Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Analysis of Intra-Industry Trade in Agri-Food Products Between Bosnia and Herzegovina and the European Union

Analysis of Intra-Industry Trade in Agri-Food Products Between Bosnia and Herzegovina and the... Although research of intra-industry trade (IIT ) has been intensive in the last several decades, the empirical lit- erature focusing on this phenomenon in the Western Balkans countries remains limited, especially in agricul- tural sector. Aiming to contribute to the existing literature, the paper analyses patterns and country-specific determinants of IIT in agri-food products between Bosnia and Herzegovina (BiH) and the European Union (EU) during the period of their mutual trade liberalisation (2008-2018). The analysis employs Grubel-Lloyd indices and Greenaway-Hine-Milner methodology for measurement of IIT, and applies an econometric panel data model using a Poisson Pseudo-maximum likelihood approach in order to estimate effects of IIT determi- nants. The findings suggest that intra-industry agri-food trade of BiH with the EU is of low intensity and main- ly of vertical nature, viewed totally and bilaterally. We found significant positive effects of trading countries’ Snježana Brkić, PhD (corresponding author) sizes, common border and history on IIT, and negative Associate Professor effects of the geographic distance and differences in School of Economics and Business agricultural productivity. University of Sarajevo E-mail: snjezana.brkic@efsa.unsa.ba Keywords: Intra-industry trade (IIT ), agri-food prod- Address: Trg oslobođenja – Alija Izetbegović 1 ucts, Bosnia and Herzegovina (BiH), the European 71 000 Sarajevo Union (EU), Poisson Pseudo-maximum likelihood Bosnia and Herzegovina ORCID: https://orcid.org/0000-0003-1004-9855 (PPML) approach JEL classification: F14; Q17; O52. Radovan Kastratović, PhD Teaching Assistant University of Belgrade Faculty of Economics Belgrade, Republic of Serbia 1. Introduction E-mail: radovan.kastratovic@ekof.bg.ac.rs For decades, intra-industry trade (IIT) as a phenom- ORCID: https://orcid.org/0000-0002-6138-906X enon mostly related to product differentiation, scale economy and imperfect competition, has been a sub- Mirela Abidović Salkica, MSc ject of intensive research on the example of industrial Senior Assistant products and developed countries. At the same time, Faculty for Technical Studies University of Travnik IIT in agricultural sector has been neglected, due to Travnik, Bosnia and Herzegovina different characteristics of the products and markets E-mail: mirelaabidovic@hotmail.com compared to industrial sector, such as resource inten- ORCID: https://orcid.org/0000-0002-5325-5600 sity, low product differentiation and assumption of Copyright © 2021 by the School of Economics and Business Sarajevo 53 ANALYSIS OF INTRA-INDUSTRY TRADE IN AGRI-FOOD PRODUCTS BETWEEN BOSNIA AND HERZEGOVINA AND THE EUROPEAN UNION perfect competition. IIT in agricultural products is still and less developed countries when trading with more insufficiently researched, particularly for developing developed countries. In that case, the trade pattern of countries. less developed countries is dominantly inter-industry, In order to fill a gap in the empirical literature on for both industrial and agri-food products, while IIT is IIT in agri-food sector of less advanced countries, es- mostly of the vertical type . pecially transition countries of the Western Balkans If we observe characteristics of the countries with (WB) , the paper focuses on the analysis of intra-indus- which a given country trades, those characteristics will try agri-food trade of Bosnia and Herzegovina (BiH). Of allow us to predict some features of its trade pattern. all the WB countries, BiH has the total IIT the largest According to traditional trade theories, it is likely that body of research. There is however the single study of inter-industry trade (international trade in products IIT in agri-food sector of BiH, which was conducted by of different industries, based on differences in com- Mrdalj et al. (2017), but without modelling of IIT de- parative advantages) will prevail in relations between termination. The research was based on the K-means countries at different levels of economic develop - cluster analysis, aiming at identifying different clusters ment. By contrast, an intensive intra-industry trade i.e. of agri-food product groups related to comparative international trade in products belonging to the same advantages, IIT level and the ratio between unit values sector/industry, is expected to occur between similar of exports and imports. countries (Krugman and Obstfeld 2009), and especial- Our study covers the country’s IIT with the ly in the form of a prominent horizontal IIT. European Union (EU) in the agri-food sector for the Namely, the structure of IIT is composed of two period 2008-2018 which is characterised by the mu- components – the horizontal component and the tual trade liberalisation within the framework of the vertical one, which differ from each other in terms Stabilisation and Association Agreement (SAA). The of types of product differentiation. Although IIT in WB countries mostly have the EU candidate status, ex- homogenous products can emerge as well (due to cept BiH which still is only a potential candidate coun- seasonal variations, border trade, etc.), the phenom- try and very slow on its path to the EU membership. enon is primarily related to more differentiated prod- It was expected that the institutionalisation of trade ucts. Horizontal differentiation is based on the actual relations between BiH and the EU would result in the or perceived differences in products’ characteristics growth of the overall trade and IIT in many sectors, which do not result in the systematic variation in pric- including the agri-food one, and would change IIT es. Vertical differentiation implies varieties of different structure as well. This research provides a more com- levels of quality and, consequently, of different prices. prehensive insight into the characteristics (intensity, There are two different approaches to modelling trend and composition) of the agri-food IIT of BiH with horizontal differentiation. The love-for-varieties mod- the EU, and possible changes. The paper also aims at el, which was developed by Dixit and Stiglitz (1977), is identification and evaluation of country-specific IIT based on the assumption that consumers will use as determinants by estimating an empirical model us- many varieties of the same product as possible if they ing a Poisson Pseudo-maximum likelihood method are available. Lancaster (1980) however developed (PPML). the core-attributes model, which starts from the as- The paper is structured as follows: The next sec- sumption that each consumer prefers one variety with tion provides a short theoretical framework for the a special combination of characteristics. The theo- analysis of IIT characteristics and econometric analy- retical basis of horizontal IIT (HIIT) was further devel- sis. The third section describes the used methodology oped by Krugman (1980), Helpman (1981), Helpman and data. Research findings on the basic patterns of and Krugman (1985), etc. Vertical IIT (VIIT) models, as IIT of BiH with the EU in agri-food sector and country- developed by Falvey (1981), Falvey and Kierzkowski specific IIT determinants are presented and discussed (1985), and others, include the redefined traditional in the fourth section. The last part contains conclud- concept of comparative advantages as explanation ing remarks, including policy implications. of IIT, starting from the assumption that differences in the product quality result from the differences in production processes i.e. in factor endowment and technology. 2. Conceptual framework The intensity of total IIT and its components is Our research is based on the hypothesis that the driven by different factors, both country-specific and pattern and determinants of IIT of BiH in agri-food industry-specific ones. With respect to IIT determi- products are consistent with theoretical assumptions nants, our analysis starts from the hypothesis that and up-to-date empirical findings on IIT of transition country-specific determinants, such as economic size, 54 SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 ANALYSIS OF INTRA-INDUSTRY TRADE IN AGRI-FOOD PRODUCTS BETWEEN BOSNIA AND HERZEGOVINA AND THE EUROPEAN UNION difference in the level of development, geographic (2003) believes that FDI is a determinant which pri- distance, trade intensity, foreign direct investment, marily and positively affects VIIT, which was confirmed common past etc., influence IIT in agri-food products for the agri-food sector in the research by Jámbor and the same way as in the case of industrial products. It Leitão (2016). means that determinants which relates to any kind Almost every model of bilateral trade, whether it of proximity (economic, geographic etc.) of trading is the total trade or one of its components such as IIT, countries have a positive effect on their mutual IIT, takes into account the geographic distance. The dis- while differences between trading countries have a tance indirectly “measures” the effects of transport, negative effect. transaction and information costs on trade. Balassa We hypothesised a significant and positive effect (1986) established that all types of trade decrease of trading countries’ economic size, measured by their with increasing geographic distance, and it particu- gross domestic products (GDP). Reasons can be found larly applies to IIT. The negative effect of geographic on both the demand and the supply side. According distance on the IIT share in agri-food products was to Lancaster (1980), greater average size of two coun- also pointed out by Jing, Leitão, and Faustino (2010), tries’ markets will result in the greater import demand Jámbor (2015), Balogh and Leitão (2019), etc. On the for differentiated products, and more prominent IIT. other hand, geographically close, especially adjacent Krugman (1980) argues that a larger market creates countries typically have similar economic, cultural and a greater possibility for achieving economy of scale, other characteristics. In economic terms, it pertains which in turn leads to greater IIT. A positive impact to the similarity of the product structure and the de- of economic size on IIT was confirmed by research in mand structure between them (Venables, Rice, and agri-food sector as well, such as in Jing, Leitão, and Stewart 2003). Therefore, as opposed to the deter- Faustino (2010), Jámbor (2013), Bojnec and Fertő minant “geographic distance“, it is expected that the (2016), etc. determinant “common border“ has a significant and Difference in the level of economic development, positive impact on IIT. mostly expressed as difference in income distribution, Trade intensity is taken as the approximation of as an IIT determinant is rooted both in the theories the geographic trade orientation, indicating how im- of increasing returns and in the H-O theory. On the portant a trading partner is to the observed country. demand side, differences in incomes per capita (p/c) The positive effect of trade intensity on IIT in agri-food reflect differences in the structure of demand (Linder products was established in papers by Fidrmuc (2004) 1961, Lancaster 1980), while, on the supply side, they and Łapinska (2014). represent differences in factor endowment (Helpman Historical and political factors such as colonial his- 1981). The negative effect of the difference in incomes tory or a common state in the past are significant for p/c on the IIT was established in a number of stud- trade ties in general, and for IIT in particular. Countries ies, and, on the example of the agri-food products, in with a common past usually share some economic, research by Jing, Leitão and Faustino (2010), Leitão political and cultural characteristics that could inten- (2011), etc. The effect of the difference in incomes p/c sify IIT between them. Effects of political and histori- depends on the type of IIT, e.g. the difference typically cal factors on the example of intra-regional trade in has a positive effect on VIIT. the South East Europe were investigated by Trivić and In a similar way, Jámbor (2015) established the Klimczak (2015). negative effect of difference in productivity of trad- ing countries on their mutual IIT in agri-food products. However, the effect of difference in productivity could 3. Applied methodology vary depending on IIT type. Intensive VIIT occurs more 3.1. Measurement of IIT frequently among countries that are different in terms of productivity. Measurement of IIT mostly refers to calculating the IIT Difference in other economic performances such share or intensity in a given industry trade or trade of as success in attracting foreign direct investment (FDI) all industries of a country, using Grubel-Loyd indices. also affects IIT. However, in general, FDI effect on IIT Standard Grubel-Lloyd index (Grubel and Lloyd 1975) can be ambivalent. Some studies in agri-food sector is calculated as follows: such as those by Jámbor (2015), and Jámbor, Bologh �𝑋𝑋 � 𝐶𝐶 ���𝑋𝑋 � 𝐶𝐶 � �� �� �� �� and Kucsera (2016) determined a negative impact of 𝐺𝐺𝐺𝐺 �   �� 𝑋𝑋 � 𝐶𝐶 �� �� FDI inflow on IIT, while Leitão (2011) found a positive effect. Ambroziak (2012) and Jámbor (2013) found � � 𝐺𝐺𝐺𝐺 � � (1) that FDI can increase both HIIT and VIIT. Kandogan 𝐺𝐺𝐺𝐺 � � 𝐺𝐺𝐺𝐺 𝑤𝑤 �� �� SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 55 ��� � � 𝐺𝐺𝐺𝐺 � � (2) �� �� � �� �� 𝑤𝑤 � (3) �� � �� �� � �� �� ��� �� �� � (4) �� � �� �� � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� �� � � � � � � � ���� � � ���� � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� ��� � � ���� (5) � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � ���� � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� ��� � � ���� (6) � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � ���� � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� ��� � � ���� ��� � ��� ��� � ��� ��� ∑ ∑ �𝑋𝑋 � 𝐶𝐶 � � �𝑋𝑋 � 𝐶𝐶 � ��� ��� �� �� �� �� ��� 𝑉𝑉𝑉𝑉𝑉𝑉 � �𝑋𝑋 � 𝐶𝐶 � �� �� ��� ��� � � 𝑉𝑉𝑉𝑉𝑉𝑉 � � ��� � � � � � 𝑉𝑉𝑉𝑉𝑉𝑉 � ��𝐷𝐷𝑉𝑉�𝐵𝐵 ∆ 𝐺𝐺𝐷𝐷 𝐷𝐷𝑉𝑉𝐷𝐷𝑉𝑉 �� �� � �� � � � � � � � ∆ 𝑉𝑉𝑉𝑉 𝑅𝑅 ∆ 𝑉𝑉 � � �� �� �� ����� 𝐵𝐵𝐵𝐵 𝑅𝑅 �� 𝐷𝐷𝐷𝐷𝐶𝐶 𝐶𝐶 �� � � �� (9) � � � � � � �� 𝐶𝐶𝐶𝐶 𝐷𝐷𝑅𝑅𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷 𝑅𝑅𝑅𝑅𝑅𝑅 �𝑋𝑋 � 𝐶𝐶 ���𝑋𝑋 � 𝐶𝐶 � �� �� �� �� 𝐺𝐺𝐺𝐺 �   �� 𝑋𝑋 � 𝐶𝐶 �� �� �𝑋𝑋 � 𝐶𝐶 ���𝑋𝑋 � 𝐶𝐶 � � � 𝐺𝐺𝐺𝐺 � � (1) �� �� �� �� 𝐺𝐺𝐺𝐺 �   �� 𝑋𝑋 � 𝐶𝐶 �� �� 𝐺𝐺𝐺𝐺 � � 𝐺𝐺𝐺𝐺 𝑤𝑤 �� �� � � 𝐺𝐺𝐺𝐺 � � (1) ��� � � 𝐺𝐺𝐺𝐺 � � (2) 𝐺𝐺𝐺𝐺 � � 𝐺𝐺𝐺𝐺 𝑤𝑤 �� �� ��� GLij – standard Grubel-Lloyd index (IIT share) for the Trade in products whose unit values are beyond �� �� � �� �� � � 𝐺𝐺𝐺𝐺 � � (2) 𝑤𝑤 � (3) given industry i in trade between the given country this interval is iden �� tified as v � ertical IIT. Blanes and ∑ �� �� � �� �� ��� and another country/country group j; Xij – exports Martın (2000) divide VIIT into two categories – high of industry i from the given country to another coun- VIIT (VIITh) and low VIIT (VIITl) by using also the rela- �� �� � �� �� 𝑤𝑤 � (3) �� try/country group j; Mij – imports of industry i of the tive unit value: VIIT �� h (RUV abo � ve 1.15) signifies trade �� ∑ �� �� � �� �� ��� � (4) �� � �� given country from another country/country group j; in vertically differentiated products of higher quality �� i = 1, … , n – the number of industries. (meaning that quality of expor ts is higher than that of � � � �� Value 0 indicates purely inter-industry trade, while imports), while 𝑅𝑅𝑅𝑅 VIITl (RUV below 0.85) sig 𝑅𝑅𝑅𝑅 nifies trade �� �� �� � (4) �� � � � � �� � � � ���� � � � ���� � �� � index value 1 means that the entire trade is of intra- in vertically differentiated products of lower quality �� 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� industry type. For most product groups, the value of (meaning that quality of expor ts is lower than that of � � ��� � � ���� (5) G-L index is in interval between two extreme values imports). 