THE WORLD BANK SERBIA Country Economic Memorandum Serbia’s New Growth Agenda Increasing Exports By Jaime Frias, Bojan Shimbov, Elwyn Davies, and Colin Ek 0 Export acronyms 7STEE 7 small transition countries of Eastern Europe that are already EU members ECI Economic Complexity Index EPA Export promotion agency EU European Union EXPY Sophistication of exports FDI Foreign direct investment GVA Gross value-added GVC Global value chain M&E Monitoring and evaluation M&EE Machinery and electrical equipment MSMEs Micro-, small, and medium-sized companies NTMs Non-Tariffs Measures NQI National Quality Infrastructure PRODY Individual Product Index R&D Research and development R&P Rubber and plastics SBRA Agency for Business Registers Serbia SDPs Supplier development programs SITC Standard International Trade Classification TPOs Trade promotion organizations UN United Nations UNCTAD UN Conference on Trade and Development WDI World Development Indicators WEF World Economic Forum WITS World Integrated Trade Solution, World Bank 1 Introduction Evidence from several countries reveals that nations that have open economies tend to enjoy higher income than those with closed economies (Lind and Ramondo 2018). Openness to hosting multinational firms can lead to firms in receiving countries acquiring new technology and skills (Harrison and Rodriguez-Clare 2010), and to productivity-enhancing spillovers, particularly through vertical commercial relationships between foreign and domestic suppliers. Learning by exporting offers positive knowledge externalities, and it comprises myriad ways in which exports can stimulate growth in productivity, including development of exporter capabilities, such as marketing new products; upgrading product quality; and acquiring expertise in managing customer relationships by dealing with foreign buyers. The value from knowledge spillovers and the promise of job creation are often seen as positive externalities and are usually brought in to justify policy interventions in the form of tax incentives, grants, and other concessions (access to land sites at minimal or low cost). It is often thought that spillovers from foreign firms are driven by transfers of technology and by learning about markets by exporting. Learning from foreign buyers is supposed to be channeled directly to the exporters or passed through to local suppliers and competitors in domestic markets. There is some evidence that in Serbia, the international competitiveness of domestic exporters has been diminishing, and government programs to support links with markets receive meager financial allocations. Recent successes in exports of manufactures have revealed the great potential of Serbia, but these have been driven by only a few firms, many of them foreign-owned. This has lowered expectations of inclusive and widespread growth. There is also a growing sense that government efforts to promote exports and attract export-oriented investment in Serbia have instead been directed to attracting foreign direct investment (FDI) at the expense of export promotion, which has not been particularly effective. A looming question has become whether the current policy mix for promoting competitive Serbian exports needs realignment. In addition, costly inefficiencies may arise when market incentives are used to attract foreign investors and multinational firms without dealing with fundamental structural issues that may had been preventing them from coming in the first place. A recent note on the effects of FDI on local suppliers in Serbia suggests that local firms can only benefit from knowledge spillovers if they already have a minimum level of absorptive capacity (Brussevich and Tan, 2019). Furthermore, incentives to attract foreign exporters could be inducing firms to turn their attention to generating cost advantages and thus investing in labor- intensive industries rather than in technological upgrades or diversifying into high-value-added economic activities. These concerns can become pronounced if disruptive global trends in automation, digitization, and labor-replacing technologies appear in Serbian export manufacturing, and services. Location decisions are often informed by some critical factors that present complementary conditions for demanding foreign operators: engineering skills, managerial talent, investor protection, and the efficacy of contracts and intellectual property rights. But not only are complementary public inputs needed; distortions must be eliminated, for example in the provision of credit, unequal enforcement of regulation, and regulatory constraints on labor. In what follows, we review Serbia’s recent achievements in exports in the context of its regional and aspirational peers. We then present the findings of an assessment of conditions for exporters operating in Serbia, whether they are foreign or domestic, public or private. We end by presenting suggestions of a few areas where Serbian policy can best begin to address problematic local conditions and discuss how to reinforce productive policies for enhancing exports from Serbia. Although the crux of this note is analysis of Serbia’s exports of goods, it does touch on exports of services. The rationale is the much-needed attention to recent gains in manufactured exports (i.e., transport and automotive and electrical equipment); the necessity of keeping the scope of analysis 2 manageable, and the more extensive data on goods exports, mainly from the customs authority. Many of the findings of this analysis also apply to exports of services. The regional peers chosen are the other five countries in the Western Balkans region: Albania, Bosnia and Herzegovina, Kosovo, North Macedonia, and Montenegro. The aspirational peers are Bulgaria, Croatia, Estonia, Latvia, Lithuania, the Slovak Republic, and Slovenia (7STEE)—the seven small transition countries of Eastern Europe that are already members of the European Union (EU). I. Despite Notable Export Achievements, Serbia Still Trails Aspirational Peers By many measures, for the last two decades Serbian exports have performed well . Exports of goods and services in 2018 topped US$26 billion and represented 51 percent of GDP (Figure 1). In 2017 Serbia’s GDP grew at 1.9 percent, barely above its 1.4 percent average annual growth rate for the last 10 years; however, exports grew by 12.1 percent. Figure 1. Exports and Economic Growth, 2007–17, Percent Source: World Bank, World Development Indicators database. Most of Serbia’s trade is conducted with the EU countries and it has risen considerably in the last decade. The EU-28 as a group is the country’s largest trading partner, accounting for almost two-thirds of its goods exports in 2018 (Annex 3). Since 2006 Serbia has managed to triple its exports to the EU-28 even though EU-28 imports have been largely stagnant. As a result, Serbia’s share in EU-28 imports of goods accounted for over 0.5 percent in 2018, having doubled since 2006 (Figure 2). 3 Figure 2. Serbia’s Exports to the EU and EU Imports, 2006–17, Percent Source: UN Comtrade data and World Bank staff calculations. Serbia is also exporting more products, especially from the automotive industry. In 2016, formation of new industrial clusters in the machinery and transport sector emerged at the core of the product space network, driven by growth in automotive production and products integrated into car manufacturing (Figure 3). The density of clusters of machinery and transportation manufactured goods have gone up since 2006; rotating electric plants and electric motors, generators, and generating sets stand out as new types of export goods (Box 1). Figure 3. Evolution of Serbia’s Product Space, 2006 to 2016 Source: Observatory of Economic Complexity, OEC, MIT 4 Box 1. Machinery and Electrical Equipment The machinery and electrical equipment (M&EE) sector has been restructuring for years. However, the process is nearly finished, and the outlook is positive based on a committed export orientation, a legacy of valuable skills, and growing demand in the form of near-shoring from European countries. In 2015 M&EE created 1.2 percent of Serbian GDP and 7.4 percent of added value in the processing industry, making it the fourth most productive industrial sector, just behind metallurgy. Along with its importance in value creation and export growth, it has relevance for Serbia’s technical-technological and social development. M&EE is often referred to as a “key enabling sector” because the skills and knowledge of its workers offer positive spillovers for other sectors. In 2015 M&EE employed over 30,000 people, many in family-owned micro and small companies, in many regions, some of them underdeveloped. However, most growth in M&EE exports is driven by large companies. The top 12 exporters in 2015, which were all foreign-owned (10 greenfield and 2 brownfield), accounted for over 70 percent of total M&EE post-crisis export growth. Of the 25 major exporters in Serbia in 2015, the 4 that were domestically owned are all former state-owned enterprises SOEs that were privatized. Exporting by domestic companies was highly diversified in terms of number of companies, export markets, and products. In 2015, 720 M&EE companies were registered, and most were micro-, small, and medium-sized companies (MSMEs), often family-owned. About 83 percent of SMEs were exporters in 2015, reaching a wide range of European markets and covering 149 product groups, according to the 4-figure SITC classification. Popular SME products exported were customized finished machines and equipment for food industry manufacturing. While Serbia has demonstrated that it has the knowledge and skills to successfully export a broad range of products and technologies, it seems to find it difficult to expand production because of a shortage of qualified staff. A qualified labor force is a critical success factor, and manufacturing requires different types of labor. Asked about the greatest limitation in responding to swift growth in demand, most companies cited finding a sufficient number of appropriately skilled workers. Growth in exports between 2006 and 2016 was accompanied not by growth in the number of jobs related with exporters but by a change in the composition of employment . Among registered exporting firms, employment has been flat, with a negative growth rate of 0.2 percent and a loss of 17,032 jobs: SOEs and newly privatized exporters shed 111,728 jobs, but new foreign and domestic exporters employed 96,097 more workers. This period saw significant privatization of SOEs, which is often associated with restructuring and loss of excess capacity in the form of jobs (Figure 4). 