Agri-food competitiveness in South East Europe

Abstract

The purpose of this article is to determine the level of competitiveness of agri-food products in South East European (SEE) countries within the processes of European Union (EU) and regional integration as well as to find the factors that determine agri-food competitiveness. This article uses the revealed comparative advantages (RCAs) index to find the level of comparative advantage of agri-food products. Additionally, a model for identifying the determinants of the SEE agri-food comparative advantage was constructed and estimated. The results show that all SEE countries (except for Albania) have comparative advantages in the agri-food sector as part of the global market. Also, the estimation of the model shows that partial productivities in agriculture have a positive impact on comparative advantage while gross domestic product (GDP) per capita has a negative impact. This article makes a useful review of competitiveness of agri-food sector in SEE countries and determines which factors are significant for an RCA index. This is essential for policymakers to identify what determinants improve or degrade competitiveness of the agri-food sector in SEE countries.

Keywords

Competitiveness revealed comparative advantages agri-food sector South East Europe land productivity labour productivity

Introduction

After the turbulent period of transformation from centrally planned economies to a market-oriented economies, all South East European (SEE) countries¹ have clearly identified integration into the European Union (EU) as a political priority, and they have determined that agricultural policy reform and modernizing agriculture (among other issues) are necessary to increase competitiveness across the entire agri-food chain (Eberlin et al., 2014). This is not a simple task for these countries, and according to experience from previous EU enlargements, this process could be long and complex. Each SEE country is at a different membership stage: Croatia has been a full member of the EU since 2013, while Albania, Bosnia and Herzegovina (B&H), North Macedonia, Montenegro and Serbia are at different stages in the accession process (Table 1). Since these countries are natural trade partners with convergent economies at the same level of competitiveness (Dragutinović-Mitrović and Bjelić, 2015) and to prepare for EU membership and improve regional cooperation, the SEE countries established a free trade zone through the Central European Free Trade Agreement (CEFTA). CEFTA is the highest contributing factor to intra-regional trade among SEE countries. Additionally, all EU Member States are members of the World Trade Organization (WTO), so some SEE countries will eventually have to complete additional steps for membership in this organization. The WTO provides the basic principles of trade cooperation between states, enables trade to proceed in a stable and predictable manner and allows each state to be an equal partner in international trade negotiations.

Table 1.

SEE countries in process of integrations.

Country	EU membership status	Member of CEFTA	Member of WTO
Albania	Candidate (received candidate status in 2014)	Member from 2007	Member from 2000
B&H	Potential candidate (application to join the EU submitted in 2016)	Member from 2007	In the process of accession
Croatia	Member state (joined the EU in 2013)	Left in 2013	Member from 2000
North Macedonia	Candidate (received candidate status in 2005)	Member from 2006	Member from 2003
Montenegro	Candidate (received candidate status in 2010)	Member from 2007	Member from 2012
Serbia	Candidate (received candidate status in 2012)	Member from 2007	In the process of accession

SEE: South East European; EU: European Union; CEFTA: Central European Free Trade Agreement; WTO: World Trade Organization.

Source: European Commission (2018); CEFTA (2018); WTO (2018).

One of the basic conditions for EU membership is a country’s ability to cope with the competitive pressures of the single European market. Since the SEE economies are ranked at the bottom in terms of global competitiveness, joining the EU is a monumental task (Bjelić, 2014). Agriculture is the most sensitive part of these countries’ economies, so it will face numerous challenges. More specifically, agriculture is an economically and politically important sector in the SEE and is characterized by structural deficit, underutilized resources and production potentials, underdeveloped agri-food chains, marginalization of rural areas and, with the exception of Serbia, net trade deficits (Erjavec et al., 2014).

As long-term sustainable development of the agri-food sector can only be assured when agri-food products are competitive in regional and global trade, an analysis of agri-food products’ competitiveness as well as determinants of its competitiveness are the main research focus of this article. Defining and measuring competitiveness and its determinants are not a simple task because competitiveness does not have a universally accepted definition, especially within macroeconomics. One of the most popular definitions of competitiveness was established by the OECD (1992), which describes it as ‘the degree to which, under open market conditions, a country can produce goods and services that meet the test of foreign competition while simultaneously maintaining and expanding domestic real income’. Another definition, and the one most widely accepted in the literature, is defined by the World Economic Forum (2015) as ‘a set of institutions, policies and factors that determine the level of productivity of a country’. Competitiveness at the macroeconomic level is related to the concept of comparative advantage, and this concept posits that trade flows exist as a result of relative cost differences between trading partners. Although competitive advantages are based on comparative advantages, these are two separate concepts. The most significant difference between comparative advantage and competitiveness is that competitiveness includes market distortions (Bojnec and Ferto, 2009). Also, comparative advantages are based on differences in labour and capital, whereas various other factors (infrastructure, technology and conducive environments) influence competitiveness (Jambor and Babu, 2016). In the agri-food sector, according to the European Commission (2009), productivity is the most reliable indicator of competitiveness in the long term. A country’s competitiveness can also be determined by a combination of several factors, including partial productivities in agriculture – land and labour productivity, land endowment, gross domestic product (GDP) per capita, WTO membership and tariffs, which, to a different extent, can be significantly related to agricultural competitiveness (Jambor and Babu, 2016). Additionally, agricultural policy also has a significant impact on shaping a country’s agri-food competitiveness.

