If the concept of CA is to be applied empirically (to estimate export opportunities of a given country), some methods of quantification need to be used. Classical and neo-classical theories are based on relative autarkic prices which are almost impossible to calculate in today’s interrelated economic world (Ballance et al., 1987). In view of this, Ballance et al. (1987) proposed a way to link the theoretical concept of CA with empirical application using the following scheme:
EC → CA → TPC → RCA
Economic conditions (EC) of a given country have a determinative impact on the CA the country holds, which copies the structure of international trade, production, and consumption (TPC). Revealed comparative advantage (RCA) is based on the indices used to calculate TPC. Ballance et al. (1987) state that it is appropriate to use post-trade indicators
12 to determine RCA. Some limitations were incorporated into the model, including: i) relating the amount of export, production, and consumption to the country size and importance of the good; ii) taking into consideration the impact of government policies that distort trade relations; iii) dealing with the aggregation of the data. Despite them, Ballance et al. (1987) state that using post-trade data will reveal the general pattern behind the export flows.
The most used index measured to reveal RCA on the basis of ex-post data is Balassa index (BI).
BI is based on available data on the exports of a certain commodity (a good) by a country, total exports of the country, world’s exports of the commodity (a good) and world’s total exports, and is measured as follows:
where x (c, i) is the amount of exports of product i from country c. The numerator expresses the share of exports of a certain good in the total exports of the country. The denominator indicates the share of exports of the same good in the world in the total world’s exports. If BI is higher than 1, the share of the exports of this good in the country is higher than the share of exports of the same good in the world’s exports, meaning the country has RCA and exports the product efficiently. BI enables not only to detect RCA, but also to compare the indexes between countries (Balassa, 1965).
BI was criticized in some respects in relation to its inability to precisely indicate CAs between countries. The demand bias in favour of a particular good can change the structure of net trade in a positive or negative way, while national preferences can only have a positive effect on net exports (Lundbäck & Torstensson, 1998). Trade and other policy measures (especially tariffs and non-tariff measures) can distort the results obtained by using RCA leading to lower possible efficiency that could have been achieved without introducing protectionist measures (Bojnec, 2001). When it comes to large economies that benefit from the economies of scale, trade liberalisation can significantly contribute to further anchoring of existing trade specialisation, thereby increasing return to scale (Bastos & Cabral, 2007).
Some researchers built on BI to define export opportunities of a given country. Russow and Okoroafo (1996), while working on their global screening model, defined three areas that have a significant impact on countries’ export opportunities and are omitted by BI:
13 i) market-size and growth of a country of destination; ii) costs and availability of factors of production; iii) economic developments of exporting country. These areas were further incorporated in other models estimating export opportunities of potential markets. While working on Decision Support Model (DSM), Cuyvers et al. (1995) decided to broaden the Balassa’s model by incorporating all above-mentioned areas in their research. After eliminating countries that are not sufficiently economically and politically stable (using macroeconomic indicators, such as credit risk ratings), the markets that do not show an adequate size or a sufficiently large growth of their economies to provide possibilities for exports are left out of the analysis (mainly based on the Gross National Product (GNP) and GNP per capita).
The second step is to eliminate markets that do not show adequate market size or sufficient growth using the following destination market indicators: short-term import growth (percentage growth of imports between two most recent years), long-term import growth (average annual percentage of growth of imports over a period of five years), and relative import market size (share of imports of country i for product j from a total import of the product j to country i).
After eliminating unpromising product groups in terms of the target market using cut-off variables, the analysis of trade restrictions is conducted to define realistic export opportunities.
The analysis contains two categories of barriers: the degree of concentration (since a partial analysis revealed that there is a significantly negative correlation between export performance and market concentration) and trade restrictions. The last step is assessing the export position of the exporting country for each of the remaining realistic export opportunities through BI.
The main shortcoming of the model is the high aggregation of data. The calculation is done at two-digit levels, which are very heterogeneous and the actual products they encompass may vary in terms of market attractiveness. Moreover, the assessment of target market is based on past developments in imports, omitting their predictions in the future (Cuyvers et al., 1995).
Another problem of BI, addressed by Lafay (1992), is that it is time invariant, which means it cannot show the evolution of CA over time. To monitor the development of CA over time, the Lafay index (LFI) was proposed. LFI is calculated as follows:
𝐿𝐹𝐼 (𝑐, 𝑖) = 100 ∗ (𝑥𝑐,𝑖−𝑚𝑐,𝑖
14 in the turnover of the item with the share of the total trade balance in the turnover of the country.