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� �� � � � � � � � ���� � � ���� 0 and 1. � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 � � �� �� �𝑋𝑋 � 𝐶𝐶 ���𝑋𝑋 � 𝐶𝐶 � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� Grubel-Lloyd inde �� x can be calcula �� �� ted as an agg �� re- 𝐺𝐺𝐺𝐺 �   𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � ���� �� � � � ��� � � ���� (5) �𝑋𝑋 � 𝐶𝐶 ���𝑋𝑋 � 𝐶𝐶 � gate index (a trade-weigh 𝑋𝑋 ted a � 𝐶𝐶 verage of the industry 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� �� �� �� �� �� �� 𝐺𝐺𝐺𝐺 �   �� indices) for a country’s individual trading partner or a 𝑋𝑋 � 𝐶𝐶 �� �� � � ��� � � ���� (6) 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 � � 𝐺𝐺𝐺𝐺 � � (1) group of a country’s trade partners, as follows: �� �� 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � ���� � � � � 𝐺𝐺𝐺𝐺 � � (1) 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 � � � �� �� 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � ���� � � � � ��� � � ���� (6) 𝐺𝐺𝐺𝐺 � � 𝐺𝐺𝐺𝐺 𝑤𝑤 �� �� 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� ��� 𝐺𝐺𝐺𝐺 � � 𝐺𝐺𝐺𝐺 𝑤𝑤 �� �� � � ��� � � ���� ��� 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 ��� �� �� 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � ���� � � 𝐺𝐺𝐺𝐺 � � (2) � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 For the purposes of calculating aggregate HIIT and �� �� � � 𝐺𝐺𝐺𝐺 � � (2) ��� ��� ��� ��� � � where VIIT according to GHM method, we applied the index ∑ �𝑋𝑋 � 𝐶𝐶 � � ∑ �𝑋𝑋 � 𝐶𝐶 � ��� � � ���� ��� ��� �� �� ��� �� �� ��� 𝑉𝑉𝑉𝑉𝑉𝑉 � in the following form: �𝑋𝑋 � �� 𝐶𝐶 �� ��� � 𝑋𝑋 � 𝐶𝐶 � �� �� � �� �� �� �� � �𝑋𝑋 � 𝐶𝐶 � 𝑤𝑤 � (3) 𝐺𝐺𝐺𝐺 �   �� �� �� � ��� �� ∑ �� �� � �� �� �� ��� ��� �𝑋𝑋� �� 𝐶𝐶 � �� �� 𝑤𝑤 � (3) �� � � ��� ��� � ��� ��� �� �� � ∑ ��� ∑ �� �� �𝑋𝑋 � 𝐶𝐶 � � �𝑋𝑋 � 𝐶𝐶 � ��� ��� ��� �� �� �� �� ��� � � 𝑉𝑉𝑉𝑉𝑉𝑉 � � ��� 𝑉𝑉𝑉𝑉𝑉𝑉 � � � 𝐺𝐺𝐺𝐺 � � (1) � � ∑ �𝑋𝑋 � 𝐶𝐶 � �� �� �� ��� �� w – a share of industry i in total trade of the given ij � � � (4) � � � � �� � � � 𝑉𝑉𝑉𝑉𝑉𝑉 � ��𝐷𝐷𝑉𝑉�𝐵𝐵 ∆ 𝐺𝐺𝐷𝐷 𝐷𝐷𝑉𝑉𝐷𝐷𝑉𝑉 �� �� �� �� � �� �� � country with the country/coun �� try group j. ��� � (4) � � 𝑉𝑉𝑉𝑉𝑉𝑉 � � ��� �� � 𝐺𝐺𝐺𝐺 � � 𝐺𝐺𝐺𝐺 �� 𝑤𝑤 The common feature of most �� �� �� methods for empiri- � � � � � � ∆ 𝑉𝑉𝑉𝑉 𝑅𝑅 ∆ 𝑉𝑉 � � � � �� �� �� 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 ��� cal division of IIT into its horizontal and vertical com- where X and M stand for industry exports or im- �� �� � � � � � �� � � � � � � � ���� � � ���� � � 𝑉𝑉𝑉𝑉𝑉𝑉 � ��𝐷𝐷𝑉𝑉�𝐵𝐵 ∆ 𝐺𝐺𝐷𝐷 𝐷𝐷𝑉𝑉𝐷𝐷𝑉𝑉 � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� � ponents is their basis on the differences in quality, i.e. ports respectively; H, V distinguishes �� HIIT fr � om VIIT, 𝑅𝑅𝑅𝑅 �� 𝑅𝑅𝑅𝑅 �� ����� 𝐵𝐵𝐵𝐵 𝑅𝑅 �� 𝐷𝐷𝐷𝐷𝐶𝐶 𝐶𝐶 �� � � �� (9) � � � � � � �� �� �� �� � � � � � � � ���� � � ���� � � 𝐺𝐺𝐺𝐺 � � (2) � � on the use of the unit value as a quality indicator. We i represents an industr � y and � j represen � ts a trading � � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� ∆ 𝑉𝑉𝑉𝑉 𝑅𝑅 ∆ 𝑉𝑉 ��� � � ���� (5) � � �𝑋𝑋 � 𝐶𝐶 ���𝑋𝑋 � 𝐶𝐶 � �� �� �� �� �� �� �� used Greenaway, Hine, and Milner (1995) approach partner of the given country (i,j = 1,…, n). 𝐺𝐺𝐺𝐺 �   �� ��� � � ���� (5) 𝑋𝑋 � 𝐶𝐶 (GHM methodology) based on the pioneering paper �� �� ����� 𝐵𝐵𝐵𝐵 𝑅𝑅 �� 𝐷𝐷𝐷𝐷𝐶𝐶 𝐶𝐶 �� � � �� (9) � � �� �� � � � � � � � �� �� �� 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� 𝑤𝑤 � (3) by Abd-El-Rahman �� (1986), � or more precisely on his ∑ �� �� � 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � ���� � �� �� � � ��� � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 � � 𝐺𝐺𝐺𝐺 � � (1) 𝑅𝑅𝑅𝑅 �� �� relative unit value index (RUV � ): �� �� 3.2. Sample and econometric specification 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � ���� � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� ��� � � ���� (6) �� We considered the decomposition of the dependent �� 𝐺𝐺𝐺𝐺 ��� � � � 𝐺𝐺𝐺𝐺 � ���� (6) 𝑤𝑤 � (4) �� �� �� � �� � variable into its horizontal and vertical components. �� 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� ��� �� However, the preliminary analysis revealed that the 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � � � ���� � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� 𝑅𝑅𝑅𝑅 �� � � �� �� RUV – ratio between exports and imports unit value vertical IIT strongly dominated the agri-food trade 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � ���� ij � �� � � 𝐺𝐺𝐺𝐺 � � (2) �� � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� � � � � � � � ���� � � ���� �� �� for industry i in tr � ade of the given country with c � oun- flows between BiH and the EU. For this reason, we fo - ��� � � ���� ��� 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� try/country group j; UV – unit value of exports for cused on examining the determinants of total IIT. ij ��� � � ���� ��� ��� � � ���� (5) �� �� � industry i; UV – unit value of imports for industry i. In order to investigate how various determinants �� �� ij 𝑤𝑤 � (3) ��� ��� ��� ��� � �� � � ∑ �� �� � Horizon ∑tal IIT e �𝑋𝑋 xists when e � 𝐶𝐶 � � xpor ∑ t unit v �𝑋𝑋 � 𝐶𝐶 alues ar �e affect the intra-industry agri-food trade of BiH with the �� �� ��� ��� �� �� ��� �� �� ��� � � ��� ��� ��� ��� 𝑉𝑉𝑉𝑉𝑉𝑉 � � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 rela �tively close to import unit values of the given EU member states we estimated an econometric mod- ∑ �𝑋𝑋 �� � 𝐶𝐶 � � � ∑ ���𝑋𝑋 � 𝐶𝐶 � ��� �� �� ��� �� �� ∑ �𝑋𝑋 � 𝐶𝐶 � ��� 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � �� �� � ���� ��� � � 𝑉𝑉𝑉𝑉𝑉𝑉 � produc � t, i.e. when the r 𝑅𝑅𝑅𝑅 atio between these unit v 𝑅𝑅𝑅𝑅 alues el, using the data on bilateral trade between BiH and �� �� ∑ �𝑋𝑋 � � 𝐶𝐶 � �� �� ��� �� �� ��� is within the ±0.15 or ±0.25 interval, depending on the 28 member states and providing a panel of 308 ob- � (4) � � 𝑉𝑉𝑉𝑉𝑉𝑉 � � ��� �� � ��� � � � ���� �� (6) �� selected dispersion factor (par ��� ameter α). servations. Due to data considerations, we adopted a � � 𝑉𝑉𝑉𝑉𝑉𝑉 � � ��� more aggregate approach in the analysis, using coun- � � � �� � � � 𝑉𝑉𝑉𝑉𝑉𝑉 � ��𝐷𝐷𝑉𝑉�𝐵𝐵 ∆ 𝐺𝐺𝐷𝐷 𝐷𝐷𝑉𝑉𝐷𝐷𝑉𝑉 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� � 𝑅𝑅𝑅𝑅 �� �� �� 𝑅𝑅𝑅𝑅 � �� �� tries as units of analysis. The IIT indices were calculat- � � 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � � ���� � �� � � � � � � � � ���� � � ���� � � 𝑉𝑉𝑉𝑉𝑉𝑉 � ��𝐷𝐷𝑉𝑉�𝐵𝐵 ∆ 𝐺𝐺𝐷𝐷 𝐷𝐷𝑉𝑉𝐷𝐷𝑉𝑉 � � �� �� � 𝑅𝑅𝑅𝑅 � � 𝑅𝑅𝑅𝑅 � �� � 𝑅𝑅𝑅𝑅 � � �𝑅𝑅𝑅𝑅 �� �� ed from the raw trade data provided by the Agency �� �� ∆ 𝑉𝑉𝑉𝑉 𝑅𝑅 ∆ 𝑉𝑉 � � �� �� �� � � � � � � for Statistics of BiH (BHAS), classified according to ∆ 𝑉𝑉𝑉𝑉 𝑅𝑅 ∆ 𝑉𝑉 ��� � � ���� ��� � ��� � � ���� (5) � �� �� �� ����� 𝐵𝐵𝐵𝐵 𝑅𝑅 �� 𝐷𝐷𝐷𝐷𝐶𝐶 𝐶𝐶 �� � � �� (9) � � � � � � �� ����� 𝐵𝐵𝐵𝐵 𝑅𝑅 �� 𝐷𝐷𝐷𝐷𝐶𝐶 𝐶𝐶 �� � � �� (9) � � � � � � � � �� 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� ��� ��� ��� ��� 56 � � SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � ���� ∑ �𝑋𝑋 � 𝐶𝐶 � � ∑ �𝑋𝑋 � 𝐶𝐶 � ��� ��� �� ��� � �� �� ��� 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 𝑉𝑉𝑉𝑉𝑉𝑉 � �� �� ∑ �𝑋𝑋 � 𝐶𝐶 � �� �� ��� ��� � � ���� (6) ��� � � 𝑉𝑉𝑉𝑉𝑉𝑉 � � ��� � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � ���� � � � � � � 𝑅𝑅𝑅𝑅 � 𝑅𝑅𝑅𝑅 𝑉𝑉𝑉𝑉𝑉𝑉 � �� �𝐷𝐷𝑉𝑉� �𝐵𝐵 ∆ 𝐺𝐺𝐷𝐷 ��𝐷𝐷𝑉𝑉𝐷𝐷𝑉𝑉 �� �� � �� � ��� � � � ���� ��� � � � � � ∆ 𝑉𝑉𝑉𝑉 𝑅𝑅 ∆ 𝑉𝑉 � � �� �� �� ����� 𝐵𝐵𝐵𝐵 𝑅𝑅 �� 𝐷𝐷𝐷𝐷𝐶𝐶 𝐶𝐶 �� � � �� (9) � � � � � � �� � ��� ��� � ��� ��� ∑ ∑ �𝑋𝑋 � 𝐶𝐶 � � �𝑋𝑋 � 𝐶𝐶 � ��� ��� �� �� �� �� ��� 𝑉𝑉𝑉𝑉𝑉𝑉 � �𝑋𝑋 � 𝐶𝐶 � �� �� ��� ��� � � 𝑉𝑉𝑉𝑉𝑉𝑉 � � ��� � � � � � 𝑉𝑉𝑉𝑉𝑉𝑉 � ��𝐷𝐷𝑉𝑉�𝐵𝐵 ∆ 𝐺𝐺𝐷𝐷 𝐷𝐷𝑉𝑉𝐷𝐷𝑉𝑉 �� �� � �� � � � � � � � ∆ 𝑉𝑉𝑉𝑉 𝑅𝑅 ∆ 𝑉𝑉 � � �� �� �� ����� 𝐵𝐵𝐵𝐵 𝑅𝑅 �� 𝐷𝐷𝐷𝐷𝐶𝐶 𝐶𝐶 �� � � �� (9) � � � � � � �� 𝐶𝐶𝐶𝐶 𝐷𝐷𝑅𝑅𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷 𝐶𝐶𝐶𝐶 𝐷𝐷𝑅𝑅𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷 𝐶𝐶𝐶𝐶 𝐷𝐷𝑅𝑅𝐷𝐷 𝐶𝐶𝐶𝐶 𝐷𝐷𝑅𝑅𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷 𝑅𝑅𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅𝑅𝑅 𝐶𝐶𝐶𝐶 𝐷𝐷𝑅𝑅𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝐶𝐶𝐶𝐶 𝐷𝐷𝑅𝑅𝐷𝐷 𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝑅𝑅𝑅𝑅𝑅𝑅 𝐷𝐷𝐷𝐷 𝑅𝑅𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅𝑅𝑅 the Standard International Trade Classification (SITC) There is considerable variability between the Revision 3. Data were aggregated for 42 individual country pairs, particularly in terms of economic size, three-digit product groups associated with the agri- income differences, distance, and differences in agri- food sector, as defined by UNCTAD (2021). cultural productivity. The panel is unbalanced due to As there is no universally accepted form of speci- 10 missing values of IIT, which are caused by the ab- fying the IIT equation, we guided our econometric sence of trade between BiH on one side and Estonia, modelling approach by following the underlying Latvia, Lithuania and Malta on the other in some ob- theoretical model and by adopting the most com- served years. Using Little’s test (Little 1988) we de- monly included variables in the closely related empiri- termined that the aforementioned observations are cal studies. The descriptive statistics of the sample are missing completely at random. Additionally, we con- provided in Table 1, along with the definitions of the sidered using the random-effects variant of Heckman’s variables considered in the model, their definitions, sample selection model (Heckman 1979), to account data sources, and expected effect on IIT. for the fact that some observations are missing due to Table 1. Description and descriptive statistics of the variables Data Exp. Variable Variable description Mean Std. dev. Min Max source effect IIT The intensity of intra-industry agri-food BHAS N/A 0.068 0.097 0 0.401 jt trade measured as the value of GL index. ASIZE The economic size expressed as the aver- IMF + 325.652 477.445 12.963 1984.866 jt age of nominal GDPs of BiH and a coun- try j (in current billions EUR) in year t. DGDPC The inequality in income per capita IMF - 28.331 21.762 2.174 113.494 jt measured as the absolute difference in per capita GDP (in current thousands EUR) between BiH and a country j in year t. DIST The geographical distance as a direct CEPII - 1.157 0.572 0.29 2.363 straight-line distance in thousands kilo- meters between capital cities of BiH and a country j. BORDER The common border represented by CEPII + 0.036 0.186 0 1 dummy variable that equals 1 if an observed pair of countries has common border and 0 otherwise. TI The trade intensity as the share of a trad- BHAS + 0.019 0.041 0 0.271 jt ing partner’s market in the foreign trade of BiH in agri-food products in year t. DPROD The difference in productivity measured FAO - 28.528 22.458 0 99.331 jt as the absolute difference of the value- added per worker in agriculture in BiH and a country j in year t in thousands EUR. DFDI The difference in FDI measured as the UNCTAD - 0.283 0.369 .004 1.882 jt absolute difference between FDI stock in BiH and in a country j in year t in millions USD. COMMON The common history represented by CEPII + 0.107 0.31 0 1 dummy variable that equals 1 if an observed pair of countries has common state or colonial relationship in the past, and 0 otherwise. Source: Authors‘ own calculation. SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 57 ANALYSIS OF INTRA-INDUSTRY TRADE IN AGRI-FOOD PRODUCTS BETWEEN BOSNIA AND HERZEGOVINA AND THE EUROPEAN UNION the lack of trade between the countries. However, the The model we estimated has the following form: preliminary estimations of the model revealed that � � � � 𝐼𝐼𝐼𝐼𝐼𝐼 � ��𝐷𝐷𝐼𝐼�𝐵𝐵 ∆ 𝐷𝐷𝐼𝐼𝐷𝐷𝐼𝐼 the Inverse Mills ratio and the correlation between �� �� � �� � the error terms in the selection and primary equations � � � � � � ∆ 𝐼𝐼𝐼𝐼 ∆ 𝐼𝐼 are insignificant, implying the absence of sample se - � � �� �� �� lection bias. For this reason, we based our empirical ����� 𝐵𝐵𝐵𝐵 𝐷𝐷 �� 𝐷𝐷𝐷𝐷𝐶𝐶 𝐶𝐶 �� � � �� (9) � � � � � � �� approach on estimating a single augmented gravity- type equation, which is particularly useful for analys- ing the economic relationships on the bilateral level. where dependent and independent variables are de- The use of the gravity-type equation as a work- fined in the same manner as presented in Table 1, α horse model is the most commonly adopted ap- denotes a constant, β and δ denote slope coefficients, proach in the related literature (Jing, Leitão, and j is the index of the EU trade partners of BiH, t signifies Faustino 2010; Łapinska 2014; Jámbor, Bologh, and the time period, and μ , λ and ε refer to individual ef- j t jt Kucsera 2016). In nearly all studies the gravity equa- fect, time effect and the error term, respectively. tion is linearised using a log-transformation and then Our baseline model is a two-way panel data model, estimated using the ordinary least squares, gener- including individual and time-specific effects, which alised least squares, and generalised method of mo- were included in the model due to the panel structure ments. However, such an approach is inadequate in of the data and the joint statistical significance of the the cases where there is a significant share of zero val- aforesaid effects. The inclusion of individual and time- ues of dependent variable, as it leads to loss of obser- specific effects allowed us to control for unobserved vations and the dependency between the error term heterogeneity across the countries and time periods and covariates, resulting in inconsistent estimates and the impact of factors not explicitly included in the (Silva and Teneyro 2006). In our sample, 23.49% of ob- model, which reduced the risk of misspecification. Our servations contain zero dependent variable (which re- model is essentially an augmented gravity equation. flect instances of perfect inter-industry trade), making We use a static specification, as there is no theoretical the aforesaid problem a non-negligible concern. The justification for using the dynamic one in the context common solution to the problem of zeros in estimat- of IIT, and as it is the most widely used approach in the ing gravity-type equations is the application of PPML related empirical literature (Fertő and Hubbard 2002; method. This method, originally proposed by Silva Jing, Leitão, and Faustino 2010). In model estimation, and Teneyro (2006) allows the gravity equation to be we were primarily interested in the significance of β estimated in its original multiplicative form, accomo- and δ coefficients, which are directly related to our ini- dating to zero values of dependent variable (Burger, tial hypotheses of the impact of various determinants Van Oort, and Linders 2009). The estimator was shown on the IIT intensity. to be robust to heteroskedasticity if the conditional Finally, in order to reduce the risk of spurious re- variance of the dependent variable is proportional to sults, we evaluated the stationarity of the panels using its conditional mean. This assumption is not violated, a Fisher-type stationarity test (Maddala and Wu 1999). as there is only one process generating zero and miss- The test results for the continuous variables are pre- ing values of the dependent variable in our sample. sented in Table 2. Table 2. Panel unit root test results for continuous variables Variable Without time trend With time trend 2 2 Modified χ p-value Modified χ p-value IIT 1.6555 0.0489 3.4774 0.0003 jt ASIZEjt 8.8983 0.0000 2.8699 0.0021 ΔDGDPCjt 6.8023 0.0000 2.6353 0.0042 ΔTIjt 3.4924 0.0002 3.9437 0.0000 DPROD 4.4063 0.0000 5.2172 0.0000 jt ΔDFDI 10.4041 0.0000 7.5441 0.0000 jt Source: Authors’ own calculations based on Maddala and Wu (1999). 