The impact of regional trade deals on North Macedonia’s wine exports

Grapes have a deep-rooted tradition and play a crucial role in the Macedonian economy. Viticulture, combined with wine production, contributes approximately 17%–20% of the country’s total agricultural production. Wine is the leading export beverage and the second most significant agricultural export, after tobacco. The sector supports: 12,000 seasonal workers, 2,500 full-time employees, 21,400 family agricultural households and 52 wine companies (Wines of Macedonia, 2024). In 2023, North Macedonia exported $63.8 million worth of wine, ranking as the 27th largest wine exporter worldwide (The Observatory of Economic Complexity (OEC), 2026).

The country is divided into three primary wine regions that cover 16 wine districts. The geographical presence of wineries is located in the Central Region, along the banks of the Vardar river. The so-called Central Wine Region represents about 83% of the total production. The Western Wine Region (Pelagonia-Polog) accounts for 13% of total production. The lowest production is evident in the Eastern Wine Region (Pchinja-Osogovo) with a contribution of 4% of total production. Wine varieties are represented with 84% and table wines with 16% (Ministry of Agriculture, Forestry and Water Economy of the R. N. (Wines of Macedonia, 2023).

According to data from the International Organisation for Vine and Wine (OIV), North Macedonia produces approximately 800,000–1,000,000 hl of wine annually, which is equivalent to 80–100 million litres. Despite its relatively small geographical size, North Macedonia holds a leading position in wine production among the Western Balkan countries, producing the largest amounts of wine in the region.

The Macedonian wine sector benefits from a network of multilateral and bilateral trade agreements, which facilitate access to key international markets and support the growth of wine exports, such as the EU Agreement, the European Free Trade Agreement (EFTA), bilateral agreements with Ukraine and Turkey and the CEFTA-2006 Agreement.

As part of the CEFTA-2006 agreement, Macedonian wine is exported without taxes to member countries, to Serbia, Bosnia and Herzegovina, Albania, Montenegro, Moldova and Kosovo ^[1]. In addition, for Macedonian exports to Croatia (former CEFTA-2006 member), a duty-free quota of 67,500 hl was agreed upon (Ministry of Agriculture, Forestry and Water Economy of the Republic of North Macedonia, 2023a). This agreement has had a positive impact on the export of Macedonian wine, particularly bottled wine, to the member countries of the CEFTA-2006 agreement. By adhering to the principles, rules and procedures of the World Trade Organization and aligning its policies with EU legislation, CEFTA-2006 serves as an effective tool for its members to accelerate their European integration agenda.

The aim of this paper is to assess the impact of the CEFTA-2006 agreement on North Macedonia’s international wine trade. Section 2 provides an overview of the country’s wine industry and its role in global markets. Sections 3 and 4 analyse Macedonian wine exports to the EU and CEFTA-2006 markets, which represent the country’s main export destinations. The following section reviews the relevant literature, while Section 6 presents the empirical methodology, applying a pooled exporter–importer–year regression estimated using Poisson pseudo–maximum likelihood (PPML) to evaluate the significance of regional markets. Finally, the Discussion section offers comments and policy recommendations based on the findings.

The wine sector has received renewed academic attention due to structural changes in global consumption patterns and increasing quality segmentation. In a comprehensive analysis of global wine trade, Gouveia et al. (2018) show that regional integration enhances both export volumes and unit values, suggesting a shift towards higher-quality market positioning after trade liberalisation. Their findings indicate that regional trade agreements (RTAs) can act as catalysts for quality improvement, especially in emerging wine-producing countries.

Recent evidence also confirms that proximity-based regional markets remain critical for wine exporters. Bao et al. (2025) and Serrano et al. (2025) used a sector-specific gravity model, finding that regional agreements significantly mitigate trade frictions in wine compared to multilateral liberalisation alone. This effect is stronger for bottled wine, where branding, distribution networks and consumer familiarity play a central role.

The growing empirical literature documents that RTAs generate substantial trade-creation effects, though the magnitude and distribution of benefits vary between members and sectors. Using structurally grounded gravity models estimated with PPML, Baier et al. (2019) demonstrate that modern RTAs significantly increase bilateral trade flows, with heterogeneous impacts depending on the production structure and initial trade intensity. Similarly, Heid et al. (2021) show that regional integration disproportionately benefits smaller economies when agreements reduce fixed trade costs and enhance market access predictability.