5 Figure 4. Employment in Exporters by Ownership Type, 2006–16 Number of employees CAGR Difference 06-16 06-16 800,000 700,000 (0.2) (1,401) Mixed SOE and Foreign 600,000 (4.7) (66,635) State-owned (20%+) (2.6) (17,115) 500,000 Privatized (4.7) (28,078) (foreign) 400,000 Privatized (domestic) 9.6 90,845 Foreign (de 300,000 novo) 0.2 5,252 Domestic (de 200,000 novo) 100,000 Total (all exporters) 0 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 (0.2) (17,032) Source: Data from Agency for Business Registers Serbia (SBRA); World Bank analysis. In the composition of export growth between 2010 and 2016, product exports were dominated by motor vehicles, fabricated metal products, other machinery and equipment, electrical equipment, and computer, electronic and optical equipment.1 These exports represented about 27.6 percent of export growth (Figure 5). 1 In this analysis repair and installation of automotive equipment are in this category. 6 Figure 5. Growth in Exports, 2010 and 2016 Source: SBRA data, World Bank analysis. 7 The sophistication of Serbia’s goods exports has risen in recent years, driven by the increasing dominance of manufactured goods. In general, countries that produce more sophisticated products will enjoy higher incomes, because such products are associated with higher productivity and more knowledge spillovers, helping push companies faster up the technological ladder to higher economic growth (UNCTAD 2013). For several years, Serbia has been a beneficiary of this nexus. The net increase in sophistication of exports (EXPY) has been largely driven by manufactures (Figure 6). The share of manufactured goods in total exports also jumped from 59 percent in 2002 to more than 70 percent in 2017. It is worth noting that the higher sophistication of the export basket has coincided with the transition from low- to medium-skill and more technology-intensive manufactures (see Figure 10). Exportof Figure 6. Evolution ofEvolution Sophistication, Serbia over overall export sophistication and High-income time Countries, in Serbia and high-income 1995–2015 countries EXPY index; US$ Country-level vs, composite (several countries) Manufacturing, non-manufacturing; country level 21,000 17,000 15,000 20,000 13,000 19,000 EXPY1 11,000 18,000 9,000 17,000 7,000 16,000 5,000 15,000 3,000 1995 2000 2005 2010 2015 1995 2000 2005 2010 2015 Serbia High-income countries EXPY in manufacturing in SRB EXPY in non-manufacturing in SRB EXPY in manufacturing in High-income countries Source: UNCTAD data; World Bank staff calculations. Serbian services exports have also grown substantially, led by computing and professional services . Between 2007 and 2018, total service exports grew by 6.8 percent (Figure 7). Computing services grew by 27.2 percent and professional services by 11.5 percent—an indication that Serbia is transitioning to higher-income status as knowledge-intensive services become more prominent (Eichengreen and Gupta 2011). 8 Figure 7. Services Exports by Sector, 2007–2018 Service export 2007-2018 Export revenue, USD million (left chart); Proportion of total service exports (right cahrt) CAGR 100% 9,000 90% 11.5 8,000 80% 7,000 27.2 70% 6,000 60% 5,000 50% 4,000 40% 3,000 30% 2,000 20% 1,000 10% 0 0% 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2007 2018 6.8 Other personal, cultural, and recreational services Audiovisual and related services Technical, trade-related, and other business services Professional and management consulting services Professional and creative services Research and development services Information services Computer services Telecommunications services Telecommunications Explicitly charged and other financial services Pension and standardized guaranteed services Auxiliary insurance services Reinsurance Financial and insurance Direct insurance Construction in the reporting economy Construction abroad Personal Business (travel) Postal and courier services Transport, travel, and freight Other (All modes of transport) Freight (All modes of transport) Passenger (Air) Air transport Manufacturing, and goods related Sea transport Maintenance and repair services n.i.e. services Manufacturing services on physical inputs owned by others Goods for processing in reporting economy – Goods returned , Goods received Goods for processing abroad - Goods sent , Goods returned Source: TC 360, World Development Indicators (WDI). 9 However, the exports-to-GDP ratio is still far below the 80 percent enjoyed by other small European transition economies. By 2017 Serbian exports had grown to more than half of GDP, but there is a yawning gap between Serbia and its 7STEE peers. Serbia’s exports-to-GDP ratio is higher than most Balkan peers except North Macedonia but lower than 7STEE peers except Croatia (Figure 8). Figure 8. Export Deepening, Serbia Export in and goods and Selected services Peers, as of 2017, Percent GDP 2017 Percentage, % 100 96.3 90 82.2 81.3 80 78.0 70 67.4 60.5 60 55.1 52.4 51.1 50 41.4 41.1 Group of Seven Small 38.9 transition economies of 40 31.5 Eastern Europe (7 STEEs) 30 26.6 6 Western 20 Balkans 10 0 Slovenia Bulgaria Herzegovina Romania Lithuania Estonia Latvia Macedonia, Albania Kosovo Croatia Republic Montenegro Serbia Bosnia and Slovak FYR Source: TC 360, WDI database. Between 2008 and 2017 Serbia added only 47 new goods (net) to its export portfolio . Croatia added 201, and Lithuania 312 during the same period. Western Balkan countries in general export only a small number of products to a small number of countries. Serbia performs better than its Balkan peers but worse than most of its 7STEE peers (Figure 9). The number of products Serbia exported in 2017 was apparently in the lower middle of the 7STEE peer group. Over time, the number of Serbian exports has been fluctuating; in 2017, the basket held 3,856 products. Like many of its peers, Serbia reached an export peak of 3,887 products in 2006; between 2007 and 2014, the number of products Serbia exported fell but their value has been recovering to pre-2008 crisis levels. However, in 2017 the number of Serbia’s trading partners, 163, was far below all 7STEE peers. 10 Figure 9. Serbian Products and Destination Countries, 2017 Source: TC 360, World Integrated Trade Solution (WITS), 2019. Since 2000 only a very small number of products, if any, created in Serbia embed high-level skills and technology content.2 The technological capacity of Serbian exporters has improved, and more export manufactures have risen to the medium-skill technology tier. In 2001–17 half of the export growth in manufactured goods is explained by rising medium-skill exports (Figure 10). However, this growth suggests that the shares of exported manufactures in the high and the low tiers have been reduced— exports of high-skill and technologically sophisticated manufactures have been flat for 20 years (Figure 10), Serbian exporters thus still have room to improve their participation in exports of highly sophisticated goods, which has become one of Serbia’s greatest export challenges. 2 UNCTAD TradeSift product groups, based on Basu 2011. 11 Figure 10. Technological Capabilities of Serbian Exports of Manufactures, 1995–2016, Percent Structure of exports of manufactures by skill and technology High skill and technology intensity manufactures intensity 8,200 100% 6,200 90% 4,200 80% 2,200 1995 2000 2005 2010 2015 70% Medium skill and technology intensity manufactures 60% 6,500 50% 4,500 40% 2,500 30% 1995 2000 2005 2010 2015 Low skill and technology intensity manufactures 20% 3,000 10% 2,000 0% 2001-2003 2015-2017 1,000 1995 2000 2005 2010 2015 Serbia High-income countries Low-skill Medium-skill High-skill Source: UNCTAD data, World Bank staff calculations. Note: The EXPY index is a measure of the sophistication of a country’s export basket, based on the methodology of Hausmann et al. 2007. It attempts to capture the productivity associated with a country’s exports, as a proxy for the most productive goods a country can produce at a given time, based on the idea that these goods reveal the production frontier. ‘ A larger share of new Serbian exports are ubiquitous products that tend to be exported by countries with less average export diversification. This change has deteriorated the economic complexity of Serbia’s aggregate export basket. Serbia had a higher ‘complexity’ ranking in 2006 than in 2016. In 2006, it was 35th out of all economies) with an economic complexity index (ECI) rating of 0.62, which was 0.04 higher than in 2016. During and immediately after the financial crisis (2008–12), Serbia’s rank fell by 8 places, falling to 0.56 but then dropping again in 2014 to an all-time low of 0.53, and a rank of 40th. The ECI index bounced back slightly in 2016 (Figure 11). Figure 11. Economic Complexity of Serbian Exports, 2006–16. Complexity of Serbia’s export basket of goods 2006-2016 Composition of exports by complexity metrics 2006-2016 ECI (index) PCI index 0.75 ECI 2006 2016 0.70 Weighted Average PCI Weighted Average PCI Share of Total Value Share of Total Value RCA Threshold 0.65 0.60 Value* Value* Count Count ECI 0.55 0.50 0.45 >= 10 25 2.1 27% -0.22 16 2.8 18% -0.48 0.40 >= 5 61 3.8 51% -0.01 40 5.5 35% -0.29 2006 2011 2016 >= 1 225 6.3 82% -0.09 204 12.4 79% -0.05 Total: 779 7.6 -- -- 730 15.7 -- -- Source: Observatory of Economic Complexity, OEC, MIT. 12 Decomposition of export margins sheds light on how changes in the export mix can lead to both greater sophistication and reduced complexity. Growth in manufactures of automotive, machinery and processed metals, and nonmanufactured goods like copper, maize, fruits and furniture can have divergent effects on aggregated export values of sophistication (EXPY) and complexity (ECI). As was shown in Figure 6, relative growth of export manufactures has driven up Serbia’s EXPY score. However, the change in the composition of the export basket can also drive down the value of the complexity score. 3 The low complexity of products that have larger relative shares of Serbian exports (insulated wire, steel, copper, maize, wheat, tobacco, apples, and frozen raspberries) has caused the aggregate ECI score to deteriorate. Raspberries illustrate how Serbia has increased exports of less complex products that have relatively little added-value (Box 2), but at the same time have significant potential to increase its value- added and complexity. 