58 SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 𝐶𝐶𝐶𝐶 𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 In the initial testing, we found that DGDPC, TI, and The highest average GL indices in BiH agri-food DFDI contained unit roots in levels. However, the first trade with the EU, which are also the only indices with differences of the variables were found to be station- the value higher than 0.50, have been registered in ary. For this reason, the three variables are included in the following product groups: 061 Sugars, molasses our model as first-order differences. and honey (0.78); 056 Vegetables, roots and tubers (0.69); 054 Vegetables, fresh, chilled, frozen or simply preserved (0.65); 046 Meal and flour of wheat and flour of meslin (0.59); and 421 Fixed vegetable fats and 4. Results and discussion oils (0.52). (Appendix, Table 1A) 4.3. Analysis of intra-industry trade patterns As indicated in Figure 2 which illustrates IIT and its structure in bilateral trade flows, inter-industry trade Over most of the analysed period agri-food trade is more significant than IIT in agri-food trade with all with the EU indicated characteristics of strong inter- member states of the EU. The highest average shares industry trade although the trend of IIT was mostly in- of IIT were registered in trading with Croatia (0.35), creasing (except in 2011 and 2014). The increase in IIT Italy (0.32), Slovenia (0.18), Austria (0.15), and France intensity was particularly large in 2013, when Croatia (0.14), followed by the Czech Republic, Sweden, as one of the main BiH trade partners joined the EU Germany and Belgium. The importance of IIT was (Figure 1). particularly low in agri-food trade with the other 19 Structure of IIT is to a large extent dominated by its countries (GL<0.05). With six countries (Baltic coun- vertical component. Besides, over several years (2010- tries – Estonia, Latvia and Lithuania, and Ireland, 2016), the share of high VIIT (VIITh) was larger than Luxembourg and Malta), the trade in agri-food prod- that of low VIIT (VIITl). The share of HIIT was relatively ucts was either non-existent in some years or the en- small, though it was more significant in the beginning tire trade was of inter-industry type (GL=0.00). The of the period than later. However, the level of HIIT and similar applied to the trade with Denmark, Poland, high VIIT taken together indicate the quality advan- Portugal and Spain, where GL index was minimal tage i.e. the situation that the quality of BiH exports of (GL=0.01), and to the remaining nine countries, where agri-food product groups in which IIT was registered, average GL indices amounted to 0.03 or 0.04. is either similar or higher than in imports. (Figure 1) Figure 1. Intra-industry trade between BiH and the EU 0.35 0,35 0.30 0,30 0.30 0.26 0.26 0.25 0,25 0.24 0.25 0.21 0.23 VIITl 0.21 0.20 0.20 0,20 VIITh 0.15 0.18 0.15 0,15 HIIT Total IIT 0.10 0,10 0.05 0,05 0.00 0,00 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Notes: 2008-2012 EU27; 2013-2018 EU28. Legend: VIITl – vertical IIT in low quality products; VIITh – vertical IIT in high quality products; 0,00 0,20 0,40 0,00 0,20 0,40 0,00 0,20 0,40 HIIT – horizontal IIT. Austria Source: Authors’ own calculation based on the BHAS data. Belgium Bulgaria SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 59 Croatia Cyprus Czech Republic Denmark Estonia Finland France Germany Greece Hungary Ireland Italy Latvia Lithuania Luxembourg Malta Netherlands Poland Portugal Romania Slovakia Slovenia Spain Sweden United Kingdom 2018 Total IIT Aver. 2008-2018 Total IIT 2008 Total IIT 2018 HIIT 2008 HIIT Aver. 2008-2018 HIIT 2018 VIIT 2008 VIIT Aver. 2008-2018 VIIT 0,35 0,30 0.30 0.26 0.26 0,25 0.24 0.25 0.21 0.23 VIITl 0.21 0.20 0,20 0.15 VIITh 0.18 0,15 HIIT Total IIT 0,10 0,05 0,00 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Figure 2. Intra-industry trade of BiH with the EU members (2008, 2018, average) 0.00 0.20 0.40 0.00 0.20 0.40 0.00 0.20 0.40 0,00 0,20 0,40 0,00 0,20 0,40 0,00 0,20 0,40 Austria Belgium Bulgaria Croatia Cyprus Czech Republic Denmark Estonia Finland France Germany Greece Hungary Ireland Italy Latvia Lithuania Luxembourg Malta Netherlands Poland Portugal Romania Slovakia Slovenia Spain Sweden United Kingdom 2018 Total IIT Aver. 2008-2018 Total IIT 2008 Total IIT 2018 HIIT Aver. 2008-2018 HIIT 2008 HIIT 2008 VIIT 2018 VIIT Aver. 2008-2018 VIIT Legend: HIIT – horizontal IIT; VIIT – vertical IIT. Source: Authors’ own calculation based on the BHAS data. In its trade with most EU member states, BiH ex- In the structure of BiH intra-industry agri-food perienced a higher intensity of IIT at the end of the trade with all EU members a vertical component sub- observed period compared to its beginning (Figure stantially prevailed, except with Italy in the period 2). However, on the above mentioned “top five” list, 2008-2011, when the dominance of HIIT was regis- the continuously increasing IIT trend was registered tered. The share of VIIT also increased in the trade with only in trading with Croatia. Greater oscillations in IIT most of the EU member states until the end of the ob- trend were observed for Austria, Belgium, the Czech served period. Republic and France. 60 SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 ANALYSIS OF INTRA-INDUSTRY TRADE IN AGRI-FOOD PRODUCTS BETWEEN BOSNIA AND HERZEGOVINA AND THE EUROPEAN UNION 4.2. Model estimation results results. However, all three specifications indicate simi- lar effects of the considered determinants. We estimated our baseline model represented by The common history of the countries has the equation (3) using PPML estimator. The estimation strongest positive effects on IIT of all the considered results for the whole sample are presented in Table factors. This reflects an intensive IIT between BiH on 3, column (1). We also checked the robustness of the one side, and Croatia and Slovenia on the other, which obtained results by estimating the baseline model on could be a result of their participation in the common subsamples for periods 2010-2018 and 2008-2016. market of the former Yugoslavia. Another variable The results of these robustness checks are presented with a robust positive effect on IIT is the economic size in Table 3, columns (2) and (3), respectively. of trade partners. This result is in line with the findings In all presented specifications Ramsey Regression of Jámbor (2014) in terms of both the intensity and Equation Specification Error test indicates that the significance of the effects. Sharing a common bor - specification of the conditional expectation is correct, der could positively affect IIT intensity, although this providing no evidence of misspecification. The model effect was found to be significant only in robustness appears to fit the data value, as evidenced by the high check, but not in the baseline specification. R values. As the Bayesian information criterion has On the other hand, differences in agricultural the lowest value for specification (1), it is considered productivity are a major determinant that signifi- a preferred specification for the interpretation of the cantly negatively and robustly affects IIT. Such a result Table 3. Estimation results Variable / model IIT (1) IIT (2) IIT (3) ASIZE 0.002* 0.002* 0.002** jt (0.001) (0.001) (0.001) ΔDGDPC -0.027 -0.037 -0.016 jt (0.028) (0.028) (0.026) DIST -0.590*** -0.596* -0.348 (0.067) (0.358) (0.331) BORDER 0.566 0.508** 0.613** (0.347) (0.255) (0.264) ΔTI 0.136 -0.492 1.506 jt (2.906) (2.299) (2.158) DPROD -0.030** -0.029** -0.031** jt (0.013) (0.013) (0.013) COMMON 1.935*** 1.902*** 2.296*** (0.389) (0.409) (0.398) ΔDFDI -0.182 -0.27 -0.355 jt (0.497) (0.560) (0.630) Constant -3.073*** -2.991*** -3.513*** (0.135) (0.433) (0.413) Observations 249 224 224 R 0.830 0.825 0.848 BIC 169.206 283.069 286.678 Log-likelihood -48.74 -44.125 -43.223 RESET test 0.1320 0.0823 0.4133 Note: Standard errors are provided in the brackets. ***, **, and * denote coefficients statistically significant at the 1%, 5% and 10% levels. BIC refers to the Bayesian information criterion and the RESET test denotes the p-values of the Ramsey Regression Equation Specification Error test. Fixed individual and time effects are estimated but not reported. Source: Authors’ own calculation. SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 61 ANALYSIS OF INTRA-INDUSTRY TRADE IN AGRI-FOOD PRODUCTS BETWEEN BOSNIA AND HERZEGOVINA AND THE EUROPEAN UNION supports the previous findings reported in the re - impact of differences in productivity on IIT was found lated empirical literature. Geographic distance was to be significantly negative. Finally, the results of our also found to be negatively related to IIT. The effect study do not support the hypotheses related to the is, however, not robust in the case of the Specification significance of effects of trade intensity, as well as dif- (3). As for the differences in GDP p/c and FDI, as well ferences in GDP p/c and FDI. as for the trade intensity, no statistically significant ef- By providing a deeper insight into characteristics fects were found in any estimated specifications. and determinants of IIT, the paper contributes to a better understanding of the position of the agri-food sector of the EU potential candidate country in one of its most important foreign markets. It also provides 5. Conclusion useful information for policymakers. The empirical evidence presented in this pa- The significantly high share of inter-industry trade per confirms the  theoretical hypotheses explaining implies a higher possibility of increasing adjustment the  patterns of IIT of less advanced countries (and in costs for BiH agri-food sector, associated with further less differentiated products), and those explaining ef- liberalisation of trade with the EU. Contrary to the fects of country-specific determinants on IIT. smooth adjustment hypothesis (Balassa, 1966) related The analysis of characteristics of IIT in agri-food to a higher share of IIT, an increase in trade as a result trade between BiH and the EU points to several main of liberalisation might cause more transfer of pro- concluding observations. Firstly, a relative importance duction factors between expanding and contracting of intra-industry trade as opposed to inter-industry product lines and more temporary unemployment in trade was significantly lower in the observed period BiH agri-food sector. – inter-industry trade proved as a strongly dominant Therefore further association of BiH to the EU by form of trade specialisation in the agri-food trade of completing the free trade area, and by harmonizing BiH with the EU, viewed totally and bilaterally. Despite regulations in the field of trade in agri-food products, the increasing level of trade liberalisation between will require restructuring of BiH agri-food sector to the trading partners, an expected change in IIT pat- increase its competitiveness. With respect to our find- terns from dominant inter- to dominant intra-industry ings on IIT patterns and following Bojnec and Fertő trade did not happen. The time series analysis indi- (2016), it is recommended to focus on two specific cated that the intensity of IIT with the EU as a whole market niches that differ by income p/c. As price com- experienced a slight growth. However, the increasing petitiveness is more important for HIIT and low VIIT, trend seems to correspond to the latest EU enlarge- competing in prices enabled by increasing economy ment in 2013, when Croatia joined, rather than the of scale in production will attract more EU consum- growth in the level and scope of mutual trade liber- ers with lower income. On the other hand, in product alisation related to BiH obtaining the potential can- groups with dominant high VIIT, meaning higher qual- didate status. Trends in IIT intensity did not point to ity and higher value-added of BiH exports, as well as a significant positive development that would lead higher prices, efforts should focus on marketing pro - to greater convergence between agri-food sectors of motion and branding at micro-level (level of compa- the analysed trading partners. Finally, the analysis of nies) or local/regional level (level of agri-food clusters). IIT structure, i. e. the distinction between HIIT and VIIT, revealed that the agri-food trade of BiH with the EU Endnotes was dominated by its vertical component, referring to trade of different quality products. In trading with 1 After Croatia‘s accession to the European Union, the EU as a whole, the high VIIT was somewhat more the Western Balkans region includes only five eco - prominent than a low VIIT over the major part of the nomies: Albania, Bosnia and Herzegovina, North observed period. Macedonia, Serbia and Kosovo*. (*The name The IIT patterns were explained through the does not prejudge the status of Kosovo* and is analysis of impact of several country-specific charac - in line with the Resolution of the United Nations teristics. Our initial hypotheses are supported by the Security Council UNSC 1244 and the opinion of the estimates of the econometric model in the case of International Court of Justice on the Declaration on economies’ size, common border, and common his- Kosovo’s Independence of 2008). tory effects, all of which were found to positively af- 2 Empirical literature contains evidence that the share fect IIT. Our hypothesis regarding the negative effect of vertical IIT is larger than that of horizontal IIT in of distance is also supported by the presented empiri- the trade between countries at different levels of cal evidence, to a certain extent. As it is expected, the economic development (Mardas and Nikas 2008). 