SEE countries’ agri-food sector competitiveness has been the focus of a limited number of papers. Most of them indicate changes in foreign trade flows as part of the processes of EU and regional integration as well as a low level of competitiveness in comparison to EU countries. Dragutinović-Mitrović and Bjelić (2015) indicate that SEE integration with the EU has a positive effect on trade and that greater positive effects on SEE trade were achieved during the first stages of EU integration when asymmetric trade preferences were applied. This was due to SEE economies’ low international competitive position vis-á-vis the EU as their main partner. However, according to Bojnec and Ferto (2009), EU enlargement has had a negative impact on agri-food relative trade advantages, and the main changes in comparative advantages were influenced by the transition from a centrally planned economy to a market economy, trade liberalization, free trade agreements, regional European reintegration and rapid adjustments to EU membership. Low-quality agri-food products are one of the main barriers to increasing this sector’s competitiveness. Bojnec and Ferto (2015) highlight the importance of agri-food sector internalization and competitive agri-food export integration into global markets, which enables greater market efficiency and transmission between national and global agri-food markets. Additionally, Matkovski et al. (2018) found that EU and CEFTA integration made considerable contributions to the trade of agri-food products in SEE countries, while CEFTA had noticeably larger effects on increasing agri-food trade than the Stabilization and Association Agreement (SAA). CEFTA countries are natural trade partners with similar levels of competitiveness, so it had been expected that CEFTA would significantly improve trade among these countries.

Since the study by Balassa (1965), which found that competitiveness could be measured by RCA, significant contributions to the literature on the RCA of trade of agri-food products have been made. Bojnec and Ferto (2017) investigated agri-food export competitiveness on the global market for 23 major countries, which together account for more than 60% of global agri-food trade, and concluded with RCA that most of these countries were competitive with agri-food exports. Using RCA, Torok and Jambor (2013) analysed the effects of EU accession on agri-food trade for new EU Member States, and their results suggest that the intensity of the agri-food trade increased significantly after accession. There have been a limited number of studies analysing the comparative advantages of the agri-food sector in SEE countries. Matkovski et al. (2016) analysed the level of comparative advantages of agri-food sector in these countries and concluded that comparative advantages are at a satisfactory level: Serbia has the strongest comparative advantages, while all other countries (except for Albania) have a satisfactory level of comparative advantages. Also, Matkovski et al. (2017) analysed Serbia’s comparative advantages for agri-food products and concluded that Serbia has comparative advantages in the export of agri-food products on the global market as well as on the EU and CEFTA markets, and they highlighted that products at a lower processing phase have the largest comparative advantages on the global market. Using RCA, Nikolić et al. (2011) indicated sectors and markets in which B&H exports are competitive in relation to their trading partners and found that B&H’s agri-food products have better overall performances on the SEE market in comparison to the EU market. These authors also confirmed that trade liberalization created opportunity that is not efficiently utilized by B&H’s agri-food sector. Several potential factors can determine an agri-food sector’s comparative advantages and competitiveness. Jambor and Babu (2016) used regression analysis to illustrate the process of identifying the determinants of agri-food competitiveness and found that land and labour productivity, land endowment, GDP per capita, product support estimate (PSE) values, WTO membership and tariffs are all to varying extents significantly related to global agricultural competitiveness. Overall, research into the trends and determinants of competitiveness for the agri-food sectors in SEE countries are limited. For this reason, this article aims to contribute to this gap in the literature. Here we will focus on levels and trends of comparative advantages for the agri-food sector in SEE countries and on identifying the determinants of agri-food competitiveness in these countries. Furthermore, information about factors that determine the agri-food sector’s competitiveness would be useful for policymakers in SEE countries. The results of this analysis should point to the main sources of competitiveness, which could influence the development of national agricultural policies. This, in turn, could create better competitive positions on the international market.

Since the crucial elements in the prosperity of agri-food products on the global market are trade liberalization and regional integration on global markets, and export competitiveness and its long-term duration (Bojnec and Ferto, 2017), the main objective of this article is to find the level of competitiveness of agri-food products in the SEE countries using the concept of revealed comparative advantages (RCA). To fulfil this objective, determinants of agri-food competitiveness in the SEE countries were also analysed.

The article is organized as follows: The introduction to the topic is developed together with a literature review as the theoretical basis for this study. The material and methods used in this article are described in the second section, where the main research hypotheses are also developed. The third section includes a presentation of the results, which are then elaborated on in the discussion. The conclusion includes implications of the results as well as consideration of future trends in agri-food competitiveness in SEE countries.

Materials and methods

There is no single measure of competitiveness, primarily because of the difficulty of evaluating agricultural competitiveness. This difficulty arises from the complexity and ambiguous nature of the phenomenon of competitiveness, as well as to agriculture’s internal diversity and the complexity of its surroundings (Nowak and Kaminska, 2016). The most common method for analysing a country’s trade-based competitiveness is RCA. In this article, an index of RCA is used in its basic form (Balassa, 1965):

{RCA}_{i j} = \frac{X_{i j} / X_{i t}}{X_{n j} / X_{n t}}

where X is the export, i is the country, j is the sector, t is the total export, and n is the group of exporting countries.