3 Growth of automotive products like wires can drive up the value of the aggregate sophistication score for Serbia, because they tend to be exported by countries with higher income per capita, and the value of the individual product index (PRODY) is higher than the average score of Serbia’s export mix (EXPY). However, wire exports tend to be exported by more countries than the products that Serbia was exporting before (explaining its ubiquity), and by countries that on average are not as diversified as Serbia, implying a complexity score that is lower than Serbia’s aggregate score (ECI). 13 Box 2. Raspberries, a strategic export for Serbia, but Gains in manufactured exports have with low complexity or value added. been driven by a few foreign exporters that are larger than domestic Raspberries are prominent among Serbia’s food exports, competitors. Foreign companies are bringing in €250 million in 2015 and benefitting more than 80,000 farms. A global leader in frozen raspberries, Serbia especially dominant among new exporters accounts for about 18 percent of the world’s exports. Its of motor vehicles, rubber and plastics main markets are wealthy countries in Europe and North electrical equipment, and fabricated America, which take more than two-thirds of Serbia’s fruit metals; it is believed that they moved into and vegetable exports. But despite its remarkable growth, Serbia seeking cost advantages from labor most exports in this segment are frozen raspberries, which efficiency and investment incentive have little value added. schemes (Box 3). However, in the past 10 A recent study of the food and beverage sector, which years some export gains have come from included analysis of opportunities for exports of higher exporter concentration in motor raspberries, suggested that extending the value chains vehicles, computer and electronic and developing new ones, can increase the benefits for equipment, and in installing the raspberry exporters. Among its recommendations were equipment (Figure 12). transitioning to retail-ready packaging by:  introducing new processing technologies, such as freeze-drying;  intensifying production and distribution of fresh raspberries through thermal processing and organic certification;  introducing new resilient varieties which can be managed under extended distribution times, such as larger fruits more suitable for consumption or with longer shelf life;  investing in supply chain facilities, such as irrigation systems and greenhouses to prolong seasonal availability, and improve yields and quality; and  expanding thermal processing and export of jams, juices, concentrates, purees, or other raspberry-based culinary supplements. Central to such scaling up are also dissemination of necessary market information, learning new production techniques, and promoting and branding final products. 14 Figure 12. Export Growth Patterns, Selected Goods and Services, by Ownership Type, 2006–16 Source: SBRA data, World Bank analysis. 15 For the past 10 years growth in exports of manufactures has coincided with higher FDI inflow and more inward processing as a percent of exports. In Serbia FDI as a percentage of GDP has gone up significantly, and in 2018 inward processing trade4 reached almost 30 percent of all final goods trade, up from just 4 percent in 2005 (Figure 13). Moreover, between 2005 and 2018 the value of Serbia’s inward processing trade rose from 1 to 7 percent of GDP. The correlation between more FDI and a higher share of inward processing trade suggests that exporters in Serbia have been participating more actively in global value chains (GVCs), especially since 2011, when SOEs and socially owned companies were phased out and new private companies established. 5 However, higher inward processing trade for export- oriented investors may indicate that sourcing from domestic suppliers is limited, preventing opportunities for knowledge spillovers. Figure 13. Inward Processing Trade and FDI, 2008–18, Percent Source: Eurostat Comext data and World Bank staff calculations. Recent expansion in transport manufacturing has brought more exports, but not necessarily much knowledge spillover. The expansion in automotive production was driven by investments like the Fiat Chrysler renovation (the Kregujevac plant), completed in 2010–2012 and the introduction of the FIAT 500 line, which expanded production capacity, productivity, and exports (Country Economic 4 Processing trade is defined as goods temporarily imported in Serbia for the purpose of being processed and which are then re-exported to the country of origin. 5 FDI has enabled Serbian exporters to specialize more in niche parts of the production chain, either through FDI directly or arms-length transactions with domestic companies; this has important implications for the country’s trade pattern and economic behavior. A recent study of Western Balkan countries by Shimbov et al. 2019 found that an increase of 10 percent in FDI or in trade in parts and components (as a proxy for participation in GVCs) is accompanied by an improvement in the export sophistication index of more than 1.6 percentage points. 16 Memorandum; Serbia’s New Growth Agenda; 2019;Box 5, “The Fiat Effect”). However, the Box 3. FDI that promises jobs, not perception about the extent to which FDI has translated substantive backward linkages. into adoption of new technology in Serbia has been falling steeply since 2008 (Public Expenditure Review An illustration from the Serbian of Small and Medium Enterprise and Competitiveness Development Agency suggests how investment incentives can contribute to Programs in Serbia; 2019). Though foreign the local economy. In a sample of 21 investments in Serbia are still substantive, they seem to investor applications for incentives in have created few links with domestic firms (Box 3). In 2016, business plans revealed that most a recent policy note, Brussevich and Tan (2019) found registered projects belonged to the no evidence of positive spillovers of FDI on the automotive industry, though other productivity of domestic transport manufacturers or on manufacturing sectors and one ICT lower tech sectors. Their study looked at the empirical company were represented. evidence relating FDI to growth of total factor productivity (TFP) in domestic firms between 2005 The new investment reached €300 million and 2016 in terms of forward, backward, and and promised to create about18,000 new jobs, generating about €1.1 billion in horizontal links. However, they did find evidence of sales, of which more than 95 percent positive FDI spillovers on domestic firms in Serbia at would be from exports. The applications the aggregate (i.e., associated with growth in revealed that domestic sourcing of productivity), mostly in the form of TFP growth in material inputs (excluding labor and domestic firms that enjoyed backward and horizontal energy) was only 8.6 percent, quite low links with foreign investors; moreover, transport by many standards. The fact that more manufacturing not only attracts one of the largest than 90 percent of raw material would shares of Serbian FDI but also benefits considerably come from foreign suppliers helped from investment incentives. explain the rise in inward processing trade. Recent gains in exports of transportation, A preliminary analysis of business plans electronic, and electrical machinery and equipment for 2015–18 confirms that sourcing are susceptible to disruptive technological trends. projections of local input supply would Use of robots in high-income economies, which hold at about 9 percent. questions the feasibility of the labor-intensive production processes used in less industrialized countries, varies by manufacturing subsector, as shown by the relative magnitude of automation, tradedness, export concentration, and services intensity as “high” or “low” (Figure 14). Adoption of automation technologies, the associated intensity of professional services input, and shifting trade patterns vary by manufacturing subsector. If Serbian exporters are to take advantage of disruptive megatrends in production that affect the manufacture of transportation equipment, electronics, electrical machinery, and machinery and equipment, they will have to move to the cutting edge of technology. Today Serbia is at risk of job displacement in these industries. 17 Figure 14. New Pressures Affecting Global Export of Goods EXTENT OF IMPACTS OF NEW TECHNOLOGY AND PRIORITIES WITHIN 3Cs GLOBALIZATION AGENDA Competitiveness Connectedness Capabilities Increasing Sectors (grouped by concentration the common of combinations of international Robots/3D Use of trends they face) production Traded printers Services Transportation High High High High Electronics High High High High Pharmaceuticals High High High High Electrical machinery High High High High Yes Yes* Yes Machinery and equipment High High High Low** Manufacturing nec. High High High Low** Textiles High High Low Low Yes Yes Rubber and plastics Low Rising High Low Yes+ Fabricated metals Low Rising High Low Food Low Low Low High Chemicals Low Low Low High Yes Coke and refined petroleum Low Low Low High Wood products Low Low Low Low Paper products Low Low Low Low Basic metals Low Low Low Low Non-metallic minerals High Low Low Low Source: Hallward-Driemeier and Nayyar 2017. According to Hallward-Driemeier and Nayyar (2017), in Figure 15 Serbia falls in the “high capabilities- high connectedness” quadrant but has only “medium competitiveness.” Only countries in the upper-right quadrant and high in competitiveness are not at risk of displacement in manufacture of transportation equipment, electronics, electrical machinery, and machinery and equipment. Improving production and exporter conditions in Serbia will mitigate the risks posed by disruptive technologies. 18 Figure 15. Competitiveness, Capabilities, and Connectedness (Index) Source: Hallward-Driemeier and Nayyar 2017. To summarize the findings so far, Serbia has plenty of room to improve its export performance. By many measures, Serbian exporters have done well; Serbia’s exports of goods and services grew from 30 percent of GDP in 2006 to 51 percent of GDP in 2018. Serbia is also exporting more products, especially from the automotive industry. Other sectors, such as Rubber and Plastics, have also benefited from Foreign Direct Investments (Box 4). 