62 SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 ANALYSIS OF INTRA-INDUSTRY TRADE IN AGRI-FOOD PRODUCTS BETWEEN BOSNIA AND HERZEGOVINA AND THE EUROPEAN UNION References Fidrmuc, J. 2004. The Endogeneity of the optimum currency area criteria, intra-industry trade, and EMU enlargement. Abd-El-Rahman, K. S. 1986. Réexamen de la définition et de Contemporary Economic Policy 22 (1): 1-12. la mesure des échanges croisés de produits similaires Food and Agriculture Organization of the United Nations entre les nations (Review of the definition and meas- (FAO) 2020. FAOSTAT (database online). http://www.fao. urement of cross-trade in similar products between na- org/faostat/en/#data/RL (accessed 20 November 2020) tions). Revue Economique 37 (1): 89-116. Greenaway, D., Hine, R., and Milner, C. 1995. Vertical and hor- Agency for Statistics of Bosnia and Herzegovina (BHAS). izontal intra-industry trade: A cross industry analysis for Statistical database – Foreign trade statistics (Available the United Kingdom. The Economic Journal 105 (433): on request only) 1505-1518. Ambroziak, Ł. 2012. FDI and intra-industry trade: Theory Grubel, H. G., and Lloyd, P. J. 1975. Intra-Industry Trade: and empirical evidence from the Visegrad countries. The Theory and Measurement of International Trade in International Journal of Economics and Business Differentiated Products. London: The Macmillan Press. Research 4 (1/2): 180–198. Heckman, J. J. 1979. Sample selection bias as a specification Balassa, B. 1966. Tariff reductions and trade in manufactures error. Econometrica 47 (1): 153-161. among the industrialised countries. American Economic Helpman, E. 1981. International trade in the presence of Review 56 (3): 466-473. product differentiation, economies of scale and mo - Balassa, B. 1986. Intra-industry trade among exporters of nopolistic competition: A Chamberlin-Heckscher-Ohlin manufactured goods. In Imperfect competition and in- approach. Journal of International Economics 11 (3): ternational trade, edited by D. Greenaway and P. K. M. 305-340. Tharakan, 108-128. Brighton: Wheatshef. Helpman, E., and Krugman, P. R. 1985. Market structure and Balogh, J. M., and Leitão, N. C. 2019. A gravity approach foreign trade. 8th edition. Cambridge, MA: MIT Press. of agricultural trade: The nexus of the EU and African, International Monetary Fund (IMF) 2019. World Economic Caribbean and Pacific countries. Agricultural Economics Outlook Database (database online). https://www.imf. 65 (11): 509-519. org/en/Publications/WEO/weo-database/2019/October Blanes, J. V., and Martín, C. 2000. The nature and causes of (accessed July 2020) intra-industry trade: Back to the comparative advantage Jámbor, A. 2013. Country-specific determinants of horizon- explanation? The case of Spain. Weltwirtschaftliches tal and vertical intra-industry agri-food trade: the case of Archiv 136 (3): 423–441. Bulgaria and Romania. Applied Studies in Agribusiness Bojnec, Š., and Fertő, I. 2016. Patterns and drivers of the and Commerce 7 (4-5): 69-76. agri-food intra-industry trade of European Union coun- Jámbor, A. 2014. Country specific determinants of horizon- tries. International Food and Agribusiness Management tal and vertical intra-industry agri-food trade: The case Review 19 (2): 53-73. of the EU new member states. Journal of Agricultural Burger, M., Van Oort, F., and Linders, G-J. 2009. On the speci- Economics 65 (3): 663-682. fication of the gravity model of trade: Zeros, excess zeros Jámbor, A. 2015. Country and industry-specific determi- and zero-inflated estimation. Spatial Economic Analysis nants of intra-industry trade in agri-food products in the 4 (2): 167-190. Visegrad countries. Studies in Agricultural Economics Centre d’études prospectives et d’informations internation- 117 (2): 93-101. als (CEPII) 2020. Distance database (database online). Jámbor, A., and Leitão N. C. 2016. Industry-specific deter - http://www.cepii.fr/anglaisgraph/bdd/distances.htm minants of vertical intra-industry trade: the case of EU (accessed August 2020) new member states’ agri-food sector. Post-Communist Dixit, A. K., and Stiglitz, J. E. 1977. Monopolistic competition Economies 28 (1): 34-48. and optimum product diversity. American Economic Jámbor, A., Bologh, J., and Kucsera, P. 2016. Country and in- Review 67 (3): 297-308. dustry specific determinants of intra-industry agri-food Falvey, R. E. 1981. Commercial policy and intra-industry trade in the Baltic countries. Agricultural Economics trade. Journal of International Economics 11 (4): 495-511. (Zemědělská Ekonomika) 62 (6): 280–291. Falvey, R. E., and Kierzkowski, H. 1987. Product quality, intra- Jing, W., Leitão, N. C., and Faustino, H. C. 2010. Intra-industry industry trade and (im)perfect competition. In Protection trade in agricultural products: The case of China. Instituto and competition in international trade, edited by H. Superior de Economia e Gestão. DE Working papers Kierzkowski, 143-161. Oxford (USA): Basil Blackwell. 6-2010/DE/SOCIUS Fertő, I., and Hubbard, L. J. 2002. Intra-industry trade in hori- Kandogan, Y. 2003. Intra-industry trade of transition coun- zontally and vertically differentiated agri-food products tries: Trends and determinants. Emerging Markets between Hungary and the EU. Hungarian Academy of Review 4 (3): 273-286. Sciences. Institute of Economics Discussion Paper 2002/2 SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 63 ANALYSIS OF INTRA-INDUSTRY TRADE IN AGRI-FOOD PRODUCTS BETWEEN BOSNIA AND HERZEGOVINA AND THE EUROPEAN UNION Krugman, P. R. 1980. Scale economies, product differen- sector of Bosnia and Herzegovina. European Journal of tiation, and the patterns of trade. American Economic Interdisciplinary Studies 3 (3): 154-163. Review 70 (5): 950-959. Silva, J. M. C., and Teneyro, S. 2006. The log of gravity. The Krugman, P. R., and Obstfeld, M. 2009. International econom- Review of Economics and Statistics 88 (4): 641-658. ics: theory and policy. 8th edition. Boston, MA: Pearson Trivić,, J., and Klimczak, L. 2015. The determinants of intra- Education Inc. regional trade in the Western Balkans. Proceedings of Lancaster, K. 1980. Intra-industry trade under perfect Rijeka Faculty of Economics: Journal of Economics and monopolistic competition. Journal of International Business 33 (1): 37-66. Economics 10 (2): 151-175. United Nations Conference on Trade and Development Łapinska, J. 2014. Determinants of intra-industry trade in (UNCTAD) 2021. Statistical database: Product groups agricultural and food products between Poland and EU and composition (SITC Rev. 3) (database online). countries. Danube: Law and Economics Review 5 (3): h ttps://unc tadsta t .unc tad.org/EN/Classifica tions/ 159-172. DimSit cR ev3P r oduc ts_DsibSpecialGr oupings_ Leitão, N. C. 2011. Intra-industry trade in the agriculture sec- Hierarchy.pdf (accessed March 2021) tor: The experience of United States. African Journal of United Nations Statistics Division (UNSTAT) 2021. Statistical Agricultural Research 6 (1): 186-190. database (database online). https://unstats.un.org/ Linder, S. B. 1961. An essay on trade and transformation. home/ (accessed August 2020) New York: John Wiley & Sons. Venables, A. J., Rice, P. G., and Stewart, M. 2003. The geogra- Little, R. J. 1988. A test of missing completely at random phy of intra-industry trade: Empirics. Topics in Economic for multivariate data with missing values. Journal of the Analysis and Policy 3 (1): 1-25. American Statistical Association 83 (404): 1198-1202. World Bank (WB) 2020. Agriculture value added per worker Maddala, G. S., and Wu, S. 1999. A comparative study of unit (database online). https://datacatalog.worldbank.org/ root tests with panel data and a new simple test. Oxford agriculture-value-added-worker-constant-2010-us (ac- Bulletin of Economics and Statistics 61 (S1): 631-652. cessed July 2020) Mardas D., and Nikas, Ch. (2008): Economic integration and World Bank (WB) 2020. World development indicators (da- intra-industry trade between the European Community tabase online). https://databank.worldbank.org/indica- and the Western Balkan countries. Transition Studies tor/NY.GDP.PCAP.CD/1ff4a498/Popular-Indicators# (ac - Review 15 (3): 511-523. cessed July 2020) Mrdalj, V., Ostojić, A., Vasko, Z., and Brković, D. 2017. Structure and determinants of intra-industry trade in agri-food 64 SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 ANALYSIS OF INTRA-INDUSTRY TRADE IN AGRI-FOOD PRODUCTS BETWEEN BOSNIA AND HERZEGOVINA AND THE EUROPEAN UNION APPENDIX Table 1A. Values of GL index by product groups in BiH agri-good trade with the EU SITC 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Average 061 0.95 0.96 0.48 0.49 0.77 0.65 0.91 0.89 0.89 0.99 0.58 0.78 056 0.57 0.63 0.69 0.70 0.76 0.83 0.88 0.70 0.65 0.59 0.55 0.69 054 0.24 0.47 0.69 0.41 0.61 0.74 0.61 0.70 0.93 0.90 0.82 0.65 046 0.85 0.74 0.90 0.86 0.94 0.20 0.29 0.32 0.45 0.48 0.50 0.59 421 0.44 0.71 0.73 0.45 0.46 0.84 0.54 0.44 0.31 0.31 0.53 0.52 058 0.64 0.64 0.56 0.48 0.68 0.56 0.46 0.40 0.43 0.40 0.37 0.51 075 0.58 0.57 0.30 0.79 0.47 0.30 0.47 0.50 0.40 0.60 0.38 0.49 048 0.29 0.28 0.26 0.29 0.28 0.67 0.70 0.67 0.62 0.57 0.54 0.47 121 0.55 0.97 0.25 0.50 0.02 0.21 0.84 0.46 0.61 0.18 0.11 0.43 057 0.11 0.10 0.33 0.23 0.26 0.62 0.27 0.40 0.35 0.44 0.40 0.32 111 0.01 0.11 0.12 0.16 0.07 0.46 0.43 0.41 0.47 0.58 0.60 0.31 071 0.29 0.28 0.31 0.30 0.16 0.19 0.19 0.23 0.21 0.22 0.25 0.24 431 0.10 0.27 0.37 0.27 0.34 0.06 0.10 0.11 0.18 0.32 0.34 0.22 059 0.15 0.22 0.25 0.05 0.23 0.17 0.21 0.14 0.28 0.24 0.25 0.20 112 0.13 0.14 0.17 0.18 0.34 0.16 0.13 0.13 0.16 0.20 0.23 0.18 081 0.06 0.11 0.12 0.09 0.18 0.16 0.17 0.15 0.15 0.22 0.22 0.15 044 0.00 0.26 0.12 0.27 0.03 0.03 0.14 0.07 0.14 0.16 0.37 0.15 072 0.00 0.05 0.00 0.00 0.11 0.10 0.14 0.00 0.01 0.43 0.75 0.14 098 0.16 0.13 0.13 0.10 0.06 0.17 0.19 0.14 0.14 0.14 0.12 0.13 012 0.01 0.00 0.01 0.24 0.15 0.18 0.21 0.19 0.32 0.01 0.00 0.12 091 0.05 0.01 0.00 0.00 0.01 0.06 0.17 0.28 0.13 0.13 0.27 0.10 022 0.00 0.00 0.00 0.00 0.01 0.38 0.00 0.02 0.11 0.35 0.19 0.10 047 0.03 0.04 0.04 0.04 0.08 0.14 0.12 0.17 0.11 0.06 0.08 0.08 223 0.00 0.00 0.00 0.04 0.01 0.09 0.12 0.01 0.13 0.21 0.28 0.08 074 0.35 0.21 0.05 0.00 0.01 0.00 0.00 0.01 0.04 0.11 0.02 0.07 073 0.00 0.17 0.03 0.04 0.03 0.09 0.10 0.08 0.07 0.06 0.05 0.07 041 0.00 0.00 0.00 0.00 0.00 0.07 0.00 0.00 0.00 0.00 0.43 0.05 122 0.00 0.01 0.06 0.13 0.07 0.02 0.01 0.02 0.03 0.04 0.11 0.04 024 0.00 0.00 0.00 0.00 0.00 0.17 0.00 0.02 0.06 0.07 0.08 0.04 017 0.16 0.00 0.00 0.00 0.00 0.19 0.00 0.00 0.00 0.00 0.00 0.03 025 0.02 0.00 0.00 0.00 0.00 0.32 0.01 0.00 0.00 0.00 0.00 0.03 222 0.02 0.01 0.01 0.06 0.02 0.01 0.06 0.01 0.00 0.02 0.01 0.02 001 0.02 0.01 0.02 0.02 0.03 0.02 0.02 0.03 0.03 0.01 0.00 0.02 062 0.01 0.01 0.01 0.01 0.01 0.01 0.02 0.02 0.03 0.07 0.01 0.02 023 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.03 0.05 0.05 0.01 045 0.00 0.00 0.00 0.02 0.00 0.00 0.00 0.00 0.01 0.01 0.00 0.00 016 0.00 0.00 0.00 0.00 0.00 0.04 0.00 0.00 0.00 0.00 0.00 0.00 422 0.00 0.00 0.00 0.00 0.00 0.01 0.02 0.00 0.00 0.00 0.00 0.00 043 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.02 0.00 042 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 011 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 411 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Note: The period 2008-2012 refers to the EU27. The period 2013-2018 refers to the EU28. Source: Authors‘ own calculation. SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 65 ANALYSIS OF INTRA-INDUSTRY TRADE IN AGRI-FOOD PRODUCTS BETWEEN BOSNIA AND HERZEGOVINA AND THE EUROPEAN UNION Table 2A. Vertical IIT in BiH agri-food trade with the EU member countries 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Average Austria 0.08 0.09 0.09 0.08 0.13 0.13 0.16 0.14 0.16 0.17 0.14 0.12 Belgium 0.00 0.02 0.02 0.00 0.04 0.09 0.14 0.03 0.14 0.07 0.09 0.06 Bulgaria 0.00 0.01 0.00 0.01 0.00 0.01 0.01 0.03 0.04 0.05 0.09 0.02 Croatia 0.24 0.28 0.29 0.28 0.26 0.31 0.27 0.28 0.33 0.34 0.32 0.29 Cyprus 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.15 0.00 0.00 0.00 0.01 Czech Republic 0.15 0.03 0.01 0.07 0.01 0.11 0.09 0.04 0.06 0.12 0.21 0.08 Denmark 0.00 0.00 0.00 0.01 0.00 0.00 0.03 0.02 0.01 0.02 0.03 0.01 Estonia 0.00 0.00 0.00 0.00 0.00  NT 0.00 0.00 0.00 0.00 0.00 0.00 Finland 0.00 0.00 0.00 0.13 0.19 0.00 0.00 0.00 0.02 0.00 0.00 0.03 France 0.05 0.09 0.09 0.08 0.16 0.12 0.14 0.07 0.29 0.17 0.13 0.13 Germany 0.05 0.04 0.05 0.05 0.06 0.05 0.10 0.11 0.10 0.11 0.11 0.08 Greece 0.04 0.01 0.00 0.00 0.00 0.02 0.02 0.05 0.03 0.03 0.04 0.02 Hungary 0.01 0.01 0.01 0.04 0.04 0.02 0.03 0.04 0.07 0.05 0.11 0.04 Ireland 0.00 0.00 0.00 0.01 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 Italy 0.16 0.18 0.22 0.14 0.39 0.37 0.32 0.29 0.23 0.26 0.19 0.25 Latvia  NT 0.00 0.00 0.00  NT  NT 0.00 0.00 0.00 0.00 0.00 0.00 Lithuania 0.00 0.00 0.00 NT  0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Luxembourg 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Malta  NT 0.00  NT  NT NT   NT 0.00 0.00 0.00 0.00 0.00 0.00 Netherlands 0.02 0.04 0.04 0.03 0.00 0.02 0.02 0.02 0.02 0.03 0.04 0.03 Poland 0.00 0.00 0.00 0.01 0.01 0.00 0.01 0.01 0.01 0.02 0.02 0.01 Portugal 0.04 0.01 0.03 0.06 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.01 Romania 0.00 0.00 0.00 0.00 0.00 0.00 0.07 0.06 0.16 0.03 0.08 0.04 Slovakia 0.00 0.07 0.02 0.00 0.01 0.00 0.01 0.09 0.01 0.02 0.03 0.02 Slovenia 0.15 0.18 0.17 0.13 0.15 0.15 0.16 0.18 0.19 0.16 0.16 0.16 Spain 0.00 0.01 0.00 0.00 0.03 0.04 0.02 0.01 0.00 0.01 0.00 0.01 Sweden 0.07 0.13 0.10 0.10 0.10 0.02 0.05 0.03 0.02 0.08 0.10 0.07 United 0.02 0.01 0.03 0.02 0.02 0.03 0.04 0.02 0.04 0.04 0.05 0.03 Kingdom Legend: NT – no agri-food trade between BiH and the EU member country in the given year. Source: Authors‘ own calculation. 66 SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 ANALYSIS OF INTRA-INDUSTRY TRADE IN AGRI-FOOD PRODUCTS BETWEEN BOSNIA AND HERZEGOVINA AND THE EUROPEAN UNION Table 3A. Horizontal IIT in BiH agri-food trade with the EU member countries 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Average Austria 0.01 0.10 0.00 0.01 0.05 0.07 0.01 0.04 0.00 0.01 0.00 0.03 Belgium 0.00 0.00 0.00 0.00 0.04 0.02 0.00 0.01 0.00 0.00 0.00 0.01 Bulgaria 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.02 0.00 Croatia 0.10 0.09 0.06 0.08 0.10 0.03 0.05 0.00 0.00 0.03 0.06 0.05 Cyprus 0.00 0.11 0.00 0.00 0.00 0.20 0.00 0.00 0.00 0.00 0.00 0.03 Czech Republic 0.06 0.00 0.00 0.00 0.00 0.00 0.00 0.04 0.07 0.00 0.00 0.02 Denmark 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Estonia 0.00 0.00 0.00 0.00 0.00 NT 0.00 0.00 0.00 0.00 0.00 0.00 Finland 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 France 0.01 0.00 0.01 0.00 0.01 0.00 0.00 0.01 0.01 0.03 0.05 0.01 Germany 0.00 0.01 0.00 0.00 0.01 0.02 0.00 0.01 0.01 0.00 0.00 0.01 Greece 0.00 0.04 0.02 0.00 0.00 0.05 0.00 0.00 0.00 0.00 0.00 0.01 Hungary 0.00 0.02 0.01 0.00 0.00 0.02 0.00 0.00 0.00 0.00 0.00 0.00 Ireland 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Italy 0.20 0.21 0.18 0.20 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.07 Latvia NT 0.00 0.00 0.00 NT NT 0.00 0.00 0.00 0.00 0.00 0.00 Lithuania 0.00 0.00 0.00 NT 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Luxembourg 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Malta NT 0.00 NT NT NT NT 0.00 0.00 0.00 0.00 0.00 0.00 Netherlands 0.02 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.01 0.00 0.00 0.00 Poland 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 Portugal 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Romania 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.01 0.00 Slovakia 0.00 0.00 0.00 0.00 0.00 0.03 0.00 0.00 0.00 0.00 0.00 0.00 Slovenia 0.00 0.03 0.02 0.03 0.04 0.02 0.02 0.02 0.00 0.00 0.01 0.02 Spain 0.00 0.01 0.00 0.01 0.02 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Sweden 0.00 0.00 0.01 0.00 0.00 0.04 0.03 0.04 0.08 0.00 0.00 0.02 United 0.00 0.00 0.00 0.00 0.02 0.00 0.00 0.00 0.01 0.00 0.00 0.00 Kingdom Legend: NT – no agri-food trade between BiH and the EU member country in the given year. Source: Authors‘ own calculation. SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 67 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png South East European Journal of Economics and Business de Gruyter

Analysis of Intra-Industry Trade in Agri-Food Products Between Bosnia and Herzegovina and the European Union

Loading next page...