An index of RCA larger than 1 indicates that comparative advantages are present, and when the RCA value increases, so do the sector’s comparative advantages. This index has been used in several studies in the literature to determine comparative advantages at the national level, for sectors, and at the commodities level (Balogh and Jambor, 2017; Bojnec and Ferto, 2017; Bojnec and Ferto, 2018; Ferto, 2008; Ignjatijević et al., 2014; Nikolić et al., 2011). An alternative measure of comparative advantages is the Lafay Index (LFI) (Lafay, 1992), which is often used in the literature at a lower level of trade data aggregation (division, commodity group or product):

{LFI}_{j}^{i} = 100 (\frac{x_{j}^{i} - m_{j}^{i}}{x_{j}^{i} + m_{j}^{i}} - \frac{\sum_{j = 1}^{N} (x_{j}^{i} - m_{j}^{i})}{\sum_{j = 1}^{N} (x_{j}^{i} + m_{j}^{i})}) \frac{x_{j}^{i} + m_{j}^{i}}{\sum_{j = 1}^{N} (x_{j}^{i} + m_{j}^{i})}

where is $x$ is the export, m is the import, i is the country, j is the sector, division or commodity group, and N is the number of items. When LFI > 0, it implies that comparative advantages exist, and the greater value of this index is, the greater a county’s level of specialization will be in the trade of the sector, product or commodity group. LFI is referred to in the literature as being a more complete method because it provides a more complete analysis of a product’s specific positions in terms of its participation in a particular country’s foreign trade exchange.

In this article, an analysis of the impact of different determinants on RCA such as land productivity, labour productivity, GDP per capita, inflation and WTO membership was conducted for the period 2005–2016. A model was then constructed based on the previous empirical studies (Figure 1).

Figure 1.

Model for identifying the determinants of SEE agri-food comparative advantages.

The model includes an RCA indicator as а dependent variable while land and labour productivity, GDP per capita and macroeconomic stability are determined as explanatory variables, and membership status in WTO is included as а dummy variable. This study includes several hypotheses based on this article’s objectives, which are defined as follows:

H1: Partial productivities of agriculture have a positive influence on SEE countries’ agri-food comparative advantages.

H2: GDP per capita and macroeconomic stability have a positive influence on SEE countries’ agri-food comparative advantages.

H3: WTO membership has a positive influence on SEE countries’ agri-food comparative advantages.

This study analyses the six SEE countries (Albania, B&H, Croatia, North Macedonia, Montenegro and Serbia) over a 12-year period (2005–2016). The study also includes a panel model estimation that covers the time and space dimension. The pooled OLS model was used to analyse the impact of indicators on the RCA in the SEE countries:

{RCA}_{it} = β_{0} + β_{1} {LAND}_{it} + β_{2} {LABOUR}_{it} + β_{3} {GDPpc}_{it} + β_{4} {INF}_{it} + β_{5} {Dummy}_{it} + ∊_{i}

where RCA is the revealed comparative advantage, LAND is the land productivity, LABOUR is the labour productivity, GDPpc is the GDP per capita, INF is the macroeconomic stability – inflation, Dummy is the membership of WTO – 1 if a country is a member of WTO, otherwise 0, $∊_{i}$ is the residual.

The concept of agri-food products covers several different sections and divisions, according to the Standard International Trade Classification (SITC)–Revision 4 (Table 2):

Table 2.

Concept of agri-food products.

Section	Division
0 Food and live animals	00 – Live animals
	01 – Meat and meat preparations
	02 – Dairy products and birds’ eggs
	03 – Fish (not marine mammals), crustaceans, molluscs and aquatic invertebrates, and preparations thereof
	04 – Cereals and cereal preparations
	05 – Vegetables and fruit
	06 – Sugars, sugar preparations and honey
	07 – Coffee, tea, cocoa, spices and manufactures thereof
	08 – Feeding stuff for animals (not including unmilled cereals)
	09 – Miscellaneous edible products and preparations
1 Beverages and tobacco	11 – Beverages
1 Beverages and tobacco	12 – Tobacco and tobacco manufactures
2 Crude materials, inedible, except fuels	21 – Hides, skins and furskins, raw
	22 – Oil seeds and oleaginous fruits
	26 – Silk, cotton, jute, vegetable textile fibres, wool and other animal hair (261, 263, 264, 265, 268)
	29 – Crude animal and vegetable materials, n.e.s
4 Animal and vegetable oils, fats and waxes	41 – Animal oils and fats
	42 – Fixed vegetable fats and oils, crude, refined, or fractioned
	43 – Animal or vegetable fats and oils, processed

Source: The authors’ composition.

This article includes annual data obtained from several sources (Table 3): the Food and Agriculture Organization Statistical data base (FAOSTAT, 2018), World Bank data base (WB, 2018) and the United Nations Comtrade data base (UNComtrade).

Table 3.

Description of variables.

Variable	Variable description	Data source
RCA	RCA calculated according to formula 1	United Nations Comtrade data base (UNComtrade)
Land productivity (LAND)	Land productivity – Value of agricultural production in US$ (constant prices 2004–2006) per hectare of agricultural land	Food and Agriculture Organization Statistical data base (FAOstat)
Labour productivity (LABOUR)	Labour productivity – Value of agricultural production in US$ (constant prices 2004–2006) per number employed in agriculture	Food and Agriculture Organization Statistical data base (FAOstat)
GDPpc	Gross Domestic Product per capita in current US$	World Bank data base (WB)
Macroeconomic stability –INF	Macroeconomic stability – Inflation, consumer prices (annual %)	World Bank data base (WB)
Membership of WTO (Dummy)	Membership of WTO –1 if a country is a member of WTO, otherwise 0	WTO

WTO: World Trade Organization; RCA: revealed comparative advantage; GDPpc: GDP per capita; INF: Inflation.