19 Box 4. Rubber and Plastics in Serbia: from favorable However, a few other factors conditions for export-oriented investments to post-crisis somewhat darken the picture and recovery it does not look that hopeful when taken into consideration a few Rubber and plastics (R&P) is one of the few sectors in the additional issues. Serbian economy that has seen dynamic growth since 2009. With a more than 8 percent share of gross value added (GVA), The gains from incremental in 2016 this was the second largest manufacturing industry. exports have been from products Foreign direct investment has led growth in R&P. Since the that are more common in global early 2000s, nearly 100 foreign investments have entered the markets, contributing to the R&P sector in Serbia. The first batch of investors came during a deterioration of export privatization drive, but the second batch entered through complexity. Compared to its greenfield investments. aspirational peers, Serbia’s range of product exports is also more Serbian opportunities, and global trends in relocating production, have created an attractive environment for FDIs. limited, and they have not First, labor costs in Serbia were among the lowest in Europe, penetrated as many markets as the and skills, knowledge, and flexibility were relatively high. In 7 STEEs. Furthermore, the 2016, according to Eurostat data, average salaries for margin of exports has come from managers in the EU28 were 5.03 times higher than in Serbia, a few foreign companies and has salaries of technical experts were 3.92 times higher; of machine a more concentrated exporter operators 2.93 times higher; and of low-qualified and unqualified base. These foreign companies, workers 2.67 times higher, with the expectation of Bulgaria and particularly in transport Romania. However, productivity in this sector in Serbia is manufacturing, seem to have sufficiently higher compared to these two countries, so that the contributed to employment, and a ration between productivity and unit labor cost is more favorable for Serbia. recent analysis of links between foreign and domestic firms does In addition to this , electricity costs are also among the lowest in support evidence of positive Europe. The average price per kW/h, which is a key input for knowledge spillovers. Over time, the industry, is lower than in most EU countries. however, Serbia has shown that it has not been able to diversify its exports into the high-technology- Finally, the expertise, diversity, and adaptability of Serbian skill segment. Finally, the recent engineers and technical workers were also deciding factors in growth in exports in transport and the R&P sector success in Serbia. The favorable balance equipment manufacturing is between productivity and input costs has enabled foreign operators in Serbia to be highly cost-competitive. vulnerable to disruptive trends of automation and the digitization of production in smart factories. Serbian exporters must move to the frontier of technological upgrading so that they can take advantage of disruptive megatrends in production and technology. 20 II. Despite Better Conditions for Exporters, Serbia Still Trails Aspirational Peers. To look at exporting conditions in Serbia, we adapted “Policies to Enhance Participation in GVCs” (World Development Report; 2020), which takes into consideration market size, geography, the quality of local institutions and endowments, and productive capabilities available to exporters. Market Size Serbia has opened its trade regime through stabilization agreements, and its tariffs are low . Trade openness is probably a result of Serbia’s advance toward membership in the World Trade Organization (WTO). However, the few tariffs remaining may affect the exports of companies that rely on raw materials, which must cross borders twice, making the cumulative impact of tariffs high (Figure 16). Trade facilitation measures, such as harmonizing of regulations and procedures, the National Single Window System, digitization of border-queuing management, and transport management systems, have complemented the reduction in tariffs. Further decreases could add 0.1 percent to annual growth. Figure 16. Tariffs on Selected Product Categories Compared, 2017 Source: World Bank, UNCTAD; WITS; Official Gazette, Serbia As part of the reforms required for WTO accession the Serbian government has also eliminated nontariff barriers to trade, such as import quotas, licensing requirements, and prohibitions on certain imported goods. Geography Though Serbia still trails the 7STEEs on capacity to ease trading across borders, it does much better than its Balkan peers. In recent years, Serbia has streamlined customs procedures in its effort to comply with WTO requirements. However, its greatest problem is the comparatively high compliance costs for customs procedures compared to all peers (Figure 17). 21 Figure 17. Trading across Borders Compared, 2017 Source: World Bank Doing Business database. Serbia ranks between Balkan and 7STEE peers in terms of efficiency in moving goods across borders. Logistics capabilities are critical to remain competitive, because Serbian suppliers who want to participate in global trade can only be as strong as the weakest link. Based on the international scorecard 6 Serbia trails six of the 7 STEEs (Latvia is the exception) in terms of aggregate logistic performance, but it is ahead of its Western Balkan peers. The efficiency through which border control agencies, such as customs officers, clear exports and imports seems particularly deficient—the speed with which officers can complete formalities for value chain operators is below par, and documentation requirements are unpredictable. The quality and competence of transport, warehousing, and brokerage services were also lower than in peers (see Figure 14), even Balkan countries. Other areas where Serbia showed deficiencies were the quality of ports, railroad systems, roads, and ICT infrastructure, which is necessary for moving goods, and inability to track and trace consignments (Figure 18). These deficiencies can also negatively affect some export sectors, like agricultural foods. A sector study found that lack of post-harvest and supply chain infrastructure prevents local firms from meeting the high-quality standards of European markets and penetrating attractive consumer niches. Agro-industrial facilities are limited as land is further divided into ever smaller agricultural plots and state-owned land, is trapped by institutional burdens that make its use sub-optimal. The prevailing post-harvest techniques and facilities, logistics, and effective market links prevent the highest-quality fruits from reaching international markets. Instead, most fruit is exported frozen, which lowers its value. All processing stages in local supply chains experience losses due to obsolete machinery and poor management. Additional investment is needed, first and foremost, in equipment, transport vehicles, and production technology upgrades. 6 The international score benchmarks country performance against the six indicators of the Logistics Performance Index (LPI). which allows comparisons with global and regional peers. 22 Figure 18. Logistic Performance of Serbia and Selected Peers, 2018 Source: LPI. Institutions Serbia’s enforcement of contracts is mid-level—far behind its STEE peers but ahead of its neighbors in the Western Balkan. Its costs for enforcing contracts are the highest in both samples (Figure 19). A poor court system probably deters foreign investment, which may affect exports. Serbian courts are slower than in most STEEs (except for those in the Slovak Republic and Slovenia), with legal procedures averaging 635 days. The main problem is the time it takes the courts to conduct a trial and reach a verdict; Serbia seems to be on par in terms of time for filing suits and enforcing a verdict. Lower courts seem to be overwhelmed and reliant on the appellate court to reach a conclusion. Insolvency cases seem to be particularly slow. One area of major concern is the high costs of claims: attorney fees, court fees, and enforcement fees range from 12.4 percent to 14.5 percent of the costs of the claims—except for Kosovo the highest in the entire sample of Balkan and STEE peers. 23 Figure 19. Enforcing Contracts, Serbia and Selected Peers, 2018 Source: World Bank, Doing Business. 24 Enforcing protection of intellectual property rights should become more efficient (Figure 20). The Serbian FDI regime is relatively liberal: Foreign and domestic companies are treated equally, with no discrimination and few limits on foreign ownership except for defense and a few other sectors; and foreign companies have been welcomed in privatization. The law protects foreign companies from expropriation and guarantees repatriation rights. However, according to the World Economic Forum (WEF), Serbia’s protection of intellectual property (IP) is midway between that of the STEEs and Balkan nations. Laws protecting investors’ IP and efficient enforcement of their rights is important to assure them that Serbia protects intangible knowledge capital. Policy makers should also make it a high priority to reduce excessive bureaucratic red tape and eradicate corruption. Figure 20. IP Protection and Competition, Serbia and Peers, 2018 Source: WEF. The WEF ranking indicates that Serbia’s domestic competition framework is inadequate, and Serbia trails not only most STEEs but even some Balkan countries. Because previous studies have found that the national commission to promote competition is not well funded, it may be susceptible to interest group influence. Accession to the WTO and perhaps the EU will bring Serbia’s competition to a higher standard, but that level may not be achieved in the short term. Building up domestic competition is important in Serbia, particularly to ensure that private firms are treated equally with public firms, and by accelerating State Owned Enterprise (SOE) restructuring. 