 
/lp/de-gruyter/analysis-of-intra-industry-trade-in-agri-food-products-between-bosnia-jn1gxhQFFJ
Publisher
de Gruyter
Copyright
© 2021 Snježana Brkić et al., published by Sciendo
ISSN
2233-1999
eISSN
2233-1999
DOI
10.2478/jeb-2021-0014
Publisher site
See Article on Publisher Site

Abstract

Although research of intra-industry trade (IIT ) has been intensive in the last several decades, the empirical lit- erature focusing on this phenomenon in the Western Balkans countries remains limited, especially in agricul- tural sector. Aiming to contribute to the existing literature, the paper analyses patterns and country-specific determinants of IIT in agri-food products between Bosnia and Herzegovina (BiH) and the European Union (EU) during the period of their mutual trade liberalisation (2008-2018). The analysis employs Grubel-Lloyd indices and Greenaway-Hine-Milner methodology for measurement of IIT, and applies an econometric panel data model using a Poisson Pseudo-maximum likelihood approach in order to estimate effects of IIT determi- nants. The findings suggest that intra-industry agri-food trade of BiH with the EU is of low intensity and main- ly of vertical nature, viewed totally and bilaterally. We found significant positive effects of trading countries’ Snježana Brkić, PhD (corresponding author) sizes, common border and history on IIT, and negative Associate Professor effects of the geographic distance and differences in School of Economics and Business agricultural productivity. University of Sarajevo E-mail: snjezana.brkic@efsa.unsa.ba Keywords: Intra-industry trade (IIT ), agri-food prod- Address: Trg oslobođenja – Alija Izetbegović 1 ucts, Bosnia and Herzegovina (BiH), the European 71 000 Sarajevo Union (EU), Poisson Pseudo-maximum likelihood Bosnia and Herzegovina ORCID: https://orcid.org/0000-0003-1004-9855 (PPML) approach JEL classification: F14; Q17; O52. Radovan Kastratović, PhD Teaching Assistant University of Belgrade Faculty of Economics Belgrade, Republic of Serbia 1. Introduction E-mail: radovan.kastratovic@ekof.bg.ac.rs For decades, intra-industry trade (IIT) as a phenom- ORCID: https://orcid.org/0000-0002-6138-906X enon mostly related to product differentiation, scale economy and imperfect competition, has been a sub- Mirela Abidović Salkica, MSc ject of intensive research on the example of industrial Senior Assistant products and developed countries. At the same time, Faculty for Technical Studies University of Travnik IIT in agricultural sector has been neglected, due to Travnik, Bosnia and Herzegovina different characteristics of the products and markets E-mail: mirelaabidovic@hotmail.com compared to industrial sector, such as resource inten- ORCID: https://orcid.org/0000-0002-5325-5600 sity, low product differentiation and assumption of Copyright © 2021 by the School of Economics and Business Sarajevo 53 ANALYSIS OF INTRA-INDUSTRY TRADE IN AGRI-FOOD PRODUCTS BETWEEN BOSNIA AND HERZEGOVINA AND THE EUROPEAN UNION perfect competition. IIT in agricultural products is still and less developed countries when trading with more insufficiently researched, particularly for developing developed countries. In that case, the trade pattern of countries. less developed countries is dominantly inter-industry, In order to fill a gap in the empirical literature on for both industrial and agri-food products, while IIT is IIT in agri-food sector of less advanced countries, es- mostly of the vertical type . pecially transition countries of the Western Balkans If we observe characteristics of the countries with (WB) , the paper focuses on the analysis of intra-indus- which a given country trades, those characteristics will try agri-food trade of Bosnia and Herzegovina (BiH). Of allow us to predict some features of its trade pattern. all the WB countries, BiH has the total IIT the largest According to traditional trade theories, it is likely that body of research. There is however the single study of inter-industry trade (international trade in products IIT in agri-food sector of BiH, which was conducted by of different industries, based on differences in com- Mrdalj et al. (2017), but without modelling of IIT de- parative advantages) will prevail in relations between termination. The research was based on the K-means countries at different levels of economic develop - cluster analysis, aiming at identifying different clusters ment. By contrast, an intensive intra-industry trade i.e. of agri-food product groups related to comparative international trade in products belonging to the same advantages, IIT level and the ratio between unit values sector/industry, is expected to occur between similar of exports and imports. countries (Krugman and Obstfeld 2009), and especial- Our study covers the country’s IIT with the ly in the form of a prominent horizontal IIT. European Union (EU) in the agri-food sector for the Namely, the structure of IIT is composed of two period 2008-2018 which is characterised by the mu- components – the horizontal component and the tual trade liberalisation within the framework of the vertical one, which differ from each other in terms Stabilisation and Association Agreement (SAA). The of types of product differentiation. Although IIT in WB countries mostly have the EU candidate status, ex- homogenous products can emerge as well (due to cept BiH which still is only a potential candidate coun- seasonal variations, border trade, etc.), the phenom- try and very slow on its path to the EU membership. enon is primarily related to more differentiated prod- It was expected that the institutionalisation of trade ucts. Horizontal differentiation is based on the actual relations between BiH and the EU would result in the or perceived differences in products’ characteristics growth of the overall trade and IIT in many sectors, which do not result in the systematic variation in pric- including the agri-food one, and would change IIT es. Vertical differentiation implies varieties of different structure as well. This research provides a more com- levels of quality and, consequently, of different prices. prehensive insight into the characteristics (intensity, There are two different approaches to modelling trend and composition) of the agri-food IIT of BiH with horizontal differentiation. The love-for-varieties mod- the EU, and possible changes. The paper also aims at el, which was developed by Dixit and Stiglitz (1977), is identification and evaluation of country-specific IIT based on the assumption that consumers will use as determinants by estimating an empirical model us- many varieties of the same product as possible if they ing a Poisson Pseudo-maximum likelihood method are available. Lancaster (1980) however developed (PPML). the core-attributes model, which starts from the as- The paper is structured as follows: The next sec- sumption that each consumer prefers one variety with tion provides a short theoretical framework for the a special combination of characteristics. The theo- analysis of IIT characteristics and econometric analy- retical basis of horizontal IIT (HIIT) was further devel- sis. The third section describes the used methodology oped by Krugman (1980), Helpman (1981), Helpman and data. Research findings on the basic patterns of and Krugman (1985), etc. Vertical IIT (VIIT) models, as IIT of BiH with the EU in agri-food sector and country- developed by Falvey (1981), Falvey and Kierzkowski specific IIT determinants are presented and discussed (1985), and others, include the redefined traditional in the fourth section. The last part contains conclud- concept of comparative advantages as explanation ing remarks, including policy implications. of IIT, starting from the assumption that differences in the product quality result from the differences in production processes i.e. in factor endowment and technology. 2. Conceptual framework The intensity of total IIT and its components is Our research is based on the hypothesis that the driven by different factors, both country-specific and pattern and determinants of IIT of BiH in agri-food industry-specific ones. With respect to IIT determi- products are consistent with theoretical assumptions nants, our analysis starts from the hypothesis that and up-to-date empirical findings on IIT of transition country-specific determinants, such as economic size, 54 SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 ANALYSIS OF INTRA-INDUSTRY TRADE IN AGRI-FOOD PRODUCTS BETWEEN BOSNIA AND HERZEGOVINA AND THE EUROPEAN UNION difference in the level of development, geographic (2003) believes that FDI is a determinant which pri- distance, trade intensity, foreign direct investment, marily and positively affects VIIT, which was confirmed common past etc., influence IIT in agri-food products for the agri-food sector in the research by Jámbor and the same way as in the case of industrial products. It Leitão (2016). means that determinants which relates to any kind Almost every model of bilateral trade, whether it of proximity (economic, geographic etc.) of trading is the total trade or one of its components such as IIT, countries have a positive effect on their mutual IIT, takes into account the geographic distance. The dis- while differences between trading countries have a tance indirectly “measures” the effects of transport, negative effect. transaction and information costs on trade. Balassa We hypothesised a significant and positive effect (1986) established that all types of trade decrease of trading countries’ economic size, measured by their with increasing geographic distance, and it particu- gross domestic products (GDP). Reasons can be found larly applies to IIT. The negative effect of geographic on both the demand and the supply side. According distance on the IIT share in agri-food products was to Lancaster (1980), greater average size of two coun- also pointed out by Jing, Leitão, and Faustino (2010), tries’ markets will result in the greater import demand Jámbor (2015), Balogh and Leitão (2019), etc. On the for differentiated products, and more prominent IIT. other hand, geographically close, especially adjacent Krugman (1980) argues that a larger market creates countries typically have similar economic, cultural and a greater possibility for achieving economy of scale, other characteristics. In economic terms, it pertains which in turn leads to greater IIT. A positive impact to the similarity of the product structure and the de- of economic size on IIT was confirmed by research in mand structure between them (Venables, Rice, and agri-food sector as well, such as in Jing, Leitão, and Stewart 2003). Therefore, as opposed to the deter- Faustino (2010), Jámbor (2013), Bojnec and Fertő minant “geographic distance“, it is expected that the (2016), etc. determinant “common border“ has a significant and Difference in the level of economic development, positive impact on IIT. mostly expressed as difference in income distribution, Trade intensity is taken as the approximation of as an IIT determinant is rooted both in the theories the geographic trade orientation, indicating how im- of increasing returns and in the H-O theory. On the portant a trading partner is to the observed country. demand side, differences in incomes per capita (p/c) The positive effect of trade intensity on IIT in agri-food reflect differences in the structure of demand (Linder products was established in papers by Fidrmuc (2004) 1961, Lancaster 1980), while, on the supply side, they and Łapinska (2014). represent differences in factor endowment (Helpman Historical and political factors such as colonial his- 1981). The negative effect of the difference in incomes tory or a common state in the past are significant for p/c on the IIT was established in a number of stud- trade ties in general, and for IIT in particular. Countries ies, and, on the example of the agri-food products, in with a common past usually share some economic, research by Jing, Leitão and Faustino (2010), Leitão political and cultural characteristics that could inten- (2011), etc. The effect of the difference in incomes p/c sify IIT between them. Effects of political and histori- depends on the type of IIT, e.g. the difference typically cal factors on the example of intra-regional trade in has a positive effect on VIIT. the South East Europe were investigated by Trivić and In a similar way, Jámbor (2015) established the Klimczak (2015). negative effect of difference in productivity of trad- ing countries on their mutual IIT in agri-food products. However, the effect of difference in productivity could 3. Applied methodology vary depending on IIT type. Intensive VIIT occurs more 3.1. Measurement of IIT frequently among countries that are different in terms of productivity. Measurement of IIT mostly refers to calculating the IIT Difference in other economic performances such share or intensity in a given industry trade or trade of as success in attracting foreign direct investment (FDI) all industries of a country, using Grubel-Loyd indices. also affects IIT. However, in general, FDI effect on IIT Standard Grubel-Lloyd index (Grubel and Lloyd 1975) can be ambivalent. Some studies in agri-food sector is calculated as follows: such as those by Jámbor (2015), and Jámbor, Bologh �𝑋𝑋 � 𝐶𝐶 ���𝑋𝑋 � 𝐶𝐶 � �� �� �� �� and Kucsera (2016) determined a negative impact of 𝐺𝐺𝐺𝐺 �   �� 𝑋𝑋 � 𝐶𝐶 �� �� FDI inflow on IIT, while Leitão (2011) found a positive effect. Ambroziak (2012) and Jámbor (2013) found � � 𝐺𝐺𝐺𝐺 � � (1) that FDI can increase both HIIT and VIIT. Kandogan 𝐺𝐺𝐺𝐺 � � 𝐺𝐺𝐺𝐺 𝑤𝑤 �� �� SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 55 ��� � � 𝐺𝐺𝐺𝐺 � � (2) �� �� � �� �� 𝑤𝑤 � (3) �� � �� �� � �� �� ��� �� �� � (4) �� � �� �� � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� �� � � � � � � � ���� � � ���� � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� ��� � � ���� (5) � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � ���� � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� ��� � � ���� (6) � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � ���� � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� ��� � � ���� ��� � ��� ��� � ��� ��� ∑ ∑ �𝑋𝑋 � 𝐶𝐶 � � �𝑋𝑋 � 𝐶𝐶 � ��� ��� �� �� �� �� ��� 𝑉𝑉𝑉𝑉𝑉𝑉 � �𝑋𝑋 � 𝐶𝐶 � �� �� ��� ��� � � 𝑉𝑉𝑉𝑉𝑉𝑉 � � ��� � � � � � 𝑉𝑉𝑉𝑉𝑉𝑉 � ��𝐷𝐷𝑉𝑉�𝐵𝐵 ∆ 𝐺𝐺𝐷𝐷 𝐷𝐷𝑉𝑉𝐷𝐷𝑉𝑉 �� �� � �� � � � � � � � ∆ 𝑉𝑉𝑉𝑉 𝑅𝑅 ∆ 𝑉𝑉 � � �� �� �� ����� 𝐵𝐵𝐵𝐵 𝑅𝑅 �� 𝐷𝐷𝐷𝐷𝐶𝐶 𝐶𝐶 �� � � �� (9) � � � � � � �� 𝐶𝐶𝐶𝐶 𝐷𝐷𝑅𝑅𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷 𝑅𝑅𝑅𝑅𝑅𝑅 �𝑋𝑋 � 𝐶𝐶 ���𝑋𝑋 � 𝐶𝐶 � �� �� �� �� 𝐺𝐺𝐺𝐺 �   �� 𝑋𝑋 � 𝐶𝐶 �� �� �𝑋𝑋 � 𝐶𝐶 ���𝑋𝑋 � 𝐶𝐶 � � � 𝐺𝐺𝐺𝐺 � � (1) �� �� �� �� 𝐺𝐺𝐺𝐺 �   �� 𝑋𝑋 � 𝐶𝐶 �� �� 𝐺𝐺𝐺𝐺 � � 𝐺𝐺𝐺𝐺 𝑤𝑤 �� �� � � 𝐺𝐺𝐺𝐺 � � (1) ��� � � 𝐺𝐺𝐺𝐺 � � (2) 𝐺𝐺𝐺𝐺 � � 𝐺𝐺𝐺𝐺 𝑤𝑤 �� �� ��� GLij – standard Grubel-Lloyd index (IIT share) for the Trade in products whose unit values are beyond �� �� � �� �� � � 𝐺𝐺𝐺𝐺 � � (2) 𝑤𝑤 � (3) given industry i in trade between the given country this interval is iden �� tified as v � ertical IIT. Blanes and ∑ �� �� � �� �� ��� and another country/country group j; Xij – exports Martın (2000) divide VIIT into two categories – high of industry i from the given country to another coun- VIIT (VIITh) and low VIIT (VIITl) by using also the rela- �� �� � �� �� 𝑤𝑤 � (3) �� try/country group j; Mij – imports of industry i of the tive unit value: VIIT �� h (RUV abo � ve 1.15) signifies trade �� ∑ �� �� � �� �� ��� � (4) �� � �� given country from another country/country group j; in vertically differentiated products of higher quality �� i = 1, … , n – the number of industries. (meaning that quality of expor ts is higher than that of � � � �� Value 0 indicates purely inter-industry trade, while imports), while 𝑅𝑅𝑅𝑅 VIITl (RUV below 0.85) sig 𝑅𝑅𝑅𝑅 nifies trade �� �� �� � (4) �� � � � � �� � � � ���� � � � ���� � �� � index value 1 means that the entire trade is of intra- in vertically differentiated products of lower quality �� 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� industry type. For most product groups, the value of (meaning that quality of expor ts is lower than that of � � ��� � � ���� (5) G-L index is in interval between two extreme values imports). 