Recent literature on RTAs increasingly combines gravity trade models with heterogeneous firm frameworks, emphasising that exporters respond unevenly to trade liberalisation. Building on Melitz-type models, this research highlights that only sufficiently productive firms can overcome fixed export costs, implying that RTA benefits vary systematically with firm size, productivity and global value chain (GVC) integration (Melitz, 2023). Empirically, modern gravity models with exporter-time, importer-time and dyadic fixed effects have become standard tools to identify such heterogeneous effects within agreements.

A growing body of empirical evidence at the firm level confirms that exporter characteristics strongly shape RTA outcomes. Neri-Lainé et al. (2023) show that deep RTAs, those extending beyond tariffs to regulatory cooperation and behind-the-border measures, generate larger export gains for bigger, more productive firms and firms embedded in GVCs, while smaller firms benefit less or even face adverse competitive effects. Similarly, Cernat et al. (2025), analysing the Comprehensive Economic and Trade Agreement between the EU and Canada (CETA), find that although the agreement increased the number of exporting firms, the intensity of export growth varied substantially between firms and sectors.

Gravity-based studies further reveal asymmetric effects within RTAs, indicating that gains are unevenly distributed even among members of the same agreement. Nagengast and Yotov (2025) demonstrate that a significant share of trade impact variation occurs within agreements rather than across them, reflecting exporter-specific advantages.

Several mechanisms underpin this heterogeneity. Agreement depth and content play a central role: deep RTAs disproportionately benefit exporters capable of meeting complex standards and exploiting regulatory harmonisation. Exporter heterogeneity is often sector- and destination-specific, shaped by comparative advantage, technological capability and destination market characteristics (Avesani, 2026).

The international wine trade is segmented by product form, notably bottled versus bulk wine, which corresponds to distinct export strategies. Bottled wine is typically associated with value-driven strategies, emphasising branding, quality differentiation and higher unit values, while bulk wine reflects volume-driven strategies, focused on cost efficiency, large quantities and downstream bottling or industrial use (del Rey and Mueller Loose, 2025). Recent trends point to market polarisation, with premiumization in bottled wine coexisting alongside resilient bulk trade flows.

Global trade data reveal persistent differences between these segments. Bottled wine accounts for roughly two-thirds of the global wine export value despite stagnating volumes, while bulk wine represents a substantial share of export volume, but contributes less to total value due to low unit prices. Trade theory explains this divergence through differential demand elasticity: Bottled wine benefits from brand loyalty and perceived quality, allowing exporters to sustain higher prices, whereas bulk wine competes in highly price-sensitive markets (Jakubowska and Sadílek, 2023).

Empirical evidence supports the relevance of product form in trade responses. Ridley et al. (2022), using a structural gravity model, find that bottled wine exports exhibit greater sensitivity to tariff increases than bulk wine during trade retaliation episodes, indicating that quality and branding shape exporters’ responses to trade costs. Although much of the literature remains aggregated, these findings underscore the importance of product segmentation.

Recent research on export strategies further reinforces this distinction. Carbone and Henke (2021) show that EU exporters that pursue higher quality positioning and target sophisticated destinations achieve stronger value-based export performance, consistent with value-driven bottled wine strategies. Complementary evidence from del Rey and Mueller Loose (2025) indicates that bulk wine exports have demonstrated volume resilience amid declining bottled volumes, acting as a stabilising component of global trade.

Recent studies focusing on Southeast Europe confirm that CEFTA-2006 has played a meaningful role in fostering intraregional trade, although sector-specific evidence remains limited. Tasić (2020) and Bartlett and Uvalić (2022) show that CEFTA-2006 has contributed to trade expansion among western Balkan economies, with stronger effects observed in the agro-food and beverage sectors. However, these studies largely rely on aggregate trade data and do not isolate product-level dynamics or exporter-specific gains.

More recent gravity-based analyses emphasise the need to account for heterogeneity among CEFTA members. The export gains from regional integration are unevenly distributed, with countries possessing established production capacity and regional brand recognition benefiting more strongly. This information is particularly relevant for North Macedonia, whose wine sector exhibits both scale advantages within the western Balkans and increased quality differentiation.

According to Miteva-Kacarski et al. (2022), North Macedonia has a comparative advantage in the period 2011–2020 in terms of its export to other countries that are part of the Open Balkan (OB) initiative, according to the trade balance index and the revealed comparative advantage export index (Balassa’s RCA) in products in the category of tobacco and beverages.