25 Box 5. Wood and Furniture Exports in Serbia: It is imperative to address issues at both ends of the export chain Serbia is moderately rich in forests, and after 2008 the wood and furniture industry made a dramatic turn to exports, primarily to developed EU countries. At that time export growth accelerated, driven by Serbian gains in external market share. The 13 percent growth in wood and furniture exports was one of the highest export growth areas. But as with the Serbian economy writ large, though this industry has done reasonably well, it could do better. A 2016 study found that of 2,155 wood and furniture companies, 696 were exporters. These companies, less than one-third of the firms in the sector, created more than 70 percent of sector revenue and employment. Foreign producers, primarily large companies, created high added-value per employee and were very export-oriented, but the sector engages many small enterprises in rural areas. The greatest potential for adding more value to Serbian wood is at the very start of the chain, in forests, where firms have supply problems, and at the very end, in furniture, where there is a shortage of design and craft skills. Currently, Serbia does not derive the value from each wooded hectare that it could. The biggest issue with acquiring raw wood is that supply is unpredictable. Without knowing the quantities sold, buyers sold to, or the logic used to allocate raw wood to different buyers, production planning is arduous. As a result, Serbia often sells raw wood, a low-value-added, unsophisticated product, while other countries keep it to produce furniture. The other opportunity lies in nonproduction activities such as design and marketing furniture for export. However, in the short term the lack of professional staff with secondary education (craftspeople, upholsterers, and carpenters) limits expansion of production. Source: “Review of the Actual Situation with SWOT Analysis in the Sector of Wood and Furniture Industry in Serbia,” World Bank 2016 Endowment and Productive Capabilities If Serbia is to attract and nurture world class exporters, work force skills need a boost. The availability of low-cost labor can create a temporary advantage, but this edge may disappear as soon as productivity benefits materialize. According to the WEF, Serbia trails most of its peers in terms of the quality of current and future skills for economic growth; it is behind Latvia and Estonia and at the same level as Montenegro and Albania. In terms of education, Serbia does somewhat better but could still improve secondary education (Figure 21). 26 Figure 21. Quality of Worker Skills, Serbia and Peers, 2018 Ranking of the quality of current an future skills in 2018 Quality of education skills in 2018 Ranking: #1 top of the class score, 1=low to 5=high Slovak Republic 5 Montenegro Bosnia & H Macedonia Slovenia Bulgaria Albania Estonia Croatia Serbia Latvia 10 4 30 50 70 3 90 123 123 123 123 123 123 123 123 123 123 123 Serbia Latvia Bulgaria Macedonia Slovenia Croatia Estonia Albania Bosnia & H Slovak Republic Montenegro 110 130 1) Skillset of graduates Skills of current workforce Skills of future workforce 2) Skillset of secondary-education graduates 3) Skillset of university graduates 6 Western Group of Seven Small transition Balkans economies of Eastern Europe (7 STEEs) Source: WEF. In 2016 Serbian firms reported finding it difficult to attract qualified workers, with shortages as high as 87 percent in skilled agriculture, 79 percent in professionals, and 78 percent in sales (Country Economic Memorandum; Serbia’s New Growth Agenda, 2019, based on the 2015–16; STEP employer survey). Lack of highly skilled managers, experienced in both integrated process management and corporate governance, is depressing Serbia’s technological progress, particularly for SMEs. The evidence suggests that managerial skills determine how well a firm can adopt and use new technologies. However, in 2016 60 percent of firms in Serbia reported problems recruiting staff for managerial jobs (Country Economic Memorandum; Serbia’s New Growth Agenda, 2019). Collaboration between private firms and between public and private entities is also necessary . There are instances where cooperation and coordination would enable private exporters to work collectively on standards for better quality and advocate for supportive regulation and public services to become more competitive. One service sector example may be instructive: Creative industries are undercapitalized and lack a unified strategy to articulate priorities for stakeholders. Incentives are predominantly directed toward audiovisual services (filming and post-production), but only the sector, where ICT, e-commerce, and the digital and sharing economy intersect, can catalyze change and build a more inclusive, connected society. Serbia needs the Strategy and an Action Plan for Creative Industry (defined both narrowly and broadly) that the Prime Minister promised a year ago; what is even more important is smooth implementation of measures to foster creative initiatives and inclusiveness. Improving the base of local suppliers is crucial if Serbia is to stay competitive in attracting foreign investment. The WEF gives Serbia one of the lowest rankings in the sample for the availability and quality of local suppliers (Figure 22), far behind Albania as well as most STEEs countries. This is consistent with reports from foreign firms operating in Serbia; these indicate a cultural dissonance between what their businesses need and how Serbian domestic firms conduct themselves in business is preventing deeper integration of domestic suppliers in global production systems. 27 Figure 22: Quality of Suppliers and ICT Infrastructure, Serbia and Selected Peers, 2017–18 Quantity and quality of local suppliers in 2018 Fixed broadband subscriptions score, 1=low to 7=high (Per 100 people) 29 6 24 19 5 14 9 Latvia Serbia Estonia Slovenia Albania Slovak Republic Bulgaria Romania Montenegro Croatia Macedonia, FYR Lithuania Bosnia and Herzegovina 4 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 Latvia Bulgaria Bosnia & H Serbia Slovenia Estonia Albania Croatia Montenegro Slovak Republic 1) Local supplier quantity 2) Local supplier quality 6 Western Group of Seven Small Balkans transition economies of Eastern Europe (7 STEEs) Source: WEF. The share of professional, scientific, and technical service inputs into manufacturing value-added vary by subsector and matter greatly to adoption of new technologies. Poor ICT infrastructure depresses both the growth of exporting service sectors and the opportunity to make manufacturing more service-intensive. Internet connectivity is substandard; without adequate fundamental infrastructure, investors are likely to locate elsewhere. Figure 22 compares ICT infrastructure among peers, using broadband subscribers per 100 as a proxy measure for ICT built capability. Serbia’s density of subscriptions was lower than all STEEs, and even lower than some Balkan neighbors. Solid technological capabilities are central to promoting exports, but Serbian firms invest little in research and development (R&D)—a problem that can reduce the potential for adapting to international competitive trends that demand the latest technological capacity. Serbian firms that invest in R&D have higher growth in productivity (Country Economic Memorandum; Serbia’s New Growth Agenda, 2019), but compared to European countries in general, in 2014 only 14.3 percent of firms with more than 10 employees invested, and even they invested only 0.3 percent of revenue in 2014. Furthermore, Serbian firms trailed all country peers except Albania in technological readiness and except for Hungary had the lowest rates of technology absorption (World Economic Forum, indicator database. 2017). Countries that are serious exporters of manufactures have established these priorities through foresight studies and strategies (OECD 2017). Keeping up what investors and customers are demanding requires that Serbian exporters increase their capacity to absorb technology and provide knowledge-intensive products. If they are to retain a competitive edge, they need to embrace the new capabilities, such as the Internet of Things, advanced robotics, and 3-D printing, and generally acquire productive know-how. Aggregate investment in science and technology in Serbia is at the mid-level of the STEEs, and analysis of the data suggests that firms are responsible for most of the gap in investment for R&D . Serbia’s R&D spending, both public and private, ranks with that of some STEE economies but is far below the EU average of 2.03 percent. Performance. measured by patent applications, puts Serbia at the STEE average and generally better than Balkan peers. These outcomes are not surprising since the 28 number of Serbian R&D researchers and technicians was at the middle of the STEE group, and generally above its Balkan peers (Figure 23). Thus, firm reluctance to invest with R&D rather than public spending seems to be at the root of the problem. According to the European innovation scoreboard, in 2017 the business sector share of R&D as a percent of GDP was 0.3 percent in Serbia, compared to 0.45 percent in Croatia. 1.0 percent in Hungary, 0.6 percent in Bulgaria, and 1.6 percent in Slovenia. R&D can also be considered a proxy for inputs; it does not reveal anything about the quality of the investment results. The production of Serbian researchers has apparently been declining, based on the falling number of publications since 2012. Their quality is also low; Serbian publications have an H-index7 of 149, compared to 205 for Bulgaria and 221 for Croatia) (World Bank 2019). Figure 23. R&D Outcomes and Professional Capability, Serbia and Comparators, 2016 R&D expenditure 2016 Patent applications 2016 Researchers and technicians in R&D 2016 (% of GDP) (residents) (per million people) 4,000 Slovenia 2.0 N/a 3,500 Estonia 1.3 29 3,000 Serbia 0.9 192 2,500 Croatia 0.9 175 Lithuania 0.8 2,000 95 Slovak Republic 0.8 220 1,500 Bulgaria 0.8 230 1,000 Romania 0.5 1,005 500 Macedonia, FYR 0.4 N/a 0 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 Latvia 0.4 95 Slovenia Lithuania Estonia Serbia Latvia Bulgaria Montenegro Slovak Republic Bosnia and Herzegovina Romania Macedonia, FYR Croatia Montenegro 0.4 10 Bosnia and… 0.2 60 Albania N/a 21 6 Western Group of Seven Small 1) Researchers in R&D (per million people) Balkans transition economies of Eastern Europe (7 STEEs) 2) Technicians in R&D (per million people) Source: WDI. Because EU consumers are increasingly concerned about health, ecology, and food safety, it has become necessary for exporting countries to master control and quality standards and product labeling and branding. A 2017 survey of Serbian quality managers concluded that the national quality infrastructure has not contributed much to economic growth, partly because adaptation of the European requirements was incomplete. Opinions were divided about how much the National Quality Infrastructure (NQI) system had contributed to the competitiveness of exports from Serbia (Ruso 2017). III. Better Operating Conditions and Better-Targeted Interventions Would Help Serbian Firms to Compete with the Country’s Aspirational Peers Bringing Serbian exports from 52.4 percent of GDP to the 80 percent level of the 7STEEs calls for several measures. The first step is to improve operating conditions for exporters so that government policy can help them to reap the benefits of exporting. Addressing at the outset gaps in cross-cutting issues using less interventionist policies based on a review of tariffs and Non-Tariff Measures (NTMs) 7The H index estimates the impact and productivity of citations, based on the publication’s linkages to other citations (this is also known as the Hirsch Index) 29 business-enabling environment policies, public infrastructure, customs bureaus, and endowments would build a solid foundation for improving operating conditions for exporters and facilitating their fruitful participation in global markets. The government can then take a bolder approach to productivity policies related to export promotion, which are typically designed to redress specific market failures affecting export-oriented investors, direct suppliers, and domestic exporters. Used correctly, these interventions have potential to generate positive externalities, such as knowledge spillovers and increases in productivity through competitive pressures and learning by exporting. Here, a sequential approach seems promising, for the following reasons (see also Table 1): 1. The decisions of investors and exporters tend to be driven by operating conditions, not incentives. Even though policymakers have traditionally prioritized incentives, it has been found that they have little influence on investment decisions (Global Investment Competitiveness Report 2017/2018). A low tax burden or grants cannot compensate for a generally unattractive operating environment. 