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� �� � � � � � � � ���� � � ���� 0 and 1. � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 � � �� �� �𝑋𝑋 � 𝐶𝐶 ���𝑋𝑋 � 𝐶𝐶 � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� Grubel-Lloyd inde �� x can be calcula �� �� ted as an agg �� re- 𝐺𝐺𝐺𝐺 �   𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � ���� �� � � � ��� � � ���� (5) �𝑋𝑋 � 𝐶𝐶 ���𝑋𝑋 � 𝐶𝐶 � gate index (a trade-weigh 𝑋𝑋 ted a � 𝐶𝐶 verage of the industry 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� �� �� �� �� �� �� 𝐺𝐺𝐺𝐺 �   �� indices) for a country’s individual trading partner or a 𝑋𝑋 � 𝐶𝐶 �� �� � � ��� � � ���� (6) 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 � � 𝐺𝐺𝐺𝐺 � � (1) group of a country’s trade partners, as follows: �� �� 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � ���� � � � � 𝐺𝐺𝐺𝐺 � � (1) 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 � � � �� �� 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � ���� � � � � ��� � � ���� (6) 𝐺𝐺𝐺𝐺 � � 𝐺𝐺𝐺𝐺 𝑤𝑤 �� �� 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� ��� 𝐺𝐺𝐺𝐺 � � 𝐺𝐺𝐺𝐺 𝑤𝑤 �� �� � � ��� � � ���� ��� 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 ��� �� �� 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � ���� � � 𝐺𝐺𝐺𝐺 � � (2) � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 For the purposes of calculating aggregate HIIT and �� �� � � 𝐺𝐺𝐺𝐺 � � (2) ��� ��� ��� ��� � � where VIIT according to GHM method, we applied the index ∑ �𝑋𝑋 � 𝐶𝐶 � � ∑ �𝑋𝑋 � 𝐶𝐶 � ��� � � ���� ��� ��� �� �� ��� �� �� ��� 𝑉𝑉𝑉𝑉𝑉𝑉 � in the following form: �𝑋𝑋 � �� 𝐶𝐶 �� ��� � 𝑋𝑋 � 𝐶𝐶 � �� �� � �� �� �� �� � �𝑋𝑋 � 𝐶𝐶 � 𝑤𝑤 � (3) 𝐺𝐺𝐺𝐺 �   �� �� �� � ��� �� ∑ �� �� � �� �� �� ��� ��� �𝑋𝑋� �� 𝐶𝐶 � �� �� 𝑤𝑤 � (3) �� � � ��� ��� � ��� ��� �� �� � ∑ ��� ∑ �� �� �𝑋𝑋 � 𝐶𝐶 � � �𝑋𝑋 � 𝐶𝐶 � ��� ��� ��� �� �� �� �� ��� � � 𝑉𝑉𝑉𝑉𝑉𝑉 � � ��� 𝑉𝑉𝑉𝑉𝑉𝑉 � � � 𝐺𝐺𝐺𝐺 � � (1) � � ∑ �𝑋𝑋 � 𝐶𝐶 � �� �� �� ��� �� w – a share of industry i in total trade of the given ij � � � (4) � � � � �� � � � 𝑉𝑉𝑉𝑉𝑉𝑉 � ��𝐷𝐷𝑉𝑉�𝐵𝐵 ∆ 𝐺𝐺𝐷𝐷 𝐷𝐷𝑉𝑉𝐷𝐷𝑉𝑉 �� �� �� �� � �� �� � country with the country/coun �� try group j. ��� � (4) � � 𝑉𝑉𝑉𝑉𝑉𝑉 � � ��� �� � 𝐺𝐺𝐺𝐺 � � 𝐺𝐺𝐺𝐺 �� 𝑤𝑤 The common feature of most �� �� �� methods for empiri- � � � � � � ∆ 𝑉𝑉𝑉𝑉 𝑅𝑅 ∆ 𝑉𝑉 � � � � �� �� �� 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 ��� cal division of IIT into its horizontal and vertical com- where X and M stand for industry exports or im- �� �� � � � � � �� � � � � � � � ���� � � ���� � � 𝑉𝑉𝑉𝑉𝑉𝑉 � ��𝐷𝐷𝑉𝑉�𝐵𝐵 ∆ 𝐺𝐺𝐷𝐷 𝐷𝐷𝑉𝑉𝐷𝐷𝑉𝑉 � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� � ponents is their basis on the differences in quality, i.e. ports respectively; H, V distinguishes �� HIIT fr � om VIIT, 𝑅𝑅𝑅𝑅 �� 𝑅𝑅𝑅𝑅 �� ����� 𝐵𝐵𝐵𝐵 𝑅𝑅 �� 𝐷𝐷𝐷𝐷𝐶𝐶 𝐶𝐶 �� � � �� (9) � � � � � � �� �� �� �� � � � � � � � ���� � � ���� � � 𝐺𝐺𝐺𝐺 � � (2) � � on the use of the unit value as a quality indicator. We i represents an industr � y and � j represen � ts a trading � � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� ∆ 𝑉𝑉𝑉𝑉 𝑅𝑅 ∆ 𝑉𝑉 ��� � � ���� (5) � � �𝑋𝑋 � 𝐶𝐶 ���𝑋𝑋 � 𝐶𝐶 � �� �� �� �� �� �� �� used Greenaway, Hine, and Milner (1995) approach partner of the given country (i,j = 1,…, n). 𝐺𝐺𝐺𝐺 �   �� ��� � � ���� (5) 𝑋𝑋 � 𝐶𝐶 (GHM methodology) based on the pioneering paper �� �� ����� 𝐵𝐵𝐵𝐵 𝑅𝑅 �� 𝐷𝐷𝐷𝐷𝐶𝐶 𝐶𝐶 �� � � �� (9) � � �� �� � � � � � � � �� �� �� 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� 𝑤𝑤 � (3) by Abd-El-Rahman �� (1986), � or more precisely on his ∑ �� �� � 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � ���� � �� �� � � ��� � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 � � 𝐺𝐺𝐺𝐺 � � (1) 𝑅𝑅𝑅𝑅 �� �� relative unit value index (RUV � ): �� �� 3.2. Sample and econometric specification 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � ���� � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� ��� � � ���� (6) �� We considered the decomposition of the dependent �� 𝐺𝐺𝐺𝐺 ��� � � � 𝐺𝐺𝐺𝐺 � ���� (6) 𝑤𝑤 � (4) �� �� �� � �� � variable into its horizontal and vertical components. �� 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� ��� �� However, the preliminary analysis revealed that the 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � � � ���� � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� 𝑅𝑅𝑅𝑅 �� � � �� �� RUV – ratio between exports and imports unit value vertical IIT strongly dominated the agri-food trade 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � ���� ij � �� � � 𝐺𝐺𝐺𝐺 � � (2) �� � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� � � � � � � � ���� � � ���� �� �� for industry i in tr � ade of the given country with c � oun- flows between BiH and the EU. For this reason, we fo - ��� � � ���� ��� 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� try/country group j; UV – unit value of exports for cused on examining the determinants of total IIT. ij ��� � � ���� ��� ��� � � ���� (5) �� �� � industry i; UV – unit value of imports for industry i. In order to investigate how various determinants �� �� ij 𝑤𝑤 � (3) ��� ��� ��� ��� � �� � � ∑ �� �� � Horizon ∑tal IIT e �𝑋𝑋 xists when e � 𝐶𝐶 � � xpor ∑ t unit v �𝑋𝑋 � 𝐶𝐶 alues ar �e affect the intra-industry agri-food trade of BiH with the �� �� ��� ��� �� �� ��� �� �� ��� � � ��� ��� ��� ��� 𝑉𝑉𝑉𝑉𝑉𝑉 � � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 rela �tively close to import unit values of the given EU member states we estimated an econometric mod- ∑ �𝑋𝑋 �� � 𝐶𝐶 � � � ∑ ���𝑋𝑋 � 𝐶𝐶 � ��� �� �� ��� �� �� ∑ �𝑋𝑋 � 𝐶𝐶 � ��� 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � �� �� � ���� ��� � � 𝑉𝑉𝑉𝑉𝑉𝑉 � produc � t, i.e. when the r 𝑅𝑅𝑅𝑅 atio between these unit v 𝑅𝑅𝑅𝑅 alues el, using the data on bilateral trade between BiH and �� �� ∑ �𝑋𝑋 � � 𝐶𝐶 � �� �� ��� �� �� ��� is within the ±0.15 or ±0.25 interval, depending on the 28 member states and providing a panel of 308 ob- � (4) � � 𝑉𝑉𝑉𝑉𝑉𝑉 � � ��� �� � ��� � � � ���� �� (6) �� selected dispersion factor (par ��� ameter α). servations. Due to data considerations, we adopted a � � 𝑉𝑉𝑉𝑉𝑉𝑉 � � ��� more aggregate approach in the analysis, using coun- � � � �� � � � 𝑉𝑉𝑉𝑉𝑉𝑉 � ��𝐷𝐷𝑉𝑉�𝐵𝐵 ∆ 𝐺𝐺𝐷𝐷 𝐷𝐷𝑉𝑉𝐷𝐷𝑉𝑉 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� � 𝑅𝑅𝑅𝑅 �� �� �� 𝑅𝑅𝑅𝑅 � �� �� tries as units of analysis. The IIT indices were calculat- � � 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � � ���� � �� � � � � � � � � ���� � � ���� � � 𝑉𝑉𝑉𝑉𝑉𝑉 � ��𝐷𝐷𝑉𝑉�𝐵𝐵 ∆ 𝐺𝐺𝐷𝐷 𝐷𝐷𝑉𝑉𝐷𝐷𝑉𝑉 � � �� �� � 𝑅𝑅𝑅𝑅 � � 𝑅𝑅𝑅𝑅 � �� � 𝑅𝑅𝑅𝑅 � � �𝑅𝑅𝑅𝑅 �� �� ed from the raw trade data provided by the Agency �� �� ∆ 𝑉𝑉𝑉𝑉 𝑅𝑅 ∆ 𝑉𝑉 � � �� �� �� � � � � � � for Statistics of BiH (BHAS), classified according to ∆ 𝑉𝑉𝑉𝑉 𝑅𝑅 ∆ 𝑉𝑉 ��� � � ���� ��� � ��� � � ���� (5) � �� �� �� ����� 𝐵𝐵𝐵𝐵 𝑅𝑅 �� 𝐷𝐷𝐷𝐷𝐶𝐶 𝐶𝐶 �� � � �� (9) � � � � � � �� ����� 𝐵𝐵𝐵𝐵 𝑅𝑅 �� 𝐷𝐷𝐷𝐷𝐶𝐶 𝐶𝐶 �� � � �� (9) � � � � � � � � �� 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� ��� ��� ��� ��� 56 � � SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � ���� ∑ �𝑋𝑋 � 𝐶𝐶 � � ∑ �𝑋𝑋 � 𝐶𝐶 � ��� ��� �� ��� � �� �� ��� 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 𝑉𝑉𝑉𝑉𝑉𝑉 � �� �� ∑ �𝑋𝑋 � 𝐶𝐶 � �� �� ��� ��� � � ���� (6) ��� � � 𝑉𝑉𝑉𝑉𝑉𝑉 � � ��� � � 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 �� �� 𝑅𝑅𝑉𝑉𝑉𝑉𝑉𝑉 : � � � � � � ���� � � � � � � 𝑅𝑅𝑅𝑅 � 𝑅𝑅𝑅𝑅 𝑉𝑉𝑉𝑉𝑉𝑉 � �� �𝐷𝐷𝑉𝑉� �𝐵𝐵 ∆ 𝐺𝐺𝐷𝐷 ��𝐷𝐷𝑉𝑉𝐷𝐷𝑉𝑉 �� �� � �� � ��� � � � ���� ��� � � � � � ∆ 𝑉𝑉𝑉𝑉 𝑅𝑅 ∆ 𝑉𝑉 � � �� �� �� ����� 𝐵𝐵𝐵𝐵 𝑅𝑅 �� 𝐷𝐷𝐷𝐷𝐶𝐶 𝐶𝐶 �� � � �� (9) � � � � � � �� � ��� ��� � ��� ��� ∑ ∑ �𝑋𝑋 � 𝐶𝐶 � � �𝑋𝑋 � 𝐶𝐶 � ��� ��� �� �� �� �� ��� 𝑉𝑉𝑉𝑉𝑉𝑉 � �𝑋𝑋 � 𝐶𝐶 � �� �� ��� ��� � � 𝑉𝑉𝑉𝑉𝑉𝑉 � � ��� � � � � � 𝑉𝑉𝑉𝑉𝑉𝑉 � ��𝐷𝐷𝑉𝑉�𝐵𝐵 ∆ 𝐺𝐺𝐷𝐷 𝐷𝐷𝑉𝑉𝐷𝐷𝑉𝑉 �� �� � �� � � � � � � � ∆ 𝑉𝑉𝑉𝑉 𝑅𝑅 ∆ 𝑉𝑉 � � �� �� �� ����� 𝐵𝐵𝐵𝐵 𝑅𝑅 �� 𝐷𝐷𝐷𝐷𝐶𝐶 𝐶𝐶 �� � � �� (9) � � � � � � �� 𝐶𝐶𝐶𝐶 𝐷𝐷𝑅𝑅𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷 𝐶𝐶𝐶𝐶 𝐷𝐷𝑅𝑅𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷 𝐶𝐶𝐶𝐶 𝐷𝐷𝑅𝑅𝐷𝐷 𝐶𝐶𝐶𝐶 𝐷𝐷𝑅𝑅𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷 𝑅𝑅𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅𝑅𝑅 𝐶𝐶𝐶𝐶 𝐷𝐷𝑅𝑅𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝐶𝐶𝐶𝐶 𝐷𝐷𝑅𝑅𝐷𝐷 𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝑅𝑅𝑅𝑅𝑅𝑅 𝐷𝐷𝐷𝐷 𝑅𝑅𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅𝑅𝑅 the Standard International Trade Classification (SITC) There is considerable variability between the Revision 3. Data were aggregated for 42 individual country pairs, particularly in terms of economic size, three-digit product groups associated with the agri- income differences, distance, and differences in agri- food sector, as defined by UNCTAD (2021). cultural productivity. The panel is unbalanced due to As there is no universally accepted form of speci- 10 missing values of IIT, which are caused by the ab- fying the IIT equation, we guided our econometric sence of trade between BiH on one side and Estonia, modelling approach by following the underlying Latvia, Lithuania and Malta on the other in some ob- theoretical model and by adopting the most com- served years. Using Little’s test (Little 1988) we de- monly included variables in the closely related empiri- termined that the aforementioned observations are cal studies. The descriptive statistics of the sample are missing completely at random. Additionally, we con- provided in Table 1, along with the definitions of the sidered using the random-effects variant of Heckman’s variables considered in the model, their definitions, sample selection model (Heckman 1979), to account data sources, and expected effect on IIT. for the fact that some observations are missing due to Table 1. Description and descriptive statistics of the variables Data Exp. Variable Variable description Mean Std. dev. Min Max source effect IIT The intensity of intra-industry agri-food BHAS N/A 0.068 0.097 0 0.401 jt trade measured as the value of GL index. ASIZE The economic size expressed as the aver- IMF + 325.652 477.445 12.963 1984.866 jt age of nominal GDPs of BiH and a coun- try j (in current billions EUR) in year t. DGDPC The inequality in income per capita IMF - 28.331 21.762 2.174 113.494 jt measured as the absolute difference in per capita GDP (in current thousands EUR) between BiH and a country j in year t. DIST The geographical distance as a direct CEPII - 1.157 0.572 0.29 2.363 straight-line distance in thousands kilo- meters between capital cities of BiH and a country j. BORDER The common border represented by CEPII + 0.036 0.186 0 1 dummy variable that equals 1 if an observed pair of countries has common border and 0 otherwise. TI The trade intensity as the share of a trad- BHAS + 0.019 0.041 0 0.271 jt ing partner’s market in the foreign trade of BiH in agri-food products in year t. DPROD The difference in productivity measured FAO - 28.528 22.458 0 99.331 jt as the absolute difference of the value- added per worker in agriculture in BiH and a country j in year t in thousands EUR. DFDI The difference in FDI measured as the UNCTAD - 0.283 0.369 .004 1.882 jt absolute difference between FDI stock in BiH and in a country j in year t in millions USD. COMMON The common history represented by CEPII + 0.107 0.31 0 1 dummy variable that equals 1 if an observed pair of countries has common state or colonial relationship in the past, and 0 otherwise. Source: Authors‘ own calculation. SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 57 ANALYSIS OF INTRA-INDUSTRY TRADE IN AGRI-FOOD PRODUCTS BETWEEN BOSNIA AND HERZEGOVINA AND THE EUROPEAN UNION the lack of trade between the countries. However, the The model we estimated has the following form: preliminary estimations of the model revealed that � � � � 𝐼𝐼𝐼𝐼𝐼𝐼 � ��𝐷𝐷𝐼𝐼�𝐵𝐵 ∆ 𝐷𝐷𝐼𝐼𝐷𝐷𝐼𝐼 the Inverse Mills ratio and the correlation between �� �� � �� � the error terms in the selection and primary equations � � � � � � ∆ 𝐼𝐼𝐼𝐼 ∆ 𝐼𝐼 are insignificant, implying the absence of sample se - � � �� �� �� lection bias. For this reason, we based our empirical ����� 𝐵𝐵𝐵𝐵 𝐷𝐷 �� 𝐷𝐷𝐷𝐷𝐶𝐶 𝐶𝐶 �� � � �� (9) � � � � � � �� approach on estimating a single augmented gravity- type equation, which is particularly useful for analys- ing the economic relationships on the bilateral level. where dependent and independent variables are de- The use of the gravity-type equation as a work- fined in the same manner as presented in Table 1, α horse model is the most commonly adopted ap- denotes a constant, β and δ denote slope coefficients, proach in the related literature (Jing, Leitão, and j is the index of the EU trade partners of BiH, t signifies Faustino 2010; Łapinska 2014; Jámbor, Bologh, and the time period, and μ , λ and ε refer to individual ef- j t jt Kucsera 2016). In nearly all studies the gravity equa- fect, time effect and the error term, respectively. tion is linearised using a log-transformation and then Our baseline model is a two-way panel data model, estimated using the ordinary least squares, gener- including individual and time-specific effects, which alised least squares, and generalised method of mo- were included in the model due to the panel structure ments. However, such an approach is inadequate in of the data and the joint statistical significance of the the cases where there is a significant share of zero val- aforesaid effects. The inclusion of individual and time- ues of dependent variable, as it leads to loss of obser- specific effects allowed us to control for unobserved vations and the dependency between the error term heterogeneity across the countries and time periods and covariates, resulting in inconsistent estimates and the impact of factors not explicitly included in the (Silva and Teneyro 2006). In our sample, 23.49% of ob- model, which reduced the risk of misspecification. Our servations contain zero dependent variable (which re- model is essentially an augmented gravity equation. flect instances of perfect inter-industry trade), making We use a static specification, as there is no theoretical the aforesaid problem a non-negligible concern. The justification for using the dynamic one in the context common solution to the problem of zeros in estimat- of IIT, and as it is the most widely used approach in the ing gravity-type equations is the application of PPML related empirical literature (Fertő and Hubbard 2002; method. This method, originally proposed by Silva Jing, Leitão, and Faustino 2010). In model estimation, and Teneyro (2006) allows the gravity equation to be we were primarily interested in the significance of β estimated in its original multiplicative form, accomo- and δ coefficients, which are directly related to our ini- dating to zero values of dependent variable (Burger, tial hypotheses of the impact of various determinants Van Oort, and Linders 2009). The estimator was shown on the IIT intensity. to be robust to heteroskedasticity if the conditional Finally, in order to reduce the risk of spurious re- variance of the dependent variable is proportional to sults, we evaluated the stationarity of the panels using its conditional mean. This assumption is not violated, a Fisher-type stationarity test (Maddala and Wu 1999). as there is only one process generating zero and miss- The test results for the continuous variables are pre- ing values of the dependent variable in our sample. sented in Table 2. Table 2. Panel unit root test results for continuous variables Variable Without time trend With time trend 2 2 Modified χ p-value Modified χ p-value IIT 1.6555 0.0489 3.4774 0.0003 jt ASIZEjt 8.8983 0.0000 2.8699 0.0021 ΔDGDPCjt 6.8023 0.0000 2.6353 0.0042 ΔTIjt 3.4924 0.0002 3.9437 0.0000 DPROD 4.4063 0.0000 5.2172 0.0000 jt ΔDFDI 10.4041 0.0000 7.5441 0.0000 jt Source: Authors’ own calculations based on Maddala and Wu (1999). 