The literature lacks sector-specific exporter-level evidence on how CEFTA-2006 affects the results of the wine trade. Existing studies focus either on aggregate trade flows or examine wine trade at the global level without isolating the role of regional agreements in Southeast Europe. Moreover, few analyses explicitly test whether individual countries derive disproportionate benefits from RTA membership. Within the context of regional integration in southeast Europe, a lot of papers have examined the impact of the Central European Free Trade Agreement on trade flows. Early studies, such as Christie (2002), highlighted CEFTA’s potential as a stepping stone toward EU integration, while later research provided quantitative evidence of trade creation among its members (Bartlett, 2009; Kikerkova, 2011). CEFTA-2006 towards the creation of a regional economic area. More recent analyses confirm that CEFTA-2006 has contributed positively to intraregional trade but with heterogeneous effects across sectors and member states (Estrin and Uvalić, 2016; Tasić, 2020).

This study addresses these gaps by applying a sector-specific PPML gravity framework to bilateral wine exports and by introducing an exporter-specific interaction term for North Macedonia. In doing so, it contributes to the recent literature on RTAs by providing new evidence on how regional integration shapes trade performance in a high-value agri-food sector and by highlighting the asymmetric benefits of CEFTA-2006 among its members.

Domestic wine consumption in North Macedonia is relatively low, amounting to approximately 176,000 hectolitres (17.6 million litres) annually, equivalent to around 7–10 litres per capita per year. As a result, approximately 85% of total commercial wine production is intended for export as presented in Figure 1 (Ministry of Agriculture, Forestry and Water Economy of the Republic of North Macedonia, 2023b).

Using the unit value of export as an indicator, the value of exported wine in 2023 was almost three times higher than in 2001. This increase in the dollar-per-quantity ratio suggests a rise in the unit value of exported wine. Several factors may have contributed to this trend, including higher wine prices, increased exports of premium and high-value wines, currency fluctuations, preferential trade agreements or global demand. All these factors indicate greater visibility and competitiveness of Macedonian wine. Macedonia has been a member of the International Organisation of Vine and Wine (OIV, 2025) since 2004, which underscores the country’s commitment to international wine standards and practices. Currently, there are 51 members that create more than 88% of the total wine production (OIV, 2024) (official web page).

The top 10 countries for Macedonian export of bottled wine in 2023, according to the value expressed in euros, are as follows: Serbia, Croatia, Slovenia, Bosnia and Herzegovina, Russia, Montenegro, Kosovo, Latvia, China and Poland. These markets represent the most significant destinations for bottled wine, reflecting both regional proximity and strong trade relations, as well as the growing demand for quality wines from international markets such as China and Russia. The presence of neighbouring countries such as Serbia, Croatia and Slovenia emphasises the importance of regional trade and cultural preferences in wine consumption. In addition, the inclusion of countries such as Latvia and Poland highlights the expansion of bottled wine exports into eastern and central Europe.

On the other hand, when it comes to bulk wine exports, which typically involve larger quantities sold for further bottling or processing, the top 10 countries in 2023 ranked by value in euros are: Germany, Serbia, Slovakia, Austria, Bulgaria, Croatia, Poland, Montenegro, Greece and Slovenia, as presented in Figure 2. This list reflects a somewhat different export strategy, focused on supplying bulk wine to countries with developed wine industries or those engaged in re-export and bottling. Overall, these rankings highlight the dual strategy of wine exporters, which aim to reach both premium bottled wine markets and high volume bulk wine markets, depending on the specific demand structure, trade barriers and market opportunities in each destination country.

According to data from the International Trade Centre (ITC), Macedonian wine exports over the past five years have been predominantly directed toward EU member states and CEFTA-2006 countries. While exports to the EU market have decreased by 16%, exports to countries in CEFTA-2006 have shown significant and consistent growth, increasing by nearly 18% in 2023 compared to 2019. Therefore, Macedonian exporters are shifting their focus to regional markets in the Western Balkan countries, where demand is growing and it is more profitable. Exports to the rest of the world have also decreased, dropping from 6.2% of total exports in 2019–3.8% in 2023. Meanwhile, the EFTA market has remained insignificant throughout the observed period.

The EU is the biggest wine producer, accounting for around 60% of the global production (144.5 million hl in 2023). The EU continues to dominate wine exports, with France, Italy and Spain holding the top three positions in both value and volume. From the “new world”’, Australia, Chile and the USA are the main suppliers in terms of value (OIV, 2023).

Since 2001, the European Union has granted autonomous trade preferences to western Balkan countries under the Stabilisation and Association Agreement. However, products such as sugar, wine, baby beef and certain fisheries are subject to preferential tariff quotas. Products originating in the Western Balkans and covered in Chapters 7 and 8 of the Combined Nomenclature of the EU (fruits and vegetables) can be imported into the EU without custom duties and without quantitative restrictions or measures having an equivalent effect. However, exceptional preferences also include a global wine quota of 30,000 hl, which can be used by each country or territory after exhaustion of its national quota established in its bilateral agreement concluded with the EU (Regulation EU 2024/823, 2024).

