. Then, any market displaying a combination of adequate size and growth (short- and/or long-term) also selects into Filter 3.

Cuyvers et al. (1995:180) stated that choosing a market based on sizable and/or growing import demand (Filter 2) does not ensure that the market will be accessible. Therefore, analysing the level of competition in each import market is also crucial (Filter 3.1). In addition, recent studies like Fan et al. (2024) and Çimen (2025), also used market concentration and competitiveness metrics to identify export opportunities.

Concentrated markets are characterised by a small number of exporting countries (competitors) with significant market shares and expertise in the industry (Vollrath 1991). A market with a high level of concentration is difficult to penetrate. Cuyvers et al. (1995:180) found a negative association between market concentration and exports. As a result, channelling resources into highly concentrated markets may be wasteful, and export survival chances are low. Therefore, Filter 3.1 assesses the accessibility of potential markets by measuring the concentration of competitors using the Herfindahl-Hirschman index (HHI) of Hirschman (1964). The HHI for market concentration of Filter 3.1 was calculated as follows (see Equation 3):

where X_k,ij represents the exports of product j to country i by country competitors k; and M_tot,ij is the total imports of product j in country i.

An HHI value closer to 1 means that a country sources a particular product from a limited number of suppliers with dominant market shares, indicating a highly concentrated market. On the other hand, HHI values closer to 0 indicate markets characterised by a diverse set of suppliers with relatively small individual shares. As noted by Cuyvers et al. (1995), markets controlled by one or two suppliers are generally more difficult to access compared to those with a broader base of competitors. Cut-off values are defined accordingly.

The data of this phase were obtained from the CEPII BACI database (CEPII 2024a).

Phase 2 matches export opportunities for South Africa identified in Phase 1 to the intensive margin product–country combinations based on the CEPII BACI database (2024a). If South Africa exported the product to the destination country in 2012, that combination was regarded as part of South Africa’s intensive margin (existing product-country relationship).

Furthermore, a potential export value was calculated for each of the matched markets (based on calculations of Cuyvers et al. [2017] and Ferreira and Steenkamp [2020]). This potential export value was calculated as the average market value of all the competitors (for the product) in the import market. South Africa’s existing (or actual) exports were then evaluated against these potential values to identify underutilisation within the intensive export growth margin for South Africa.

Therefore, export opportunities within the intensive margin that were underutilised were obtained at the end of Phase 2.

The purpose of the DSM is to provide a list of interesting markets for further investigation (Cuyvers 2004). Previous attempts to prioritise among the export opportunities identified in the model involved rankings based on the export potential values explained in Phase 2. However, these values only take into account the size of the import demand and the number of competitors. Market attractiveness is, however, a multidimensional concept that also includes other key determinants such as import growth, market accessibility and export competitiveness. This is supported by the international market selection literature covered in Section Literature review.

Composite indexing is often used to measure multidimensional concepts (OECD 2019). Therefore, Phase 3 of this study uses this method to construct an MAI to prioritise among the underutilised export opportunities with growth potential in South Africa’s intensive margin.

Figure 2 illustrates the dimensions and variables included in compiling the MAI.

The first dimension, size of import demand, includes the total import value per product–country combination in US$1000. A weighted 5-year average of these values was used with more recent years weighing more.

The second dimension, growth in import demand, includes the short- and long-term growth in import demand. The short-term growth in import demand is the simple 1-year growth in import value, while the long-term growth is the compounded growth rate over 5 years. Both growth figures were calculated using data from the CEPII BACI database.

The third dimension, market access, includes the ad valorem equivalent tariff applied to South Africa, the World Bank Logistics Performance Index (LPI), market concentration, and distance. The ad valorem equivalent tariff applied to South Africa is the latest available tariff that South Africa faces when exporting a particular product to a specific destination country (ITC 2024). The World Bank LPI measures the on-the-ground efficiency of trade supply chains, as well as the reliability and predictability of the logistic service delivery of a country (World Bank 2023). The World Bank constructed the LPI based on a global survey of logistics professionals, evaluating countries across six key dimensions: (1) the efficiency of customs and border clearance procedures; (2) the quality of trade and transport infrastructure; (3) the ease of arranging competitively priced international shipments; (4) the competence and quality of logistics services; (5) the ability to track and trace consignments; and (6) the timeliness of shipments reaching their destinations. The performance of each country is determined with the use of a five-point scale, rated from very low (1) to very high (5) in relation to generally accepted industry standards or practices (World Bank 2023). Market concentration is also part of this dimension. It is measured by means of the HHI (see Section Phase 1: Identification of product-country combinations with export potential), using data from the CEPII BACI database. Higher HHI values indicate more concentration, which is regarded as lower market access and/or attractiveness. Distance is often used in gravity analysis as a measure of market access (De Benedictis & Taglioni 2011:75). It is measured in kilometres and is calculated by the weighted distance from South Africa to the major city of the importing country, considering the share of each city in the importing country’s total population. The data were obtained from the CEPII GeoDist database (CEPII 2024b).

The last dimension, export competitiveness, includes the RCA and Revealed Trade Advantage (RTA) variables for South Africa. The RCA is a metric that indicates whether a country has an advantage in production and export capacity for a particular product (see Section Phase 1: Identification of product-country combinations with export potential). Therefore, when RCA is larger than 1, the country is assumed to be a competitive producer and exporter of that product, as opposed to a country producing and exporting that product at or below the global average. The RTA variable includes both exports and imports and is used to determine whether South Africa is a net exporter (and not a re-exporter) of a particular product (Jessen & Vignoles 2004). The RTA is determined by subtracting a country’s RMA (calculated the same as the RCA, only using import data) or a certain product from its RCA (Vollrath 1991). This variable therefore indicated the items that South Africa regularly exports competitively by using the data from the CEPII BACI database. As the international trade environment changes frequently, the most recent 5 years for which data are available at the time of analysis were used in this study (2017–2021). Where applicable, more weight was given to more recent values. The weights assigned were 30% for year 5 (most recent); 25% for year 4; 20% for year 3; 15% for year 2; and 10% for year 1. This weighted average method was used for the size and growth of import demand, market concentration (HHI), RCA and RTA variables.

As the variables included are measured in different units, they must be standardised before they can be included in a composite index (OECD 2019). The standardisation method used in this study was to calculate the distance from the best (100) and worst (0) performers.

The different variables in each dimension were then standardised using the following formula, as seen in Equation 4 (OECD 2019):

where y_ij is the value of the variable for product j in destination country i; y_l is the lower threshold; and y_u is the upper threshold as defined below.

The upper threshold is the top 1% of product-country combinations in the applicable variable, and the lower threshold is the bottom 1%. The only exception for these calculations is the tariffs variable. When calculating the index of this variable, the upper threshold was 0% (no tariff applied) and the lower threshold was 30% (considered a high tariff, ITC [2024]).

Finally, equal weights were used within each dimension and among the four dimensions (25% each) to arrive at the MAI. This MAI was then used to prioritise the underutilised export opportunities in South Africa’s intensive margin identified in Phases 1 and 2.

Ethical clearance to conduct this study was obtained from the North-West University, Economic and Management Sciences Research Ethics Committee (NWU-00675-24-A4).