Source: The authors’ composition.

Results

Comparative advantages in the export of agri-food products are largely dependent on the performance of agricultural production. These performances are directly related to agro-ecological conditions and the level of general economic development. SEE countries (and ex-Yugoslav countries, in particular) have similar historic and economic development, and mostly similar structures of production and agricultural holdings. The primary difference is in the available agricultural resources, which largely defines the differences in comparative advantages for the export of agri-food products. Serbia, which has the most utilized agricultural area and arable land per inhabitant (more than the EU), creates the highest export surpluses. Albania, which differs from other SEE countries, lags behind due to disadvantages in terms of resources and the dominance of a small, semi-natural production sector.

Considering the value of the index of RCA for agri-food products in the SEE countries (Figure 2), it can be concluded that all SEE countries (except for Albania) have a comparative advantage in the export of these products during the period analysed here. The analysis of the differences in comparative advantages of agri-food products for individual SEE countries clearly indicates that Serbia has the highest level of comparative advantages in this sector. Also, according to the index value of the comparative advantages for North Macedonia, which also has significant comparative advantages, North Macedonia experienced the largest decline in comparative advantage during this period. A satisfactory level of RCA for the export of agri-food products in B&H, Croatia, and Montenegro is noticeable, as are positive trends in this index in these countries. In Montenegro, the largest fluctuations of the RCA were present during this period and considering the value of agri-food product exports and the values of total exports were considerably lower than in other countries, such trends were to be expected. We noticed a drastic increase of RCA in 2014, which is a consequence of a dramatic increase in exports of meat and meat products recorded in Montenegro. The greatest stability in the trends of comparative advantages was in Croatia, which is the only EU member country, and to a certain extent this stabilized the position of their products on the international market. The worst situation was in Albania, which in the most years did not achieve a satisfactory level of comparative advantage for agri-food products export on the international market. As already mentioned, in some countries, there are noticeable negative trends in RCA, and one reason may be inadequate responses to improvements in competitiveness required by the world market in terms of regional and international integration. Additional reasons can be found in the production structure of agriculture, which is characterized by plant products with a lower processing level and products with low added value, had a dominant influence on the structure of agri-food export.

Figure 2.

Revealed comparative advantages of agri-food products in the SEE countries.

When considering the differences in the comparative advantages of agri-food products according to sector, only Serbia has significant comparative advantages in the food and live animals section, in which cereals and cereal preparations have the highest comparative advantages (Table 4). The beverages and tobacco section has comparative advantages in the international market in most of the countries analysed, with the highest level of comparative advantages for this section during this period recorded in North Macedonia, followed by Montenegro, Serbia and Croatia. The crude material section has comparative advantages in the international market in all countries analysed (with the exception of North Macedonia), while the animal and vegetable oils, fats and waxes section has comparative advantages only in Serbia and Bosnia and Herzegovina.

Table 4.

LFI for sections and divisions of agri-food products of SEE countries in period 2005–2016.

Section/Division	Albania	Bosnia and Herzegovina	Croatia	North Macedonia	Montenegro	Serbia
0 – Food and live animals	−2.89	−3.41	0.09	−1.02	−3.00	5.02
00 – Live animals	−0.24	−0.22	−0.13	0.01	−0.27	0.11
01 – Meat and meat preparations	−0.49	−0.34	−0.23	−0.72	−0.49	0.18
02 – Dairy products and birds’ eggs	−0.16	−0.10	−0.08	−0.26	−0.61	0.20
03 – Fish (not marine mammals), crustaceans, molluscs and aquatic invertebrates, and preparations thereof	0.59	−0.06	0.37	−0.10	−0.14	−0,17
04 – Cereals and cereal preparations	−1,37	−0,81	0.22	−0,11	−0.50	2.15
05 – Vegetables and fruit	−0.23	−0.18	−0.53	1.36	−0.03	1.70
06 – Sugars, sugar preparations and honey	−0.32	−0.20	0.35	−0.28	−0.15	0.64
07 – Coffee, tea, cocoa, spices and manufactures thereof	−0.29	−0.57	−0.09	−0.39	−0.35	−0.15
08 – Feeding stuff for animals (not including unmilled cereals)	−0.10	−0.38	−0.17	−0.20	−0.21	0.23
09 – Miscellaneous edible products and preparations	−0.29	−0.55	0.37	−0.33	−0.25	0.13
1 – Beverages and tobacco	−0.99	−1.00	0.43	1.99	0.83	0.74
11 – Beverages	−0.52	−0.73	0.19	0.67	0.88	0.57
12 – Tobacco and tobacco manufactures	−0.46	−0.28	0.24	1.32	−0.05	0.17
2 – Crude materials, inedible, except fuels	0.73	0.02	0.15	−0.08	0.23	0.22
21 – Hides, skins and furskins, raw	0.20	0.28	0.06	0.05	0.29	0.08
22 – Oilseeds and oleaginous fruits	0.00	−0.18	0.18	−0.07	−0.01	0.13
26 – Silk, cotton, jute, vegetable textile fibres, wool and other animal hair (261, 263, 264, 265, 268)	0.00	−0.03	0.00	−0.02	0.00	0.00
29 – Crude animal and vegetable materials, n.e.s	0.53	−0.06	−0.09	−0.04	−0.05	0.01
4 – Animal and vegetable oils, fats and waxes	−0.39	0.03	−0.10	−0.32	−0.09	0.48
41 – Animal oils and fats	−0.03	−0.01	0.00	−0.01	0.03	0.00
42 – Fixed vegetable fats and oils, crude, refined, or fractioned	−0.36	0.08	−0.09	−0.28	−0.12	0.47
43 – Animal or vegetable fats and oils, processed	−0.01	−0.04	0.00	−0.03	0.00	0.01

Source: The authors’ calculations on the basics of UNComtrade data base, 2018.