58 SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 𝐶𝐶𝐶𝐶 𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 In the initial testing, we found that DGDPC, TI, and The highest average GL indices in BiH agri-food DFDI contained unit roots in levels. However, the first trade with the EU, which are also the only indices with differences of the variables were found to be station- the value higher than 0.50, have been registered in ary. For this reason, the three variables are included in the following product groups: 061 Sugars, molasses our model as first-order differences. and honey (0.78); 056 Vegetables, roots and tubers (0.69); 054 Vegetables, fresh, chilled, frozen or simply preserved (0.65); 046 Meal and flour of wheat and flour of meslin (0.59); and 421 Fixed vegetable fats and 4. Results and discussion oils (0.52). (Appendix, Table 1A) 4.3. Analysis of intra-industry trade patterns As indicated in Figure 2 which illustrates IIT and its structure in bilateral trade flows, inter-industry trade Over most of the analysed period agri-food trade is more significant than IIT in agri-food trade with all with the EU indicated characteristics of strong inter- member states of the EU. The highest average shares industry trade although the trend of IIT was mostly in- of IIT were registered in trading with Croatia (0.35), creasing (except in 2011 and 2014). The increase in IIT Italy (0.32), Slovenia (0.18), Austria (0.15), and France intensity was particularly large in 2013, when Croatia (0.14), followed by the Czech Republic, Sweden, as one of the main BiH trade partners joined the EU Germany and Belgium. The importance of IIT was (Figure 1). particularly low in agri-food trade with the other 19 Structure of IIT is to a large extent dominated by its countries (GL<0.05). With six countries (Baltic coun- vertical component. Besides, over several years (2010- tries – Estonia, Latvia and Lithuania, and Ireland, 2016), the share of high VIIT (VIITh) was larger than Luxembourg and Malta), the trade in agri-food prod- that of low VIIT (VIITl). The share of HIIT was relatively ucts was either non-existent in some years or the en- small, though it was more significant in the beginning tire trade was of inter-industry type (GL=0.00). The of the period than later. However, the level of HIIT and similar applied to the trade with Denmark, Poland, high VIIT taken together indicate the quality advan- Portugal and Spain, where GL index was minimal tage i.e. the situation that the quality of BiH exports of (GL=0.01), and to the remaining nine countries, where agri-food product groups in which IIT was registered, average GL indices amounted to 0.03 or 0.04. is either similar or higher than in imports. (Figure 1) Figure 1. Intra-industry trade between BiH and the EU 0.35 0,35 0.30 0,30 0.30 0.26 0.26 0.25 0,25 0.24 0.25 0.21 0.23 VIITl 0.21 0.20 0.20 0,20 VIITh 0.15 0.18 0.15 0,15 HIIT Total IIT 0.10 0,10 0.05 0,05 0.00 0,00 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Notes: 2008-2012 EU27; 2013-2018 EU28. Legend: VIITl – vertical IIT in low quality products; VIITh – vertical IIT in high quality products; 0,00 0,20 0,40 0,00 0,20 0,40 0,00 0,20 0,40 HIIT – horizontal IIT. Austria Source: Authors’ own calculation based on the BHAS data. Belgium Bulgaria SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 59 Croatia Cyprus Czech Republic Denmark Estonia Finland France Germany Greece Hungary Ireland Italy Latvia Lithuania Luxembourg Malta Netherlands Poland Portugal Romania Slovakia Slovenia Spain Sweden United Kingdom 2018 Total IIT Aver. 2008-2018 Total IIT 2008 Total IIT 2018 HIIT 2008 HIIT Aver. 2008-2018 HIIT 2018 VIIT 2008 VIIT Aver. 2008-2018 VIIT 0,35 0,30 0.30 0.26 0.26 0,25 0.24 0.25 0.21 0.23 VIITl 0.21 0.20 0,20 0.15 VIITh 0.18 0,15 HIIT Total IIT 0,10 0,05 0,00 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Figure 2. Intra-industry trade of BiH with the EU members (2008, 2018, average) 0.00 0.20 0.40 0.00 0.20 0.40 0.00 0.20 0.40 0,00 0,20 0,40 0,00 0,20 0,40 0,00 0,20 0,40 Austria Belgium Bulgaria Croatia Cyprus Czech Republic Denmark Estonia Finland France Germany Greece Hungary Ireland Italy Latvia Lithuania Luxembourg Malta Netherlands Poland Portugal Romania Slovakia Slovenia Spain Sweden United Kingdom 2018 Total IIT Aver. 2008-2018 Total IIT 2008 Total IIT 2018 HIIT Aver. 2008-2018 HIIT 2008 HIIT 2008 VIIT 2018 VIIT Aver. 2008-2018 VIIT Legend: HIIT – horizontal IIT; VIIT – vertical IIT. Source: Authors’ own calculation based on the BHAS data. In its trade with most EU member states, BiH ex- In the structure of BiH intra-industry agri-food perienced a higher intensity of IIT at the end of the trade with all EU members a vertical component sub- observed period compared to its beginning (Figure stantially prevailed, except with Italy in the period 2). However, on the above mentioned “top five” list, 2008-2011, when the dominance of HIIT was regis- the continuously increasing IIT trend was registered tered. The share of VIIT also increased in the trade with only in trading with Croatia. Greater oscillations in IIT most of the EU member states until the end of the ob- trend were observed for Austria, Belgium, the Czech served period. Republic and France. 60 SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 ANALYSIS OF INTRA-INDUSTRY TRADE IN AGRI-FOOD PRODUCTS BETWEEN BOSNIA AND HERZEGOVINA AND THE EUROPEAN UNION 4.2. Model estimation results results. However, all three specifications indicate simi- lar effects of the considered determinants. We estimated our baseline model represented by The common history of the countries has the equation (3) using PPML estimator. The estimation strongest positive effects on IIT of all the considered results for the whole sample are presented in Table factors. This reflects an intensive IIT between BiH on 3, column (1). We also checked the robustness of the one side, and Croatia and Slovenia on the other, which obtained results by estimating the baseline model on could be a result of their participation in the common subsamples for periods 2010-2018 and 2008-2016. market of the former Yugoslavia. Another variable The results of these robustness checks are presented with a robust positive effect on IIT is the economic size in Table 3, columns (2) and (3), respectively. of trade partners. This result is in line with the findings In all presented specifications Ramsey Regression of Jámbor (2014) in terms of both the intensity and Equation Specification Error test indicates that the significance of the effects. Sharing a common bor - specification of the conditional expectation is correct, der could positively affect IIT intensity, although this providing no evidence of misspecification. The model effect was found to be significant only in robustness appears to fit the data value, as evidenced by the high check, but not in the baseline specification. R values. As the Bayesian information criterion has On the other hand, differences in agricultural the lowest value for specification (1), it is considered productivity are a major determinant that signifi- a preferred specification for the interpretation of the cantly negatively and robustly affects IIT. Such a result Table 3. Estimation results Variable / model IIT (1) IIT (2) IIT (3) ASIZE 0.002* 0.002* 0.002** jt (0.001) (0.001) (0.001) ΔDGDPC -0.027 -0.037 -0.016 jt (0.028) (0.028) (0.026) DIST -0.590*** -0.596* -0.348 (0.067) (0.358) (0.331) BORDER 0.566 0.508** 0.613** (0.347) (0.255) (0.264) ΔTI 0.136 -0.492 1.506 jt (2.906) (2.299) (2.158) DPROD -0.030** -0.029** -0.031** jt (0.013) (0.013) (0.013) COMMON 1.935*** 1.902*** 2.296*** (0.389) (0.409) (0.398) ΔDFDI -0.182 -0.27 -0.355 jt (0.497) (0.560) (0.630) Constant -3.073*** -2.991*** -3.513*** (0.135) (0.433) (0.413) Observations 249 224 224 R 0.830 0.825 0.848 BIC 169.206 283.069 286.678 Log-likelihood -48.74 -44.125 -43.223 RESET test 0.1320 0.0823 0.4133 Note: Standard errors are provided in the brackets. ***, **, and * denote coefficients statistically significant at the 1%, 5% and 10% levels. BIC refers to the Bayesian information criterion and the RESET test denotes the p-values of the Ramsey Regression Equation Specification Error test. Fixed individual and time effects are estimated but not reported. Source: Authors’ own calculation. SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 61 ANALYSIS OF INTRA-INDUSTRY TRADE IN AGRI-FOOD PRODUCTS BETWEEN BOSNIA AND HERZEGOVINA AND THE EUROPEAN UNION supports the previous findings reported in the re - impact of differences in productivity on IIT was found lated empirical literature. Geographic distance was to be significantly negative. Finally, the results of our also found to be negatively related to IIT. The effect study do not support the hypotheses related to the is, however, not robust in the case of the Specification significance of effects of trade intensity, as well as dif- (3). As for the differences in GDP p/c and FDI, as well ferences in GDP p/c and FDI. as for the trade intensity, no statistically significant ef- By providing a deeper insight into characteristics fects were found in any estimated specifications. and determinants of IIT, the paper contributes to a better understanding of the position of the agri-food sector of the EU potential candidate country in one of its most important foreign markets. It also provides 5. Conclusion useful information for policymakers. The empirical evidence presented in this pa- The significantly high share of inter-industry trade per confirms the  theoretical hypotheses explaining implies a higher possibility of increasing adjustment the  patterns of IIT of less advanced countries (and in costs for BiH agri-food sector, associated with further less differentiated products), and those explaining ef- liberalisation of trade with the EU. Contrary to the fects of country-specific determinants on IIT. smooth adjustment hypothesis (Balassa, 1966) related The analysis of characteristics of IIT in agri-food to a higher share of IIT, an increase in trade as a result trade between BiH and the EU points to several main of liberalisation might cause more transfer of pro- concluding observations. Firstly, a relative importance duction factors between expanding and contracting of intra-industry trade as opposed to inter-industry product lines and more temporary unemployment in trade was significantly lower in the observed period BiH agri-food sector. – inter-industry trade proved as a strongly dominant Therefore further association of BiH to the EU by form of trade specialisation in the agri-food trade of completing the free trade area, and by harmonizing BiH with the EU, viewed totally and bilaterally. Despite regulations in the field of trade in agri-food products, the increasing level of trade liberalisation between will require restructuring of BiH agri-food sector to the trading partners, an expected change in IIT pat- increase its competitiveness. With respect to our find- terns from dominant inter- to dominant intra-industry ings on IIT patterns and following Bojnec and Fertő trade did not happen. The time series analysis indi- (2016), it is recommended to focus on two specific cated that the intensity of IIT with the EU as a whole market niches that differ by income p/c. As price com- experienced a slight growth. However, the increasing petitiveness is more important for HIIT and low VIIT, trend seems to correspond to the latest EU enlarge- competing in prices enabled by increasing economy ment in 2013, when Croatia joined, rather than the of scale in production will attract more EU consum- growth in the level and scope of mutual trade liber- ers with lower income. On the other hand, in product alisation related to BiH obtaining the potential can- groups with dominant high VIIT, meaning higher qual- didate status. Trends in IIT intensity did not point to ity and higher value-added of BiH exports, as well as a significant positive development that would lead higher prices, efforts should focus on marketing pro - to greater convergence between agri-food sectors of motion and branding at micro-level (level of compa- the analysed trading partners. Finally, the analysis of nies) or local/regional level (level of agri-food clusters). IIT structure, i. e. the distinction between HIIT and VIIT, revealed that the agri-food trade of BiH with the EU Endnotes was dominated by its vertical component, referring to trade of different quality products. In trading with 1 After Croatia‘s accession to the European Union, the EU as a whole, the high VIIT was somewhat more the Western Balkans region includes only five eco - prominent than a low VIIT over the major part of the nomies: Albania, Bosnia and Herzegovina, North observed period. Macedonia, Serbia and Kosovo*. (*The name The IIT patterns were explained through the does not prejudge the status of Kosovo* and is analysis of impact of several country-specific charac - in line with the Resolution of the United Nations teristics. Our initial hypotheses are supported by the Security Council UNSC 1244 and the opinion of the estimates of the econometric model in the case of International Court of Justice on the Declaration on economies’ size, common border, and common his- Kosovo’s Independence of 2008). tory effects, all of which were found to positively af- 2 Empirical literature contains evidence that the share fect IIT. Our hypothesis regarding the negative effect of vertical IIT is larger than that of horizontal IIT in of distance is also supported by the presented empiri- the trade between countries at different levels of cal evidence, to a certain extent. As it is expected, the economic development (Mardas and Nikas 2008). 