Macedonia is the seventh wine import trading partner in the EU-27 countries. Germany and Croatia are the dominant trading partners, but their demand has been steadily declining. Germany remains the largest importer, but its demand has been steadily declining from $12.2 million in 2019 to $8.2 million in 2023. A similar trend can be observed in Croatia, where exports peaked at $9.6 million in 2021 before dropping to $7.1 million in 2023.

Austria saw the most significant increase in 2023 compared to the previous year, with exports rising to $1.59 million (+152%), highlighting growing demand (International Trade Centre, 2025). The significant drop in Slovakia raises questions about possible policy changes or changing market dynamics. According to Ristevska-Jovanovska et al. (2018), Macedonian wineries need to focus their export effort on the following EU markets: Germany, The Netherlands, Poland, Denmark and the Czech Republic which is presented in Figure 3.

Geographical indications (GI), often in the form of appellations of origin, have long been significant in the wine trade, in particular and among the 9,500 origin brands that are estimated to be in use globally, 2,535 are used for wines. Within the European Union, a major user of GIs, of the 3,500 GIs registered at the EU level, 1,627 are for wines. Not all GI products are equally successful, but the average value premium rate (i.e. the volume-weighted price premium for GI products compared to non-GI products) in all EU GI products was 2.07 in 2017. This indicates that the sales value of GI products was, on average, 2.07 times higher than that of comparable non-GI products. France leads the score with a value premium rate of 4.13. European GI products are mainly sold nationally (60% in terms of value and volume) or traded with other EU countries (23% in terms of volume and 20% in terms of value) (European Commission, 2021, 2023).

However, EU wine policy faces criticism for its complexity and strict regulations for third countries. Third-country producers must comply with EU standards on labelling, GIs and production methods, which can be costly and time consuming. They also face lengthy approval processes for the recognition of their wine production methods.

The system of geographical indicators is harmonised with the EU system of GIs, due to the outstanding issue between North Macedonia and the EU on the issue of the term “Macedonia”, by both sides (Regional Rural Development Standing Working Group in South Eastern Europe, 2022). The whole territory of the Republic of North Macedonia is the geographical area, in other words, a region that produces regional wine.

The legal framework (the Wine Law) is in accordance with the EU regulation on the common organisation of the wine market. According to the classification in the current wine law, more than 750 wines with protected geographical origin, that is, table wines with geographical indication (PGI), are 18%, wines with controlled origin are 70% and wines with controlled and guaranteed origin are around 12%. The new law for wine will allow reconstruction of the existing wine regions and create geographic labels for wines (as in the EU) according to Regulation 1308 / 2013, which is illustrated in the Table 1.

CEFTA-2006 or the Central European Free Trade Agreement, revised in 2006, is a regional free trade agreement that aims to eliminate customs duties and trade barriers among its members, harmonise rules in key areas such as competition policy, public procurement and intellectual property rights and encourage investment and regional cooperation. It also plays a key role in preparing the economies of member states for EU accession by aligning them with EU rules and practices.

Current members of CEFTA-2006 include Albania, Bosnia and Herzegovina, North Macedonia, Moldova, Montenegro, Serbia and Kosovo. Kosovo is excluded from the analysis, since there are no available data. Among these members, Moldova stands out as the 21st largest wine exporter in the world in 2023, with wine ranking as the country’s seventh most exported product. However, despite its membership in CEFTA-2006, Moldova’s main wine export destinations are not other CEFTA members, but primarily EU markets, such as Romania, Belarus, GA, Czechia and Poland. As mentioned above, North Macedonia produces approximately 800,000–1,000,000 hl of wine annually, making it the largest producer in the Western Balkan region, presented in Figure 4.

Figure 5 illustrates the comparison of wine export and import from western Balkan countries, where it is visible that North Macedonia and Moldova are two of the most wine-exporting countries and North Macedonia is the least wine-importing country. The main sources of wine imports from North Macedonia were Moldova ($545000), France ($534000), Italy ($524000), Spain ($185000) and Germany ($62.9000). In 2023, North Macedonia imported $1.96 million worth of wine, ranking as the 163rd largest wine importer globally.