2. Export incentives that rely on subsidies, directly or indirectly, could lead to significant distortions, especially if they are not justified as a response to market failures. For example, incentives can create a situation where policy makers artificially expand exports for which their producers do not have a long-term comparative advantage. The case for policy intervention needs to be clear. Maloney and Nayyar (2018) argue that establishing the case for a policy requires evidence of a market failure that can justify interventions. (Note that it has been possible to make a more credible case for addressing market failures for horizontal than for vertical policies). 3. To redress market failures, government agencies must be able to devise policy solutions that address the problem directly and deliver the intervention successfully. For example, too-complex incentive schemes can burden the managing agency, whether export or tax authority, treasury or revenue administration, by imposing demanding responsibilities to verify compliance, particularly when variants of the scheme apply differently by type of firm or product. According to the review by Andersen et al. (2018), an OECD survey in 2003 found that some investors were discouraged from participating in complex schemes because they lacked transparency. 4. Incentives for exporters and investors may risk state capture by investors, which could lead to corruption, unfairness, and uncertainty, considering that incentives rely on agency discretion. 5. Finally, indiscriminate use of subsidies or tax incentives will likely intensify pressure on scarce fiscal resources. Table 1. Improving Operating Conditions for Serbian Exporters Issue Recommendations Reduce tariffs  Eliminate the few remaining tariffs and any NTM restriction not necessary to and Non-Tariff protect consumers. Measures (NTMs) for raw material. Streamline  Reduce the relatively high cost of document and border compliance. border  Complete trade facilitation measures being planning in the Balkans 6 project, procedures. such as harmonizing regulations and procedures, the National Single Window System, digitization of border queuing management, and transport management systems. 30 Expand  Improve the efficiency of border control agencies, the speed with which officers transport and can complete formalities, and the predictability of document requirements. logistics  Invest in improving the quality and competence of transportation and capacity. communication infrastructure, such as roads, digital and sea routing, as well as transport, warehousing, and brokerage services. Reduce the  Build court capacity to reduce the time required for trial through to verdict. time and costs  Increase the quality of court and legal services to reduce the high costs of of enforcing processing and enforcing contracts. contracts. Build the base,  Expand the access of Serbian suppliers to capital based on their relationships and the quality, with reputable larger exporting firms. If they can promote Serbian ability to of suppliers quickly produce customized products, they could generate significant growth, given the near-shoring trend in the most developed EU countries.  Combining capability upgrading with market links could benefit the automotive, agribusiness, and wood/furniture sectors, which are Ministry of Economy priorities. There are solid opportunities to foster links between domestic SMEs and foreign investor buyers in South Eastern Europe, including Serbia. Improve  To inform policy design, Serbian policy makers should coordinate with public-private stakeholders in prominent manufacturing sectors to assess their technology coordination research priorities. for collective investments. Promote the  Promote professional, scientific, and technical service inputs into manufacturing availability of value-added, particularly in relation to adoption of new technologies. export-related  Expand the quality of export support services that could interact with traditional services. export sectors by providing, e.g., transport, warehousing and logistics. These services could be a source of employment for workers in creative services, ICT, and professional services. Build up R&D  Promote interdisciplinary partnerships and inter-institutional coordination to and help manufacturing firms and their suppliers to integrate the different areas of technological expertise required. The need to combine a variety of scientific disciplines is capabilities. becoming increasingly evident as manufacturers are confronted by challenges that call for expertise in different domains, such as production tools, operations management, advanced ICT, and advanced materials (OECD 2017).  Build capacity for technology extension services to disseminate research results, which are often not properly circulated to firms, and scale up new technological advances. Support these efforts with demonstration facilities and pilot production infrastructure, coupled with advanced scientific advice, and specialized equipment. Support  Local and international universities and the NQI system should be called on to quality- expand the pool of expertise in such areas as advanced metrology, data enhancing collection through smart sensors, characterization analysis, shared databases, infrastructure. and modelling and simulation. Also critical is ensuring financing for acquisition of advanced equipment. Serbian policy makers have relied on productivity policies to enhance exports, but it is not known how well these are working. Policies to attract export-oriented FDI, develop links with local suppliers, and promote domestic exporters are deeply rooted in the policy toolkit. Serbia offers a series of tax holidays, tax credits, and cash subsidies targeting employment outputs, promoting disadvantaged regions and incentivizing investments in R&D, IT, and training. An evaluation of the program to attract direct 31 investment (WBG 2017 showed cash subsidies have been effective in creating jobs (96 per participating firm; €2,086 per job), but that the efficiency of the program could not be established. Furthermore, a recent study (Public Expenditure Review – PER of Small and Medium Enterprise – SMEs and Competitiveness Programs in Serbia, 2019) found that the RAS presented opportunities to understand whether the programs are working as intended, and that monitoring and evaluation (M&E) systems could be strengthened. For example, agencies do not collect much data about outcomes or impact of interventions; the scope of M&E is limited to counting how much is disbursed in grants. The study concluded that it was not possible to state whether the programs reviewed were meeting their goals. A recent study found opportunities Box 6: Vouchers for Innovative Exports: The experience for rebalancing the portfolio of of Korea productive export-promotion policies (Public Expenditure Review Vouchers are small nonrepayable grants allocated to SMEs of Small and Medium Enterprise to purchase services from external knowledge providers. and Competitiveness Programs in The main objective is to induce non-innovator SMEs to start Serbia, 2019). The current targeting of collaborating with knowledge organizations and providers. policies to attract FDI suggests there Vouchers are often entitlement- rather than competition- may be a bias toward export based, and they typically require light management and manufactures. Attracting investors that effective auditing. rely on low labor costs to be The justification for the use of vouchers is that they can competitive may have little potential address the capability failures of smaller firms by inducing for transferring knowledge to the local changes in behavior to more proactive learning and economy. Serbian policymakers have sustainable collaboration with knowledge providers. There is apparently been prioritizing policies often acute information asymmetry between knowledge that create jobs. Now, however, efforts providers (particularly public-sector research organizations) to attract FDI may need to concentrate as suppliers of innovation knowledge and SMEs as potential on attracting exporters of knowledge- users (Cirera et al. World Bank. forthcoming). intensive rather than labor-intensive products, which are more likely to Korea now has a voucher program (since 2017, with a generate knowledge spillovers. budget of US$14 Million) to promote exports by domestic SMEs. It provides exporters with vouchers that list supporting services by category. All services listed are Serbian policy makers may also available for all participants regardless of the program they need to revisit current investment are engaging in. incentive schemes to verify that there are no obvious duplications. SMEs co-finance all services to avoid moral hazard, Schemes to attract natural-resource promote accountability, and leverage financial resources. and market-seeking investors should The voucher program is structured by degree of exporter especially be considered for experience in the following stages: rationalization. It is important that