The econometric analysis includes descriptive statistics, multivariate analysis of variance tests, and diagnostics tests as well as a pooled regression model of ordinary least squares. Descriptive values of explanatory variables are presented first (Table 5).

Table 5.

Descriptive statistics.

	RCA	LAND	LABOUR	GDPpc	INF
Albania
Mean	0.96	979.75	1808.25	3972.75	2.42
Std. dev.	0.24	129.67	314.23	583.49	0.74
Min	0.49	851	1386	2709	1.28
Max	1.35	1198	2320	4576	3.55
Bosnia and Herzegovina
Mean	1.04	403.67	5023.75	4518	1.92
Std. dev.	0.08	30.02	420.85	687.32	2.84
Min	0.93	363	4027	2968	-1.25
Max	1.16	465	5691	5194	7.42
Croatia
Mean	1.55	1095.42	7628.58	13098.25	2.08
Std. dev.	0.09	109.27	1890.26	1547.87	2.01
Min	1.36	926	4703	1022	-1.1
Max	1.76	1261	11086	1589	6.08
North Macedonia
Mean	1.96	645.33	6438.83	4566.17	2.02
Std. dev.	0.37	57.44	328.23	752.27	2.58
Min	1.37	556	5934	3038	-0.74
Max	2.5	745	6976	5469	8.33
Montenegro
Mean	1.88	307.25	9511.75	6362.83	2.83
Std. dev.	0.59	113.17	1852.18	1177.19	2.51
Min	1.29	179	6835	3675	-0.71
Max	3.47	458	12657	7378	8.76
Serbia
Mean	2.68	1056.08	7035.42	5522.67	7.64
Std. dev.	0.24	53.48	739.88	929.59	4.67
Min	2.34	955	5310	3528	1.12
Max	2.97	1132	8180	6702	16.12

WTO: World Trade Organization; RCA: revealed comparative advantage; GDPpc: GDP per capita; INF: Inflation.

Source: The authors’ calculations.

The results from Table 5 show there is a significant difference in levels of comparative advantages among countries. The highest values of the RCA indicator were recorded in Montenegro (3.47) and Serbia (2.97), while Albania and B&H had minimum values (below 1). Although Montenegro had the highest RCA indicator in this period, the mean RCA indicator was largest in Serbia in comparison to other countries. Standard deviation was the smallest in B&H and in Croatia, which indicates there is a narrow range between the lowest and highest values of these variables. In comparison to B&H and Montenegro, Croatia and Serbia had the highest land productivity, where land productivity was doubly or triply higher. However, labour productivity was the highest in Montenegro, which was a bit more than in other countries but far greater than Albania, which has the lowest level of land productivity. From the standpoint of partial productivities in agriculture, labour and land productivity, there is a significant lag in the SEE countries in comparison with EU countries. The main problem behind relatively poor results in labour productivity is an unfavourable resource structure for agriculture, which indicates relative overemployment in agriculture. This is mostly a consequence of fragmented ownership and the slow development of the non-agricultural sector, which does not have the capacity to accept a surplus of labour from agriculture. Furthermore, Croatia had the highest GDP per capita during this period, while other countries recorded a far less average GDP per capita from 2005 to 2016. Finally, macroeconomic stability is measured by inflation, and Serbia had the highest average value in comparison to other countries. It can explain the fact that in 2005 inflation was as high as 16.12% and that there was a similar growth trend in Serbia in 2008 and 2011. In most countries, mean inflation was below 3%, and B&H had the smallest average inflation during this period.

Since this analysis estimates the impact of indicators on the level of comparative advantages for SEE countries, it was necessary to determine whether there was a significant difference between these countries during the period 2005–2016 (Table 6).

Table 6.

Test difference of level of comparative advantages – MANOVA TEST.

Dummy	W = Wilks’ lambda		L = Lawley-Hotelling trace
Dummy	P = Pillai’s trace		R = Roy’s largest root
Source	Statistic	F(df1)	F(df2)	F	p > F
W	0.2146	5.0	66.0	48.30	0.0000
P	0.7854	5.0	66.0	48.30	0.0000
L	3.6589	5.0	66.0	48.30	0.0000
R	3.6589	5.0	66.0	48.30	0.0000
Residual	66
Total	71

Source: The authors’ calculations.

The results from Table 6 show that there are significant differences in levels of comparative advantages for these countries during this period. The values of multivariate analysis of variance tests are <0.05, which indicates a significant difference in comparative advantages. To identify which determinants are important for comparative advantages in these countries, we included determinants such as land productivity, labour productivity, GDP per capita and macroeconomic stability measured by inflation. Also, a dummy variable was involved in empirical research in terms of country status in the WTO.