The beverages sector is considered one of the few with the greatest potential to form regional value chains and increase intra-CEFTA trade, particularly in the production of wine and beer (Bjelić, 2016, p. 5). All CEFTA 2006 economies produce and export beer and wine and at the same time, they import the same products from other CEFTA 2006 economies. The import of wine from fresh grapes, including fortified wines (in USD), in the western Balkans shows that Serbia is the dominant importer in absolute terms. Serbia shows the highest absolute growth, increasing from 30.9 million US dollars in 2018–52.4 million dollars in 2023, confirming its position as the largest economy in the region. Serbia’s imports reached US$25.2m in 2022, marking an almost 70% increase compared to 2021. Beyond Serbia, other notable growth markets include Bosnia and Herzegovina, where demand increased by approximately $5m in 2022. On the contrary, Albania and Montenegro remain relatively insignificant trading partners.

As shown in Figure 5(a) and (b), the volume of wine exported to Western Balkan countries (measured in millions of litres) is on a downward trend, but its value (in millions of US dollars) is increasing. However, this decrease in quantity does not indicate weakening performance. On the contrary, it could suggest that North Macedonia is consolidating its competitive position by shifting towards the export of higher quality wines. This development underscores a strategic move from volume-driven to value-driven trade. These trends clearly demonstrate the growing competitiveness and recognition of North Macedonian wine, both within the Western Balkan region and across EU markets, opening opportunities for deeper trade integration and broader market diversification. This situation is presented in Figure 6(a) and (b).

Wine production in the Western Balkans accounts for less than 2% of global wine production, yet wines from the region are increasingly in demand on international markets. This growth has been driven primarily by expanding sales of bottled wine, which are beginning to overtake bulk wine exports, reflecting substantial improvements in quality in recent years alongside highly competitive price positioning (FAO, 2023). Building on this positive export momentum, the OB initiative creates additional opportunities to boost international trade by facilitating deeper regional integration, reducing administrative and non-tariff barriers and enabling producers to exploit economies of scale, coordinate market access strategies and strengthen the collective presence of the region in global wine markets.

The OB initiative originated as part of the Berlin Process, initially envisioned as an economic union that encompassed Macedonia, Serbia and Albania. Its primary objective is to enhance regional economic cooperation and facilitate the free movement of people and goods. Furthermore, it seeks to support the integration of the OB countries into the European Union (Rikalović et al., 2022). The OB initiative is a more recent and informal regional cooperation framework. Although it shares the goal of deepening regional integration, OB emphasises practical economic cooperation and mobility rather than a legally binding trade agreement like CEFTA-2006.

Another recent initiative is the Balkan Wines Association, which builds on the Memorandum of Cooperation that led to the establishment of the Balkan Wine Network in 2019. The Network was created to support the definition of strategic directions for the sustainable development of the wine industry, viticulture and wine tourism throughout the region. The initiative is supported by the Food and Agriculture Organization of the United Nations (FAO) and the European Bank for Reconstruction and Development (EBRD) (Bloomberg Adria, 2024).

Although tariffs on wine have declined globally, non-tariff measures (NTMs) remain a major constraint on trade (Anderson and Pinilla, 2022). NTMs remain pervasive, encompassing of: sanitary and phytosanitary (SPS) standards, technical barriers to trade and commercial standards (for instance, rules on labelling for health warnings about wine consumption or production methods), pre-shipment inspections (requirements on the characteristics of products for export enforced by a supervisory entity in the exporting country) and GIs for place-designated varieties. Santeramo and Lamonaca (2019) provide meta-analytic evidence showing that NTMs disproportionately disadvantage smaller exporters lacking regulatory capacity.

There are also significant barriers that hinder wine trade in the regional market. A survey conducted by the ITC in 2015 and 2016, based on more than 8,000 interviews across 26 sectors, aimed to identify trade barriers faced by EU companies exporting to key global markets, focusing on obstacles beyond tariffs. The challenges for the beverages sector in CEFTA 2006 economies are particularly numerous, with key issues including the nonrecognition of quality certificates, discriminatory excise duties and trade facilitation problems. The recommendations provided by private sector operators to enhance intra-CEFTA trade, especially for the beverages sector, address TF concerns and related measures primarily. These suggestions include harmonising standards, eliminating redundant testing, mutually recognising quality certificates (particularly for sanitation and phytosanitary measures, SPS), acknowledging the status of certified laboratories for wine inspections as listed in the EU Official Journal and implementing risk management strategies. Nontariff measures in the region contribute to the decreasing trend in the beverage sector trade (Petrović et al., 2018).