incentive schemes have positive 1. Preparation: Translation of webpages and data into foreign languages, optimizing design, education on economic returns, and that the trade and marketing program targets the best potential 2. Beginning: Marketing through media, support for global investors for Serbia. A revision might market research and matchmaking, participation in eventually re-direct tax incentives exhibitions toward the higher-yielding investors 3. Contract: Checking buyer’s credit, writing the contract who can increase the share of document, managing export distribution knowledge-intensive exports. That 4. Global expansion: Support for building local branches, may require careful redesign of the consulting on Marketing and Advertising. scheme and a cost-benefit analysis. Source: Yong-Seok Choi, Professor, Kyung Hee University. For instance, local companies have complained that incentive schemes 32 targeting foreign companies can create unfair competition—anecdotally, Serbian firms in the ICT sector have stated recently that measures to attract foreign firms would negatively affect competition in the local labor market. Even though enabling domestic exporters is a prominent policy aspiration, this priority is not reflected in resource allocations. Although Serbian firms have shown low and diminishing competitiveness internationally, and the government’s SME Strategy 2015–20 calls for increasing exports, little financing is allocated to government programs that support domestic exporters and market links. Serbia’s incentive polices for promoting exports seem biased toward attracting export-oriented investment from abroad—there is no substantive policy support for market entry and expansion (e.g., exports and links with multinational companies). The SME PER identified only six programs supporting creation of market links and their budgets added up to RSD 91 million, just 0.5 percent of the value of the 2017 portfolio. Table 2 summarizes recommendations for revising the approach. (Annex 1 has a brief review of the literature on Promoting Exports, Attracting Investment, and Developing Suppliers programs.) Table 2. Productivity Policies for Enhancing Exports Issue Recommendations Evaluate the  Validate whether current investment incentives are fully aligned with Serbia’s current economic objectives; particularly check whether they promote not only jobs but alignment of also knowledge spillovers, such as innovation and productivity growth. investment  Ensure that current design of investment incentives is also aligned with best incentive practices, such as, e.g., use of cost-based incentives that draw on policy goals, schemes. such as investments in R&D, developing new product lines, certification, and employment. Verify that there  Schemes to attract natural-resource and market-seeking investors are often are no obvious redundant and should be considered for rationalization. It is important to avoid redundancies in incentive schemes that generate negative spillovers by creating unfair the schemes. competition for domestic firms. Ensure that  It is important that incentive schemes have positive economic returns, and that targeted are designed to attract the best potential investors to Serbia. Revision of the interventions schemes may eventually re-direct tax incentives toward activities that generate will have higher returns, e.g., knowledge spillovers. In that case, there should probably be economic careful redesign of the schemes to incorporate a full-blown cost-benefit returns. analysis. Rebalance  Serbia’s incentive for promoting exports seems biased toward attracting export- support toward oriented investment from abroad. The SME PER found that Serbian policies for domestic market entry and expansion (e.g., exports and linkages with multinational exporters. companies) do not have substantive support.  The SME PER identified only six programs that supported creating market links, and the total budget for them, RSD 91 million, was only 0.5 percent of the value of the portfolio in 2017. Enhance  RAS and other development agencies have shown they need to build capacity outreach of for targeting and selecting companies that are growth-oriented. This requires export qualitative assessment—a very different approach from the current selection promotion process which relies heavily on pre-determined screening metrics. agencies and  It is necessary to build the capacity of individuals with business experience to their selection assess business proposals and manage relationships with companies to capacity. accompany them on their upgrading journey. 33 Improve the  This can be done by introducing results-based management practices, stronger M&E capacity M&E, and analytical assessment of previous and current programs; clarifying of export the role of RAS as a relationship manager with beneficiary (and target) firms, promotion encouraging knowledge-sharing that help employees to understand the impact agencies. of RAS programs and activities; and engaging more intensively with private businesses in preparing future firm-level interventions. 34 References Alexander James Gaspar Simoes, César A. Hidalgo. 2011. The Economic Complexity Observatory: An Analytical Tool for Understanding the Dynamics of Economic Development. Andersen et al. 2018. Corporate Tax Incentives and FDI in Developing Countries, Global Assessment Competitiveness Report. Washington D.C. Basu SR. 2011. Retooling trade policy in developing countries: Does technology intensity of exports matter for GDP per capita? Policy issues in international trade and commodities (Journal). Bojan Shimbov , Maite Alguacil & Celestino Surez (2013) Fragmentation and parts and components trade in the Western Balkan countries, Post-Communist Economies, 25:3, 371-391, DOI: 10.1080/14631377.2013.813142 Bojan Shimbov, Maite Alguacil & Celestino Suárez (2016) International Production Networks and Economic Growth: The Case of the Western Balkan Countries, Eastern European Economics, 54:1, 49-70, DOI: 10.1080/00128775.2015.1112749 Bojan Shimbov, Maite Alguacil & Celestino Suárez (2019) Export Structure Upgrading and Economic Growth in the Western Balkan Countries, Emerging Markets Finance and Trade, 55:10, 2185-2210, DOI: 10.1080/1540496X.2018.1563538 Brussevich, Mariya, and Shawn Tan, Forthcoming, FDI Links in Serbia Policy Note. World Bank Group, Washington D.C. Country Economic Memorandum; Serbia’s New Growth Agenda. 2019; World Bank Group, Washington D.C. CEVES (Center of Advanced Economic Studies. 2017. Integrated Report on Performance and Value Chain Analysis of Selected Sectors within Manufacturing Industry. Belgrade. Eichengreen and Gupta.2011. The two waves of service-sector growth; Oxford Economic Papers. Ek, Colin. Forthcoming. Product Space Analysis of Serbian Exports (Draft). Economist Intelligence Unit. 2018. Policy and Business Outlook: Policy Towards Foreign Investment, London. United Kingdom. Hausmann; R. & Hwang; J. & Rodrik; D., 2007. "What you export matters," Journal of Economic Growth, Springer. Harrison, A., & Rodríguez-Clare, A. (2010). “Trade, foreign investment, and industrial policy for developing countries.” In Handbook of development economics (Vol. 5, pp. 4039-4214). Maloney; W. and Nayyar. G. 2018; Industrial Policy, Information, and Government Capacity; Oxford University Press on behalf of the World Bank. 35 Maria R. Andersen, Benjamin R. Kett, and Erik von Uexkull. 2017. “Corporate Tax Incentives and FDI in Developing Countries”. In Global Investment Competitiveness Report 2017/2018: Foreign Investor Perspectives and Policy Implications. World Bank Group M. Hallward-Driemeier and G. Nayyar. 2017. Trouble in the Making? The Future of Manufacturing-Led Development, Washington D.C Nelson Lind, Natalia Ramondo. 2018. Trade with Correlation; NBER Working Paper No. 24380; Cambridge, MA 02138 OECD (2017), The Next Production Revolution: Implications for Governments and Business, OECD Publishing, Paris. OECD. 2003. Tax Incentives and Foreign Direct Investment: A Global Survey OECD Publishing, Paris. Ruso, Jelena & Filipovic, Jovan & Tosic, Biljana. 2017. The Serbian Quality Infrastructure as Viewed by the Quality Managers. Serbian Journal of Management. World Bank. 2017. “Evaluating the Serbian Program for Attracting Direct Investments 2006–15: A Micro-econometric Study (Draft). World Bank Serbia Competitiveness Policy Note. Rebalancing Serbia’s Economy: Improving Competitiveness, Strengthening the Private Sector, and Creating Jobs. 2014; World Bank Group, Washington D.C. World Bank. 2018. “Regional Investment Reform Agenda for the Western Balkans, Policy Note on Investment Incentives.” (Draft) World Bank. 2019. Public Expenditure Review of Small and Medium Enterprise and Competitiveness Programs in Serbia; World Bank Group, Washington D.C. World Bank. Western Balkans Regular Economic Report; No.15; Spring 2019; 2014; World Bank Group, Washington D.C. World Bank. Revving Up The Engines of Growth and Prosperity; 2017; World Bank Group, Washington D.C. World Bank. 2020. World Development Report 2020: Trading for Development in the Age of Global Value Chains. Washington, DC: World Bank. World Bank. Forthcoming. Innovation Policy Practitioners’ Guide. Instruments for Capability Building and Technological Catch-up in Developing Countries 36 Annex 1: Literature on Promoting Exports, Attracting Investment, and Developing Suppliers Incentives for Export-Oriented FDI An investment incentive is measurable economic support that governments provide to specific enterprises in order to steer investments into either priority sectors or regions or to influence the character of the investments (James, Sebastian 2013). Serbian policy makers also use incentives to bring the Serbian economy closer to foreign markets and to enable Serbian firms to access advanced technology, respond to competitive pressures, raise production standards, and facilitate growth in productivity for the entire economy. Incentives used to attract foreign direct investment (FDI), especially in promoting exports, are primarily tax holidays, tariff exemptions, and direct subsidies. Justification Tax incentives are justified by their potential to generate positive externalities that can compensate for the costs of extending these subsidies (Wade 1990; Margalioth 2003; Harrison and Rodríguez-Clare 2010). Policymakers cite potential knowledge spillovers as justification for incentives to attract foreign firms. Attracting a large manufacturing facility, for example, may result in the transfer of skills and technology to firms that supply