In Table 7, the results of the Levin–Lin–Chu test, the Breitung test and the Hadri LM test confirmed stationary of time series, which implies the rejection of the null hypothesis that panels contains unit root (LLC test and BT test), as well as acceptance of the null hypothesis that panels are stationary (Hadri LM test). The condition of stationary was satisfied at all variables with a statistical significance of 0.01, 0.05 and 0.1.

Table 7.

Panel unit root tests.

Variables	Number of panels	Levin–Lin–Chu test	Breitung test	Hadri LM test
RCA	6	−2.7198 (0.0033)	−1.4033 (0.0803)	−1.5890 (0.9440)
LAND	6	−2.5199 (0.0059)	−2.5534 (0.0053)	0.2065 (0.4182)
LABOUR	6	−10.2608 (0.0000)	−2.5245 (0.0058)	0.5054 (0.3066)
GDPpc	6	−7.4935 (0.0000)	−4.3005 (0.0000)	1.8521 (0.1320)
INF	6	−5.3081 (0.0000)	−1.9305 (0.0268)	−1.8696 (0.9692)
Levin–Lin–Chu test (H₀: Panels contain unit roots; H_a: Panels are stationary)
Breitung test (H₀: Panels contain unit roots; H_a: Panels are stationary)
Hadri LM test (H₀: Panels are stationary; H_a: Panels contain unit roots)

WTO: World Trade Organization; RCA: revealed comparative advantage; GDPpc: GDP per capita; INF: Inflation.

Source: The authors’ calculations.

To ensure an appropriately defined model, we involved fundamental prerequisites in econometric modelling. Table 8 includes tests of stationary, multicollinearity, heteroscedasticity and autocorrelation to determine model soundness.

Table 8.

Model validation.

Econometric preconditions of time series
Hadri LM test	Stationary
Hadri LM test	P value	0.9969
Variable	Multicollinearity	1/VIF
GDPpc	1.84	0.544145
LABOUR	1.81	0.553468
LAND	1.44	0.692677
Dummy	1.26	0.791000
INF	1.21	0.828956
Durbin Watson test	Autocorrelation	1.949
Breusch-Pagan/Cook- Weisberg test	Heteroscedasticity
	χ²(1)	0.00
	p > χ²	0.9610
White test	chi²(1)	22.61
White test	p > χ²	0.2551

GDPpc: GDP per capita; INF: inflation; VIF: variance inflation factor.

Source: The authors’ calculations.

Based on results from Table 8, the model is well designed in terms of econometric preconditions. An LM test shows a p-value = 0.9967, which means the data are stationary and which is one of the fundamental assumptions of an appropriate econometric analysis. According to the value of a variance inflation factor test, there is no problem of multicollinearity, where the value is far less than the reference value of 10. Based on DW (Durbin Watson) test value, it can be concluded there is no autocorrelation in residuals. Finally, the result of the BP/CT (Breusch-Pagan/Cook Weisberg) test shows there is no problem of heteroscedasticity because values 0.9610 and 0.2551 are more than 0.05, and the null hypothesis that supposes homoscedasticity cannot be rejected.

Based on the POLS model (Table 9), land productivity, labour productivity and inflation had a positive impact on the RCA, while GDP per capita negatively affects the RCA. Likewise, the model shows that these variables had a significant impact on the dependent variable, while only the dummy variable did not have a statistically significant influence on the RCA. Although there is no significance for the dummy variable, positive effects show that membership in the WTO enhanced these countries’ RCA. Model validation is presented by F results and reflects the statistically significant impact of most of the factors on RCA (p value <0.05). The model explains 62.12% of explanatory variables variations, and model reliability was confirmed by tests of multicollinearity, heteroscedasticity and autocorrelation.

Table 9.

POLS model estimation.

Variable	Coef.	Std. Err.	t	p > t	95% Conf. Interval
LAND	0.291465	0.0689459	4.23	0.000	0.15381	0.805675
LABOUR	0.657967	0.0739807	8.89	0.000	0.5102604	0.805675
GDPpc	−0.323704	0.0965327	−3.35	0.001	−0.516438	−0.1309705
INF	0.0280717	0.0100109	2.80	0.007	0.0080843	0.0480592
Dummy	0.0987637	0.0732812	1.35	0.182	0.0080843	0.2450743
C	−4.476608	0.7520004	−5.95	0.000	−5.978025	−2.975191
Source	SS		Df	MS
Model	7.35685334		5	1.47137067
Residual	4.48564928		66	0.06796438
Total	11.8425026		71	0.16679581
Number of obs						72
F(5, 66)						0.0000
R ²						0.6212
Adj R ²						0.5925
Root MSE						0.2607

Source: The authors’ calculations.

Discussion

Looking at the effect intensity, labour and land productivities caused the highest change in the RCA, where a 1% increase in these determinants raised the RCA by 0.66% and 0.29% respectively (Table 9). The results confirmed the fact that partial productivities in agriculture are important indicators for shaping SEE agri-food competitiveness. Higher production and productivity are crucial for meeting growing demand for food and non-food agricultural products, while an increase in agricultural output and productivity drives up agricultural incomes and improves competitiveness for this sector. Improving competitiveness in less-developed economies like those of the SEE countries, where a considerable proportion of the rural population still depends on agriculture, is very important for maintaining the viability of rural areas and reducing the poverty gap between urban and rural populations (Lovre, 2016).