More recently, Disdier-Chauvin and Fernández-Olmos (2024) demonstrated that regulatory convergence within RTAs significantly reduces the trade-restrictive effects of NTMs in the agri-food sectors. Ridley et al. (2022) show that regulatory retaliation and compliance costs substantially reduce trade flows and consumer welfare, reinforcing the importance of stable regional frameworks that minimize policy uncertainty (Avesani, 2026). RTAs function not only through tariff elimination, but increasingly through institutional harmonisation and trade facilitation. Yotov (2022) finds that deep trade agreements, those that address standards, certification and regulatory cooperation, produce stronger and more persistent export gains than shallow tariff-based arrangements. These findings are particularly relevant for agrifood products, where compliance costs and administrative barriers often dominate traditional tariffs (Depetris-Chauvin and Fernández-Olmos, 2024). These findings suggest that RTAs such as CEFTA-2006 may be particularly valuable for wine exporters by providing predictable rules and mutual recognition mechanisms.

Empirical studies of international trade have increasingly used gravity-type models to quantify the effects of RTAs on bilateral trade flows. Originating from Tinbergen’s (1962) gravity equation, the approach has evolved to incorporate theoretical underpinnings from Anderson and van Wincoop (2003), who formalised the role of multilateral resistance terms in shaping trade outcomes. Since then, this framework has become the dominant tool for evaluating the impacts of TAs that create and divert trade between sectors and regions.

A central methodological development in this literature is the use of PPML estimation, introduced in the trade context by Santos Silva and Tenreyro (2006). Unlike ordinary least squares or log-linear estimations, PPML naturally accounts for heteroskedasticity in trade data and incorporates zero trade flows without the need for arbitrary transformations. This feature is particularly important in sectoral analyses such as wine exports, where many country pair-year observations record zero values due to specialisation and limited market coverage. Numerous applications, such as Head and Mayer (2014), Yotov et al. (2016) and Larch et al. (2019), have reaffirmed PPML as the preferred estimator for modern gravity models.

The wine industry, as a highly differentiated product sector with strong cultural and GIs, has received increasing attention in trade research. Studies such as Anderson and Pinilla (2018) and Balogh (2022) emphasise that wine exports are shaped not only by standard determinants such as income and distance, but also by RTAs, consumer preferences and branding linked to quality improvement. Evidence suggests that integration into regional and global markets can serve as a catalyst for quality improvement in the wine sectors, particularly in smaller producing countries seeking niche advantages.

Building on this literature, the present study contributes in two important ways. First, it applies a sector-specific PPML gravity model to assess the effect of CEFTA-2006 on wine exports, providing new evidence on how regional trade integration affects a product of strategic importance for smaller economies. Second, it introduces an exporter-specific interaction term for North Macedonia, allowing us to test whether the country benefits disproportionately from CEFTA membership relative to other signatories.

Following Yotov et al. (2016), our baseline model is a PPML gravity equation with three-way fixed effects:

In this specification, Y_eit is the wine export value from exporter e to importer i in year t; PairTreat_eit equals one when both the exporter and the importer are active CEFTA-2006 members in year t; MKD_e is an indicator for North Macedonia as exporter; ln GDP_it is the logarithm of destination GDP from the World Bank World Development Indicators. The three sets of fixed effects follow the structural gravity benchmark: exporter-time fixed effects (γ_et) absorb all time-varying exporter characteristics including outward multilateral resistance; importer-time fixed effects (δ_it) absorb all time-varying importer characteristics including inward multilateral resistance; and bilateral pair fixed effects (λ_ei) control for all time-invariant bilateral trade costs such as distance, common language and colonial ties. As a robustness check, Columns (3)–(4) re-estimate the model using separate exporter (γ_e), importer (δ_i) and year (τ_t) fixed effects – the specification used in earlier drafts of this paper. A further robustness check (Column 4) presents a North Macedonia-specific difference-in-differences model with importer and year fixed effects. All coefficients are semi-elasticities interpreted as percentage changes using 100×(exp(β̂)−1). Standard errors are HC1 heteroskedasticity-robust throughout.

We use an annual bilateral panel of wine export values at the exporter–importer–year level. The panel spans 1995–2022 and covers 32 countries: the EU-27 plus five Western Balkan CEFTA economies (Albania, Bosnia and Herzegovina, Montenegro, North Macedonia, Serbia). Zero trade flows are explicitly modelled rather than discarded: we complete the panel with zeros for all exporter–importer–year cells absent from the raw data, yielding 25,073 observations for the Yotov pooled models and 866 for the North Macedonia-specific model after singleton removal. Destination market size is proxied by the logarithm of importer GDP in constant USD. CEFTA-2006 participation is coded at the country–year level using documented adoption and exit dates; the dyadic CEFTA-2006 indicator equals one only in years when both the exporter and the importer hold active membership.