The weight here is the share of turnover of this item in the total turnover of trade. LFI does not measure CA in relation to other countries but shows a CA above the overall trade structure of the country. Therefore, positive LFI values show that a country has a CA and indicate the degree of specialisation of the item (the higher the value of the index, the higher the degree of specialisation). Negative values indicate a comparative disadvantage (degree of non-specialisation) of the given item. LFI average is therefore always zero. To compare the items according to their CA, calculated LFIs for all items are sorted in descending order.
In this way, it is easy to find out which of these items show the largest CA in a given country.
From the calculated values of the LFI index for individual items, it is also possible to calculate the cumulative value of the LFI (Lafay, 1992).
In 2016, the Ministry of Foreign Affairs of the Czech Republic prepared an input analysis to identify export opportunities. The analysis includes the intersection of CA development over time (using LFI), the growth dynamics of the target market, and the untapped EP of the exporter in the partner market with a two- to three-year outlook. The results were published in the Map of global industry opportunities and in the Map of strategic opportunities to provide Czech exporters with an overview of promising markets. The first step of the analysis is the selection of commodities in which the exporter shows RCA and the partner a comparative disadvantage (called export competence). The exporter’s export competence is measured by calculating CAs for individual product groups (or items) using LFI. An exporter has a CA in item A if the share of its net exports of item A in the turnover of item A is greater than the share of its total net exports in the turnover of foreign trade. Otherwise, item A has a comparative disadvantage.
According to the model, those items where the exporter has a CA and the importer comparative disadvantage are selected. The second factor is the target market’s imports growth dynamics.
When following the imports growth dynamics of the target market, it is important for the methodology that the growth rate does not slow down in the period under review.
The dynamics of the target market growth (MG) is calculated as a geometric mean as follows:
𝑀𝐺 (𝑐, 𝑖) = ∏ ( 𝑚𝑡,𝑐,𝑖
𝑚𝑡−1,𝑐,𝑖)
1 𝑛 𝑛−1
𝑡=2 ,
where mt,c,i is the amount of imports of product i to country c in year t. The model is interested in such items that have at least 10% year-on-year growth in imports in the target market over the last 4 years. Items whose import growth has been faster in the last two years than in the first
15 two years are selected in order to exclude the items for which import growth is saturated.
The third factor in the methodology is EP of the Czech Republic on a global scale. In this step, items that do not meet the expected EP of the Czech Republic in the target market are selected.
This means that the share of Czech exports of the item in the partner market is at least twice as small as its share in the world market. The final selection of items with export opportunities then includes those items that meet all three conditions (Tlapa et al., 2019).
The methodology presented in the Map of global industry opportunities has several limitations.
Firstly, its ambition is not an in-depth analysis of the competitive environment, which would actually affect the extent to which companies could establish themselves in foreign markets.
Moreover, the methodology is not capable of estimating to which extent the item is prospective, which does not allow the comparison of given items and choice of the most promising ones.
Another aspect is strictly defined criteria for import growth, which includes the minimum rate of 10% year-on-year growth in import in the target market over the last 4 years and the condition of higher imports growth in the last two years compared to the previous two. These conditions create high barriers, which impede most of the items from entering the category of products with export opportunities. It is also emphasised by the Ministry of Foreign Affairs that the model is not capable of capturing many of the factors affecting trade between the exporter and the target market (such as security and political risks, trade agreements, tariff advantages, cultural proximity, etc.). The model relies mostly on experts from each target territory to identify significant technical and non-technical barriers to entry into the target market for each promising item identified (Tlapa et al., 2019).
Several successful alternative attempts to create a model that would identify export opportunities have also been made, for example, by Green and Allaway (1985). They used the shift-share model, the core of which resides in the calculation of market share changes over a selected period of time. Data on imports of a certain product to a given country are used to calculate average market share change for each product-country combination (actual growth). Expected import growth is calculated based on the average growth of all importing countries. The difference between the actual and expected growth of each market is defined as a net shift and will be positive for markets that have gained market share in that country for a defined period of time. For comparison between product groups, the percentage net shift is divided by the total net shift of all markets included in the analysis. The main weakness of the analysis is the focus on import-only measures. What is more, the results may be biased
16 based on the years chosen since the model is based on only two points in time (Papadopoulos et al., 2002).