62 SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 ANALYSIS OF INTRA-INDUSTRY TRADE IN AGRI-FOOD PRODUCTS BETWEEN BOSNIA AND HERZEGOVINA AND THE EUROPEAN UNION References Fidrmuc, J. 2004. The Endogeneity of the optimum currency area criteria, intra-industry trade, and EMU enlargement. Abd-El-Rahman, K. S. 1986. Réexamen de la définition et de Contemporary Economic Policy 22 (1): 1-12. la mesure des échanges croisés de produits similaires Food and Agriculture Organization of the United Nations entre les nations (Review of the definition and meas- (FAO) 2020. FAOSTAT (database online). http://www.fao. urement of cross-trade in similar products between na- org/faostat/en/#data/RL (accessed 20 November 2020) tions). Revue Economique 37 (1): 89-116. Greenaway, D., Hine, R., and Milner, C. 1995. Vertical and hor- Agency for Statistics of Bosnia and Herzegovina (BHAS). izontal intra-industry trade: A cross industry analysis for Statistical database – Foreign trade statistics (Available the United Kingdom. The Economic Journal 105 (433): on request only) 1505-1518. Ambroziak, Ł. 2012. FDI and intra-industry trade: Theory Grubel, H. G., and Lloyd, P. J. 1975. Intra-Industry Trade: and empirical evidence from the Visegrad countries. The Theory and Measurement of International Trade in International Journal of Economics and Business Differentiated Products. London: The Macmillan Press. Research 4 (1/2): 180–198. Heckman, J. J. 1979. Sample selection bias as a specification Balassa, B. 1966. Tariff reductions and trade in manufactures error. Econometrica 47 (1): 153-161. among the industrialised countries. American Economic Helpman, E. 1981. International trade in the presence of Review 56 (3): 466-473. product differentiation, economies of scale and mo - Balassa, B. 1986. Intra-industry trade among exporters of nopolistic competition: A Chamberlin-Heckscher-Ohlin manufactured goods. In Imperfect competition and in- approach. Journal of International Economics 11 (3): ternational trade, edited by D. Greenaway and P. K. M. 305-340. Tharakan, 108-128. Brighton: Wheatshef. Helpman, E., and Krugman, P. R. 1985. Market structure and Balogh, J. M., and Leitão, N. C. 2019. A gravity approach foreign trade. 8th edition. Cambridge, MA: MIT Press. of agricultural trade: The nexus of the EU and African, International Monetary Fund (IMF) 2019. World Economic Caribbean and Pacific countries. Agricultural Economics Outlook Database (database online). https://www.imf. 65 (11): 509-519. org/en/Publications/WEO/weo-database/2019/October Blanes, J. V., and Martín, C. 2000. The nature and causes of (accessed July 2020) intra-industry trade: Back to the comparative advantage Jámbor, A. 2013. Country-specific determinants of horizon- explanation? The case of Spain. Weltwirtschaftliches tal and vertical intra-industry agri-food trade: the case of Archiv 136 (3): 423–441. Bulgaria and Romania. Applied Studies in Agribusiness Bojnec, Š., and Fertő, I. 2016. Patterns and drivers of the and Commerce 7 (4-5): 69-76. agri-food intra-industry trade of European Union coun- Jámbor, A. 2014. Country specific determinants of horizon- tries. International Food and Agribusiness Management tal and vertical intra-industry agri-food trade: The case Review 19 (2): 53-73. of the EU new member states. Journal of Agricultural Burger, M., Van Oort, F., and Linders, G-J. 2009. On the speci- Economics 65 (3): 663-682. fication of the gravity model of trade: Zeros, excess zeros Jámbor, A. 2015. Country and industry-specific determi- and zero-inflated estimation. Spatial Economic Analysis nants of intra-industry trade in agri-food products in the 4 (2): 167-190. Visegrad countries. Studies in Agricultural Economics Centre d’études prospectives et d’informations internation- 117 (2): 93-101. als (CEPII) 2020. Distance database (database online). Jámbor, A., and Leitão N. C. 2016. Industry-specific deter - http://www.cepii.fr/anglaisgraph/bdd/distances.htm minants of vertical intra-industry trade: the case of EU (accessed August 2020) new member states’ agri-food sector. Post-Communist Dixit, A. K., and Stiglitz, J. E. 1977. Monopolistic competition Economies 28 (1): 34-48. and optimum product diversity. American Economic Jámbor, A., Bologh, J., and Kucsera, P. 2016. Country and in- Review 67 (3): 297-308. dustry specific determinants of intra-industry agri-food Falvey, R. E. 1981. Commercial policy and intra-industry trade in the Baltic countries. Agricultural Economics trade. Journal of International Economics 11 (4): 495-511. (Zemědělská Ekonomika) 62 (6): 280–291. Falvey, R. E., and Kierzkowski, H. 1987. Product quality, intra- Jing, W., Leitão, N. C., and Faustino, H. C. 2010. Intra-industry industry trade and (im)perfect competition. In Protection trade in agricultural products: The case of China. Instituto and competition in international trade, edited by H. Superior de Economia e Gestão. DE Working papers Kierzkowski, 143-161. Oxford (USA): Basil Blackwell. 6-2010/DE/SOCIUS Fertő, I., and Hubbard, L. J. 2002. Intra-industry trade in hori- Kandogan, Y. 2003. Intra-industry trade of transition coun- zontally and vertically differentiated agri-food products tries: Trends and determinants. Emerging Markets between Hungary and the EU. Hungarian Academy of Review 4 (3): 273-286. Sciences. Institute of Economics Discussion Paper 2002/2 SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 63 ANALYSIS OF INTRA-INDUSTRY TRADE IN AGRI-FOOD PRODUCTS BETWEEN BOSNIA AND HERZEGOVINA AND THE EUROPEAN UNION Krugman, P. R. 1980. Scale economies, product differen- sector of Bosnia and Herzegovina. European Journal of tiation, and the patterns of trade. American Economic Interdisciplinary Studies 3 (3): 154-163. Review 70 (5): 950-959. Silva, J. M. C., and Teneyro, S. 2006. The log of gravity. The Krugman, P. R., and Obstfeld, M. 2009. International econom- Review of Economics and Statistics 88 (4): 641-658. ics: theory and policy. 8th edition. Boston, MA: Pearson Trivić,, J., and Klimczak, L. 2015. The determinants of intra- Education Inc. regional trade in the Western Balkans. Proceedings of Lancaster, K. 1980. Intra-industry trade under perfect Rijeka Faculty of Economics: Journal of Economics and monopolistic competition. Journal of International Business 33 (1): 37-66. Economics 10 (2): 151-175. United Nations Conference on Trade and Development Łapinska, J. 2014. Determinants of intra-industry trade in (UNCTAD) 2021. Statistical database: Product groups agricultural and food products between Poland and EU and composition (SITC Rev. 3) (database online). countries. Danube: Law and Economics Review 5 (3): h ttps://unc tadsta t .unc tad.org/EN/Classifica tions/ 159-172. DimSit cR ev3P r oduc ts_DsibSpecialGr oupings_ Leitão, N. C. 2011. Intra-industry trade in the agriculture sec- Hierarchy.pdf (accessed March 2021) tor: The experience of United States. African Journal of United Nations Statistics Division (UNSTAT) 2021. Statistical Agricultural Research 6 (1): 186-190. database (database online). https://unstats.un.org/ Linder, S. B. 1961. An essay on trade and transformation. home/ (accessed August 2020) New York: John Wiley & Sons. Venables, A. J., Rice, P. G., and Stewart, M. 2003. The geogra- Little, R. J. 1988. A test of missing completely at random phy of intra-industry trade: Empirics. Topics in Economic for multivariate data with missing values. Journal of the Analysis and Policy 3 (1): 1-25. American Statistical Association 83 (404): 1198-1202. World Bank (WB) 2020. Agriculture value added per worker Maddala, G. S., and Wu, S. 1999. A comparative study of unit (database online). https://datacatalog.worldbank.org/ root tests with panel data and a new simple test. Oxford agriculture-value-added-worker-constant-2010-us (ac- Bulletin of Economics and Statistics 61 (S1): 631-652. cessed July 2020) Mardas D., and Nikas, Ch. (2008): Economic integration and World Bank (WB) 2020. World development indicators (da- intra-industry trade between the European Community tabase online). https://databank.worldbank.org/indica- and the Western Balkan countries. Transition Studies tor/NY.GDP.PCAP.CD/1ff4a498/Popular-Indicators# (ac - Review 15 (3): 511-523. cessed July 2020) Mrdalj, V., Ostojić, A., Vasko, Z., and Brković, D. 2017. Structure and determinants of intra-industry trade in agri-food 64 SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 ANALYSIS OF INTRA-INDUSTRY TRADE IN AGRI-FOOD PRODUCTS BETWEEN BOSNIA AND HERZEGOVINA AND THE EUROPEAN UNION APPENDIX Table 1A. Values of GL index by product groups in BiH agri-good trade with the EU SITC 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Average 061 0.95 0.96 0.48 0.49 0.77 0.65 0.91 0.89 0.89 0.99 0.58 0.78 056 0.57 0.63 0.69 0.70 0.76 0.83 0.88 0.70 0.65 0.59 0.55 0.69 054 0.24 0.47 0.69 0.41 0.61 0.74 0.61 0.70 0.93 0.90 0.82 0.65 046 0.85 0.74 0.90 0.86 0.94 0.20 0.29 0.32 0.45 0.48 0.50 0.59 421 0.44 0.71 0.73 0.45 0.46 0.84 0.54 0.44 0.31 0.31 0.53 0.52 058 0.64 0.64 0.56 0.48 0.68 0.56 0.46 0.40 0.43 0.40 0.37 0.51 075 0.58 0.57 0.30 0.79 0.47 0.30 0.47 0.50 0.40 0.60 0.38 0.49 048 0.29 0.28 0.26 0.29 0.28 0.67 0.70 0.67 0.62 0.57 0.54 0.47 121 0.55 0.97 0.25 0.50 0.02 0.21 0.84 0.46 0.61 0.18 0.11 0.43 057 0.11 0.10 0.33 0.23 0.26 0.62 0.27 0.40 0.35 0.44 0.40 0.32 111 0.01 0.11 0.12 0.16 0.07 0.46 0.43 0.41 0.47 0.58 0.60 0.31 071 0.29 0.28 0.31 0.30 0.16 0.19 0.19 0.23 0.21 0.22 0.25 0.24 431 0.10 0.27 0.37 0.27 0.34 0.06 0.10 0.11 0.18 0.32 0.34 0.22 059 0.15 0.22 0.25 0.05 0.23 0.17 0.21 0.14 0.28 0.24 0.25 0.20 112 0.13 0.14 0.17 0.18 0.34 0.16 0.13 0.13 0.16 0.20 0.23 0.18 081 0.06 0.11 0.12 0.09 0.18 0.16 0.17 0.15 0.15 0.22 0.22 0.15 044 0.00 0.26 0.12 0.27 0.03 0.03 0.14 0.07 0.14 0.16 0.37 0.15 072 0.00 0.05 0.00 0.00 0.11 0.10 0.14 0.00 0.01 0.43 0.75 0.14 098 0.16 0.13 0.13 0.10 0.06 0.17 0.19 0.14 0.14 0.14 0.12 0.13 012 0.01 0.00 0.01 0.24 0.15 0.18 0.21 0.19 0.32 0.01 0.00 0.12 091 0.05 0.01 0.00 0.00 0.01 0.06 0.17 0.28 0.13 0.13 0.27 0.10 022 0.00 0.00 0.00 0.00 0.01 0.38 0.00 0.02 0.11 0.35 0.19 0.10 047 0.03 0.04 0.04 0.04 0.08 0.14 0.12 0.17 0.11 0.06 0.08 0.08 223 0.00 0.00 0.00 0.04 0.01 0.09 0.12 0.01 0.13 0.21 0.28 0.08 074 0.35 0.21 0.05 0.00 0.01 0.00 0.00 0.01 0.04 0.11 0.02 0.07 073 0.00 0.17 0.03 0.04 0.03 0.09 0.10 0.08 0.07 0.06 0.05 0.07 041 0.00 0.00 0.00 0.00 0.00 0.07 0.00 0.00 0.00 0.00 0.43 0.05 122 0.00 0.01 0.06 0.13 0.07 0.02 0.01 0.02 0.03 0.04 0.11 0.04 024 0.00 0.00 0.00 0.00 0.00 0.17 0.00 0.02 0.06 0.07 0.08 0.04 017 0.16 0.00 0.00 0.00 0.00 0.19 0.00 0.00 0.00 0.00 0.00 0.03 025 0.02 0.00 0.00 0.00 0.00 0.32 0.01 0.00 0.00 0.00 0.00 0.03 222 0.02 0.01 0.01 0.06 0.02 0.01 0.06 0.01 0.00 0.02 0.01 0.02 001 0.02 0.01 0.02 0.02 0.03 0.02 0.02 0.03 0.03 0.01 0.00 0.02 062 0.01 0.01 0.01 0.01 0.01 0.01 0.02 0.02 0.03 0.07 0.01 0.02 023 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.03 0.05 0.05 0.01 045 0.00 0.00 0.00 0.02 0.00 0.00 0.00 0.00 0.01 0.01 0.00 0.00 016 0.00 0.00 0.00 0.00 0.00 0.04 0.00 0.00 0.00 0.00 0.00 0.00 422 0.00 0.00 0.00 0.00 0.00 0.01 0.02 0.00 0.00 0.00 0.00 0.00 043 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.02 0.00 042 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 011 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 411 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Note: The period 2008-2012 refers to the EU27. The period 2013-2018 refers to the EU28. Source: Authors‘ own calculation. SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 65 ANALYSIS OF INTRA-INDUSTRY TRADE IN AGRI-FOOD PRODUCTS BETWEEN BOSNIA AND HERZEGOVINA AND THE EUROPEAN UNION Table 2A. Vertical IIT in BiH agri-food trade with the EU member countries 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Average Austria 0.08 0.09 0.09 0.08 0.13 0.13 0.16 0.14 0.16 0.17 0.14 0.12 Belgium 0.00 0.02 0.02 0.00 0.04 0.09 0.14 0.03 0.14 0.07 0.09 0.06 Bulgaria 0.00 0.01 0.00 0.01 0.00 0.01 0.01 0.03 0.04 0.05 0.09 0.02 Croatia 0.24 0.28 0.29 0.28 0.26 0.31 0.27 0.28 0.33 0.34 0.32 0.29 Cyprus 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.15 0.00 0.00 0.00 0.01 Czech Republic 0.15 0.03 0.01 0.07 0.01 0.11 0.09 0.04 0.06 0.12 0.21 0.08 Denmark 0.00 0.00 0.00 0.01 0.00 0.00 0.03 0.02 0.01 0.02 0.03 0.01 Estonia 0.00 0.00 0.00 0.00 0.00  NT 0.00 0.00 0.00 0.00 0.00 0.00 Finland 0.00 0.00 0.00 0.13 0.19 0.00 0.00 0.00 0.02 0.00 0.00 0.03 France 0.05 0.09 0.09 0.08 0.16 0.12 0.14 0.07 0.29 0.17 0.13 0.13 Germany 0.05 0.04 0.05 0.05 0.06 0.05 0.10 0.11 0.10 0.11 0.11 0.08 Greece 0.04 0.01 0.00 0.00 0.00 0.02 0.02 0.05 0.03 0.03 0.04 0.02 Hungary 0.01 0.01 0.01 0.04 0.04 0.02 0.03 0.04 0.07 0.05 0.11 0.04 Ireland 0.00 0.00 0.00 0.01 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 Italy 0.16 0.18 0.22 0.14 0.39 0.37 0.32 0.29 0.23 0.26 0.19 0.25 Latvia  NT 0.00 0.00 0.00  NT  NT 0.00 0.00 0.00 0.00 0.00 0.00 Lithuania 0.00 0.00 0.00 NT  0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Luxembourg 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Malta  NT 0.00  NT  NT NT   NT 0.00 0.00 0.00 0.00 0.00 0.00 Netherlands 0.02 0.04 0.04 0.03 0.00 0.02 0.02 0.02 0.02 0.03 0.04 0.03 Poland 0.00 0.00 0.00 0.01 0.01 0.00 0.01 0.01 0.01 0.02 0.02 0.01 Portugal 0.04 0.01 0.03 0.06 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.01 Romania 0.00 0.00 0.00 0.00 0.00 0.00 0.07 0.06 0.16 0.03 0.08 0.04 Slovakia 0.00 0.07 0.02 0.00 0.01 0.00 0.01 0.09 0.01 0.02 0.03 0.02 Slovenia 0.15 0.18 0.17 0.13 0.15 0.15 0.16 0.18 0.19 0.16 0.16 0.16 Spain 0.00 0.01 0.00 0.00 0.03 0.04 0.02 0.01 0.00 0.01 0.00 0.01 Sweden 0.07 0.13 0.10 0.10 0.10 0.02 0.05 0.03 0.02 0.08 0.10 0.07 United 0.02 0.01 0.03 0.02 0.02 0.03 0.04 0.02 0.04 0.04 0.05 0.03 Kingdom Legend: NT – no agri-food trade between BiH and the EU member country in the given year. Source: Authors‘ own calculation. 66 SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 ANALYSIS OF INTRA-INDUSTRY TRADE IN AGRI-FOOD PRODUCTS BETWEEN BOSNIA AND HERZEGOVINA AND THE EUROPEAN UNION Table 3A. Horizontal IIT in BiH agri-food trade with the EU member countries 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Average Austria 0.01 0.10 0.00 0.01 0.05 0.07 0.01 0.04 0.00 0.01 0.00 0.03 Belgium 0.00 0.00 0.00 0.00 0.04 0.02 0.00 0.01 0.00 0.00 0.00 0.01 Bulgaria 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.02 0.00 Croatia 0.10 0.09 0.06 0.08 0.10 0.03 0.05 0.00 0.00 0.03 0.06 0.05 Cyprus 0.00 0.11 0.00 0.00 0.00 0.20 0.00 0.00 0.00 0.00 0.00 0.03 Czech Republic 0.06 0.00 0.00 0.00 0.00 0.00 0.00 0.04 0.07 0.00 0.00 0.02 Denmark 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Estonia 0.00 0.00 0.00 0.00 0.00 NT 0.00 0.00 0.00 0.00 0.00 0.00 Finland 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 France 0.01 0.00 0.01 0.00 0.01 0.00 0.00 0.01 0.01 0.03 0.05 0.01 Germany 0.00 0.01 0.00 0.00 0.01 0.02 0.00 0.01 0.01 0.00 0.00 0.01 Greece 0.00 0.04 0.02 0.00 0.00 0.05 0.00 0.00 0.00 0.00 0.00 0.01 Hungary 0.00 0.02 0.01 0.00 0.00 0.02 0.00 0.00 0.00 0.00 0.00 0.00 Ireland 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Italy 0.20 0.21 0.18 0.20 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.07 Latvia NT 0.00 0.00 0.00 NT NT 0.00 0.00 0.00 0.00 0.00 0.00 Lithuania 0.00 0.00 0.00 NT 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Luxembourg 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Malta NT 0.00 NT NT NT NT 0.00 0.00 0.00 0.00 0.00 0.00 Netherlands 0.02 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.01 0.00 0.00 0.00 Poland 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 Portugal 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Romania 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.01 0.00 Slovakia 0.00 0.00 0.00 0.00 0.00 0.03 0.00 0.00 0.00 0.00 0.00 0.00 Slovenia 0.00 0.03 0.02 0.03 0.04 0.02 0.02 0.02 0.00 0.00 0.01 0.02 Spain 0.00 0.01 0.00 0.01 0.02 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Sweden 0.00 0.00 0.01 0.00 0.00 0.04 0.03 0.04 0.08 0.00 0.00 0.02 United 0.00 0.00 0.00 0.00 0.02 0.00 0.00 0.00 0.01 0.00 0.00 0.00 Kingdom Legend: NT – no agri-food trade between BiH and the EU member country in the given year. Source: Authors‘ own calculation. SOUTH EAST EUROPEAN JOURNAL OF ECONOMICS AND BUSINESS, VOLUME 16 (2) 2021 67

Journal

South East European Journal of Economics and Businessde Gruyter

Published: Dec 1, 2021

Keywords: Intra-industry trade (IIT); agri-food products; Bosnia and Herzegovina (BiH); the European Union (EU); Poisson Pseudo-maximum likelihood (PPML) approach; F14; Q17; O52

There are no references for this article.