Table 2 presents the PPML estimates across four specifications. Column (1), our Yotov et al. (2016) baseline, shows that dyadic CEFTA-2006 membership raises bilateral wine exports by approximately 28% (coefficient 0.250, p < 0.01) – an effect consistent with the 30–100% range of trade-creation effects documented by meta-analyses of RTAs (Head and Mayer, 2014). Column (2) adds an interaction term between the CEFTA dummy and an indicator for North Macedonia as exporter: the mkd_x_pair coefficient of 0.500 (p < 0.01) implies a further + 65% country-specific gain for Macedonian exporters, yielding a combined MKD effect of approximately + 91%. The pair_treated coefficient in Column (2) attenuates to 0.145 and loses statistical significance, suggesting that the aggregate trade creation captured in Column (1) is largely concentrated in North Macedonia rather than uniformly distributed across all CEFTA-2006 member pairs. In the complementary North Macedonia-specific difference-in-differences (Column 4), the CEFTA indicator yields a coefficient of 0.846 (p < 0.01), corresponding to a + 133% increase in Macedonian wine export values.

Columns (3) and (4) present the robustness checks using a standard three-way specification with separate exporter, importer and year fixed effects. This less-restrictive structure yields substantially larger point estimates: Column (3) gives a CEFTA pair effect of + 2,837% (coefficient 3.380, p < 0.01) and a North Macedonia interaction of + 399% (coefficient 1.608, p < 0.01). The contrast between the Yotov baseline (Columns 1–2) and the standard specification (Columns 3–4) illustrates the well-known sensitivity of gravity estimates to the fixed-effect structure. The Yotov specification reduces omitted-variable bias by absorbing time-varying multilateral resistance through country-year dummies and all time-invariant bilateral trade costs through pair dummies. The simple fixed-effect estimates should therefore be interpreted as an upper bound and are presented primarily for comparability with earlier versions of this paper.

The North Macedonia-specific difference-in-differences model (Column 4) focuses exclusively on MKD bilateral wine flows, using importer and year fixed effects as controls. The estimated coefficient of 0.846 (p < 0.01) corresponds to a + 133% increase in wine export values upon CEFTA-2006 membership. This result is directionally consistent with the combined interaction effect in Column (2) and confirms that North Macedonia has derived substantial and statistically robust export gains from CEFTA-2006 participation.

Taken together, the four specifications consistently indicate that CEFTA-2006 membership raises bilateral wine export values. The Yotov et al. (2016) baseline – our preferred specification – yields a + 28% average effect and a + 91% combined effect for North Macedonia, both economically plausible and in line with the gravity literature. These findings confirm that CEFTA-2006 is a meaningful driver of wine trade expansion in the Western Balkans, with North Macedonia as a primary beneficiary.

The results underscore two key insights. First, CEFTA-2006 membership exerts a statistically significant and economically meaningful positive effect on intra-regional wine trade, confirming its role as a facilitator of export expansion among member states. Second, North Macedonia experiences an above-average benefit: approximately + 91% (Yotov baseline, combined effect) compared to the average member pair’s + 28%. This finding underscores the strategic relevance of CEFTA-2006 regional markets for the Macedonian wine sector, particularly given the observed contraction of exports to the European Union in recent years.

The findings provide robust evidence that CEFTA-2006 membership has a significant and positive effect on wine exports within the region. The pooled PPML regression using the Yotov et al. (2016) baseline confirms the trade-creating role of RTAs: the average effect of CEFTA-2006 membership on bilateral wine exports is approximately + 28% (p < 0.01) – a magnitude within the 30%–100% range documented by Head and Mayer (2014) in their meta-analysis of gravity models. Under a less-restrictive fixed-effect structure (the robustness specification), the estimated effect is substantially larger (+2,837%), underscoring the importance of controlling for multilateral resistance as recommended by Yotov et al. (2016). The preferred Yotov estimates therefore represent a conservative and credible lower bound for the trade-creation effect of CEFTA-2006 in wine.

An important contribution of this analysis is the identification of North Macedonia as a relative winner within CEFTA-2006. The interaction term in Column (2) of Table 2 shows that Macedonian exporters obtain an additional + 65% gain relative to the average CEFTA-2006 pair (p < 0.01), yielding a combined effect of approximately + 91% for North Macedonia. The North Macedonia-specific difference-in-differences model corroborates this finding with an estimated + 133% effect (p < 0.01), confirming the robustness of the result. These magnitudes suggest that Macedonian producers possess comparative advantages in wine production – such as cost efficiency, established distribution channels or brand recognition within the Western Balkans – which allow them to capitalise more effectively on the preferential market access provided by CEFTA-2006.