Interestingly, the inflation rate had the opposite effect of what had been predicted. According to results of the model, macroeconomic stability did not have a significant influence on agri-food competitiveness, and GDP per capita had a negative influence on agri-food competitiveness. This can be explained by the example of small European countries with lower levels of GDP per capita as in SEE economies, which can be successful on the global agri-food market. Furthermore, these results are in line with Jambor and Babu (2016) who concluded that in most regions GDP per capita is negatively related to agricultural competitiveness. Additionally, these authors concluded that in Europe all these determinants (except labour productivity) have a significant influence on the agri-food sector’s competitiveness. Land productivity and endowment along with WTO membership have a positive influence on agri-food competitiveness for European countries, while GDP per capita, PSE and tariffs are negatively related to competitiveness, which means that agri-food sectors in small European countries with low GDP per capita can also have successful global competitiveness.

Since there are a limited number of studies analysing determinants of competitiveness for the agri-food sector of SEE countries, this article will contribute to filling this gap in the literature. More specifically, previous research on this topic focused mainly on levels of competitiveness or agriculture’s economic performance in a specific SEE country (e.g. Bojnec and Ferto, 2009; Bojnec and Ferto, 2010; Lovre, 2016; Matkovski et al., 2016; Nikolić et al., 2011), without finding factors that affect agri-food competitiveness. Bojnec and Ferto (2010) compared SEE countries’ agri-food trade specialization patterns of with those of the EU. The authors concluded that, with the exception of Serbia, Montenegro and North Macedonia, the rest of the SEE countries had an increasing food deficit with the EU-15, and the exports of SEE countries are highly concentrated on bulk raw materials. Also, their results indicate that EU integration is important for trade specialization. After analysing the export performances for agricultural sectors in the Western Balkan countries, Matkovski et al. (2016) concluded that, even though export performances were poor, comparative advantages in the majority of these countries were at a satisfactory level.

The results of this research clearly highlight how important partial productivity in agriculture is for the level of agri-food competitiveness. By considering the clear influence partial productivity in agriculture has on the comparative advantages of agri-food products in the SEE countries, it appears that countries with lower levels of agricultural productivity also achieve lower levels of comparative advantages. Also worthy of note is that Serbia has the highest level of RCAs for agri-food products and, at the same time, has achieved above-average levels for both labour and land productivity in comparison to the SEE region average. However, Albania has the lowest level of labour productivity due to overemployment in agriculture, and consequently it has not achieved comparative advantages in this sector. The situation in other SEE countries is very similar, and the levels of partial productivity in agriculture that were analysed correspond to agri-food products’ levels of RCAs. The difference in labour productivity among the SEE and EU countries is due to the fact that EU countries have high intensity production and therefore achieve a level of agricultural production per number employed in agriculture that is several times higher. The low level of labour productivity in agriculture in the SEE countries can be interpreted as a result of a surplus of workers in this sector, covert unemployment, and a predominance of small agricultural holdings with semi-natural production. Because of this, the agricultural sector should not be a source of additional employment, and employment could possibly be found the field of livestock production, which would most likely produce a positive effect on the level of production intensity.

Conclusions

Considering that competitiveness is an essential issue in the EU agri-food market, SEE countries should focus on improving competitiveness in the agri-food sector so as to be able to withstand market pressure from the EU. Improving agri-food competitiveness will be a major challenge for policy makers in all SEE countries, and especially for those that are still in process of European integration. The focus will be on resolving issues inherited from the past that are still creating burdens for these countries’ agricultural sectors. In the plant production sector, the level of competitiveness is not particularly low, but improving storage capacities, transport infrastructure and irrigation systems, as well as better organization of farmers would certainly be beneficial. A much larger problem is the livestock sector, which is predominantly concentrated in small holdings, where it is far more difficult to achieve adequate production and meet required standards. Also, the creation of a competitive food industry would enable SEE countries to reduce their dependence on exporting raw materials, and instead export products with higher added value. The whole process involves multitrack action, ranging from budget support and structural transformation, to large-scale improvement of agricultural institutions. In other words, building a modern agribusiness sector should be an imperative for future agricultural policy in all SEE countries.

Although the framework in this article is conceptually innovative in identifying determinants that influence agri-food competitiveness, there are some limitations. Namely, the choice of variables is highly significant, because choosing different variables will yield different results, and often the data for certain variables in these countries is problematic. Furthermore, it is possible to measure the same variables in several different ways. Because of that, this article enables a better understanding of agri-food comparative advantages in SEE countries and identifies factors that determine them for these countries, but a more complex set of factors determine these countries’ competitive positions on the international market.

Finally, these results provide informational support to policymakers concerning agri-food competitiveness and which factors are significant for this sector’s competitiveness in SEE countries. Future research should focus on analyses of EU countries in terms of agri-food competitiveness levels and comparisons with SEE countries. Additionally, future research should determine which segments of the agri-food market in SEE countries are the most competitive, and should define directions that would help policymakers to improve those segments that are not sufficiently competitive.

Footnotes

Authors’ note

This paper represents a part of the research on the project of the Ministry of Education, Science and Technological Development, Republic of Serbia, No. 46006, entitled ‘Sustainable Agriculture and Rural Development in terms of the Republic of Serbia strategic goals implementation within Danube region’.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Bojan Matkovski

Note

References

Balassa

(1965) Trade liberalization and revealed comparative advantage. The Manchester School of Economic and Social Studies 33:99–123.

Balogh

Jambor

(2017) The global competitiveness of European wine producers. British Food Journal 119:2076–2088.

Bjelić

(2014) Building competitiveness and increasing trade potential in the western balkans: Economic policy making in preparing for European integration. In Margo

Bojcic-Dzelilovic

(ed.) Public Policy Reform in the Western Balkans: Case Studies of Actors, Networks and Coalitions. Netherlands: Springer Svienve and Business Media B.V, pp. 69–89.

Bojnec

Ferto

(2010) Southeastern European agrofood trade specialization. Eastern European Economics 48(3):22–51.

Bojnec

Ferto

(2009) Agro-food trade competitiveness of Central European and Balkan countries. Food Policy 34:417–425.

Bojnec

Ferto

(2015) Agri-food export competitiveness in European Union Countries. Journal of Common Market Studies 53:476–492.

Bojnec

Ferto

(2017) The duration of global agri-food export competitiveness. British Food Journal 119:1378–1393.

Bojnec

Ferto

(2018) Drivers of the duration of comparative advantage in the European Union’s agri-food exports. Agricultural Economics - Czech 64:51–60.

Central European Free Trade Agreement (CEFTA) (2018) CEFTA Parties. Available at: http://cefta.int/cefta-parties/ (accessed 27 July 2018).

10.

Dragutinović-Mitrović

Bjelić

(2015) Trade regimes and bilateral trade in the EU enlargement process: focus on the Western Balkans. Acta Oeconomica 65:249–270.

11.

Eberlin

Ludvig

Dzimrevska

, et al. (2014) Objectives and approach. In Erjavec

Volk

Mortensen

(ed.) Agricultural policy and European integration in Southeastern Europe. Budapest: Food and Agriculture Organization of the United Nations, pp. 3–8.

12.

Erjavec

Mortensen

Volk

, et al. (2014) Gap analysis and recommendations. In: Erjavec

Volk

Mortensen

(ed.) Agricultural Policy and European Integration in Southeastern Europe. Budapest: Food and Agriculture Organization of the United Nations, pp. 39–60.

13.

EU Commission (2009) European Competitiveness Report 2008. Brussels. Available at ec.europa.eu/enterprise/newsroom/cf/_getdocument.cfm?doc_id (accessed 10 September 2018).

14.

European Commission (2018) EU Enlargement - State of play, Available at: https://europa.eu/newsroom/highlights/special-coverage/enlargement_en (accessed 27 July 2018).

15.

FAOSTAT (2018) FAO Database. Available at: http://www.fao.org/faostat/en/#data (accessed 20 August 2018).

16.

Ferto

(2008) The evolution of agri-food trade patterns in Central European countries. Post-Communist Economies 20:1–10.

17.

Ignjatijević

Matijašević

Milojević

(2014) Revealed comparative advantages and competitiveness of the processed food sector for the Danube countries. Custose@gronegocio 10:256–281.

18.

Jambor

Babu

(2016) Competitiveness of Global Agriculture - Policy Lessons for Food Security. Switzerland: Springer International.

19.

Lafay

(1992) The Measurement of Revealed Comparative Advantages. International Trade Modeling. London: Chapman and Hill.

20.

Lovre

(2016) Technical change in agricultural development of the Western Balkan countries. In: Tomić

Lovre

Subić

, et al. (eds) 152nd EAAE Seminar- Emerging technologies and the development of agriculture. Novi Sad: Serbian Association of Agricultural Economists, Faculty of Economics in Subotica, Institute of Agricultural Economics, pp. 1–14.

21.

Matkovski

Djokic

Zekić

(2016) Export performances of agricultural sector of the Western Balkan countries. In: Tomić

Lovre

Subić

, et al. (eds) 152nd EAAE Seminar - Emerging Technologies and the development of agriculture. Novi Sad: Serbian Association of Agricultural Economists, Faculty of Economics in Subotica, Institute of Agricultural Economics, pp. 294–303.

22.

Matkovski

Lovre

Zekić

(2017) The foreign trade liberalization and export of agri-food products of Serbia. Agricultural Economics - Czech 63:331–345.

23.

Matkovski

Zekić

Savić

, et al. (2018) Trade of agri-food products in the EU enlargement process: evidence from the Southeastern Europe. Agricultural Economics - Czech 64(8):357–366.

24.

Nikolić

Bajramović

Ognjenović

, et al. (2011) SEE trade liberalization - New opportunity for B&H agribusiness? British Food Journal 113:78–95.

25.

Nowak

Kaminska

(2016) Agricultural competitiveness: the case of the European Union countries. Agricultural Economics – Czech 62(11):507–516.

26.

Organization for Economic Cooperation and Development (OECD) (1992) Programme on Technology and the Economy. Paris: Organization for Economic Cooperation and Development.

27.

Torok

Jambor

(2013) Agri-food trade of the new member states since EU accession. Agricultural Economics – Czech 59:101–112.

28.

UN Comtrade Database (2018). Available at: http://comtrade.un.org/data (accessed 25 August 2018).

29.

World Bank (WB) (2018) World Development Indicators. Available at: http://databank.worldbank.org/data/reports.aspx?source=world-development-indicators (accessed 7 September 2018).

30.

World Economic Forum (2015) The global competitiveness report 2014–2015. Cologny: World Economic.

31.

World Trade Organization (WTO) (2018) International Trade Statistics. Available at: https://www.wto.org/english/res_e/statis_e/statis_e.htm (accessed 4 September 2018).