The study demonstrates that CEFTA-2006 has produced meaningful and statistically robust trade gains in the wine sector, with North Macedonia standing out as a particularly strong beneficiary. The Yotov et al. (2016) specification places the average trade-creation effect at approximately + 28% and the North Macedonia-specific combined effect at + 91%, both consistent with the broader gravity literature. These results contribute to the debate on the effectiveness of RTAs in fostering regional integration and economic development and they provide practical insights for policymakers seeking to strengthen the competitiveness of export-oriented industries in the Western Balkans.

To fully unlock its potential, a comprehensive and multidisciplinary approach is required, including the development of wine routes and events, the education and training of hospitality staff, the strengthening of wine diplomacy and regional cooperation, as well as the more effective use of digital marketing. Expanding the cultivation of unique local varieties (Vranec, Temjanika, Kratoshija, Smederevka, Stanushina) will allow the production of distinctive wines that can secure niche markets and strengthen competitiveness. Such a strategy aligns with the growing global demand for authentic, origin-based products and experiential consumption, particularly in premium and super-premium wine segments. North Macedonia’s rich winemaking heritage, diverse grape varieties and favourable wine-to-price ratio create a strong foundation for producing high-quality wines with GIs. These certifications attract global consumers seeking premium wines, enhancing market value and international recognition. At the same time, stronger institutional support for GI protection and enforcement within regional and EU markets could further amplify the positive trade effects identified in this study. To remain competitive, North Macedonia must focus on differentiating its exports through high-quality products, strong branding and an emphasis on indigenous origins.

By addressing these areas, the Macedonian wine sector can build resilience, take advantage of new opportunities and reinforce its position in the global market.

There are three the most important study implications: The study contributes to the international trade literature by applying a PPML pooled exporter-importer-year framework that explicitly accounts for zero trade flows. This strengthens the empirical foundations of gravity-type models by demonstrating how RTAs can be robustly evaluated without discarding non-trading dyads, which are economically meaningful observations. By improving the precision and inclusiveness of empirical trade analysis, the study indirectly supports more evidence-based policymaking, which is essential for designing trade policies.

Beyond estimating the average effect of CEFTA-2006, the model introduces an interaction term that isolates North Macedonia’s exporter-specific gains. This enriches the theoretical understanding of RTAs by showing that the benefits are not homogeneous between members but can vary significantly depending on country characteristics, production specialisation and market integration strategies. By focusing on a single product sector (wine), the study provides evidence that CEFTA-2006 has been strongly trade-creating within the bloc. Although not directly measuring diversion, the results contribute to the theoretical debate by suggesting that RTAs can amplify intra-regional flows in ways that may reshape comparative advantage dynamics among smaller economies. Such differentiated outcomes are particularly relevant for public attitudes toward regional integration, as visible sector-specific gains can strengthen societal support for cooperation frameworks like CEFTA-2006, especially in smaller economies that often perceive themselves as asymmetrically positioned in international trade.

These trade-creation effects are particularly relevant for the North Macedonia wine industry, where extensive rural outmigration has already reduced vineyard acreage and intensified labour shortages. In this context, expanding access to export markets becomes not only a trade outcome, but also a mechanism to strengthen rural employment and stabilise household income, given the substantial direct and indirect job creation of the sector. Improved income stability in rural areas can positively affect quality of life by reducing economic insecurity, slowing depopulation trends and preserving local communities, traditions and cultural landscapes associated with viticulture. To fully realise these benefits, sustained support for young farmers, continued modernisation of viticulture and more effectively designed state incentives remain essential. Such policies may also influence the social perceptions of agriculture as a viable and future-oriented profession, particularly among younger generations.

At the same time, the results invite reflection on potential limitations. While CEFTA-2006 clearly fosters intraregional trade, it may also contribute to trade diversion away from extraregional markets. Given the observed decline in North Macedonia’s exports to the EU, the country’s increasing reliance on CEFTA-2006 partners raises questions about long-term market diversification and resilience. Furthermore, the analysis focuses on trade values, which, while informative, do not capture possible changes in export quality, product upgrading or market concentration. Future research could extend the framework to examine whether CEFTA-2006 not only increases export volumes, but also shapes the structure and competitiveness of the wine industry, thus influencing long-term welfare outcomes, consumer perceptions of domestic products and the broader social sustainability of export-oriented growth.

This paper is co-funded by the Western Balkans–Visegrad Fellowship, under the project “Characteristics of the EU Wine Market”.