and distribute products from that facility (Zolt 2014). A related justification for use of incentives relates to the incomplete ability of appropriate rents of domestic firms when they invest in innovation. Inviting foreign firms with the necessary know-how and customer relationships would fill this gap. The basic argument is that because firms typically under-invest in innovation, they are unlikely to recoup the initial fixed investment when the intangible knowledge results become available to the wider population of firms, effectively becoming a public good and opening the possibility of learning by imitation. Fiscal incentives can motivate enterprises to invest in innovation by reducing their tax liability and the cost of the investment. Evidence The evidence of any economic benefits from promoting export-oriented FDI is scant, and for developing countries, the evidence that these policies are effective in promoting exports is mixed. A review from the global investment competitiveness report (Andersen et al. 2018) suggests that for Singapore, Taiwan, Thailand, and Korea, the use of tax incentives policy has yielded new export-oriented investments and contributed to their industrialization—but in all these countries, tax incentives have not been introduced in a vacuum but as part of a holistic strategy (Wade 1990; Tanzi and Shome 1992). However, there is evidence that the use of tax incentive policy has resulted in mediocre new investment or none at all (IMF 2015), and found it to have limited effectiveness at the aggregate level (Allen et al 2001; James 2009; James and van Parys 2010; Klemm and van Parys 2012; van Parys 2012). In fact, some studies have found that their use can even be detrimental: tax concessions in the Eastern Caribbean Currency Union not only attracted little if any FDI but instead aggravated fiscal deficits through erosion of the tax base (Chai and Goyal, 2005). Tax incentives have worked better when they target investors who are more concerned with cost efficiency than other types of investors, such as those seeking resources or market access. The specific country context matters if tax incentives are to have an effect, such as whether there is a supportive business climate (James 2009), sound infrastructure (Bellak et al. 2009), and reasonable logistics and transportation costs (Kinda 2014). Based on such findings, Gondor and Nistor (2012) concluded that “a low tax burden cannot compensate for a generally weak or unattractive FDI environment”. Efficiency- seeking investment also considers conditions in the host country, such as the business environment, and incentives are unlikely to overcome these requirements. However, although investment decisions were not 37 driven by incentives, they do have an effect at the margin of decision-making, e.g., when choosing between similar alternatives. The Andersen et al. review (2018) concluded that good practices for advancing tax incentives are transparency and decent governance of the incentives scheme. Supplier Development Programs Supplier development programs (SDPs) help link supply and demand within global value chains by identifying opportunities and assisting suppliers and potential suppliers to upgrade so that they can meet the demands of large buyers. SDPs are commonly used to facilitate links between domestic SMEs and foreign investors, or other large firms present in the country. Critically, any intervention to assist a current or potential supplier should be informed by a diagnostic of that firm and an improvement plan to help them meet the requirements of relevant buyers. The intent of SDPs is to induce knowledge transfer, upgrade supplier capabilities, enable SME access to markets, and increase supplier access to finance. SDP interventions use advisory services, matching grants, and provision of information to help suppliers to better respond to buyer sourcing needs. Justification The justification for programs that support supplier development typically rests on the need to address coordination problems. For example, the channels for communication between foreign investors and domestic SMEs are not always clear, and are often complicated by, e.g., work standards, organizational practices, and business culture. Both foreign investors and domestic companies also suffer from information asymmetry, for example. in how aware and confident investors are in the capabilities of suppliers. Also, many policy makers have begun to use SDPs on the assumption that foreign investors possess frontier technology that, in the right conditions, could permeate local SMEs, effectively serving as a mechanism for knowledge transfer, with an obvious market orientation. And proponents of these programs have often argued that entry of competitive foreign players into the local economy can stimulate competition in the domestic market, raising production standards and the performance bar. Evidence SDPs in developing countries have not been studied much, so there is no evidence on which to base conclusive statements about their effectiveness. A case study often cited is Chile’s Supplier Development Program, which was designed to strengthen established links between local SMEs and large exporting firms. The program, which ran from 2003 to 2008, offered matching grants for SMEs to upgrade their management skills. The impact evaluation found that it increased revenues, employment, and wages, and the survival of SMEs (Portugal 2018). A study of Costa Rica’s PROVEE program, in effect from 2001 to 2014, found that each year it led to 126 new product and service links. The first supplier-buyer transactions stemming from the new links alone generated a total cumulative value of over US$80 million (Cirera et al. forthcoming). And a study of the Czech Supplier Development Program found that in 2003, a year and a half after the program began, 15 of the 45 companies participating had additional business revenue attributable to the program in the form of contracts worth about US$18 million annually. A few companies had also expanded operations to overseas locations, and three stated that the program had led to engagements with higher value-added content. An assessment by Metz et al. (2017) suggests that central to any successful supplier SDP are a core team to lead and manage the program through its different stages and to facilitate relationships; a diagnostic of current and potential suppliers to inform the firm interventions to be supported; a pool of industry-specific experts to act as consultants and advisors; and funding or in-kind support for suppliers to make necessary improvements. Also crucial for these programs were analytical advice in understanding market 38 requirements, individual performance, and root causes of gaps; analysis of investor needs and requirements; a supplier database; and methods for screening and matching services. Export Promotion Programs for Domestic Firms Export promotion programs are “specific measures that generally amount to the government bearing a portion of the private cost of production of exports” (OECD 1984). These measures and programs seek to support exporters, current and potential, who want to internationalize their products; they use a range of policy instruments, among them grants, direct transfers, indirect incentives like tax credits and drawbacks, export finance, technical assistance, information and advice, and export processing zones (EPZs). Other measures do not require direct market interventions but rely on public inputs to promote exports; these include reducing requirements for export licenses, easing technology controls for exported goods, reducing antitrust concerns in the export sector, reducing exchange rate volatility, and building a sound NQI system. Justification Justification for export promotion programs comes in a variety of forms. Exporter firms are generally more productive than non-exporters (Bernard and Jensen 1995, 1999, 2004; Melitz 2003). The general argument is that for trade to be welfare-enhancing, there often need to be present certain conditions for international trade to function well, such as competition and access to market information. If there are no distortions, productive resources would then be reallocated from less to more efficient firms. However, the government should intervene to reduce distortions when there are signs that any of the necessary conditions are missing—without government intervention the number of exporting firms may be suboptimal. The following justifications go well beyond mere correcting of markets. In learning by exporting (Krugman 1987), firms improve their productivity after entering export markets. Learning by exporting presents positive knowledge externalities, and it comprises a myriad ways in which exports can induce growth in productivity, such as building up exporter capabilities, marketing new products, upgrading product quality, and acquiring expertise in managing customer relationships by dealing with foreign buyers. Virtuous learning from foreign buyers can be channeled directly to the exporters or passed through to domestic suppliers and competitors. Other justifications are Marshallian externalities and protection of infant industries (Harrison and Rodrìguez-Clare 2009). These justifications cite potential spillovers that increase with industry size. The drivers could be industry-level spillovers (Marshall 1920), upstream and downstream links (Krugman 1991), or technology-based economies of scale, which may reveal latent comparative advantages in a particular sector. A similar but more general justification relates to externalities of self-discovery (Hausmann and Rodrik 2003). Evidence There is evidence that export promotion programs revolve around export promotion agencies (EPAs), export promotion policy instruments (programs), and their strategies for targeting beneficiary exporters. EPAs have proved to be generally effective, and their presence has had positive effects not only on exports but also on income per capita (FERDI 2017, Lederman 2010, Volpe, 2014). Olarreaga (2015) suggests that on average one dollar spent on export promotion generates 15 dollars of exports. Among EPA features that seem to be associated with higher exports are representation of the private sector in the governing board and targeting only a few sectors (FERDI 2017). In a related study, Lederman et al. (2007) found that proliferation of EPAs in a country (e.g., through decentralization) had hurt exports. There is no evidence that funding foreign offices for EPAs has any impact on exports. 39 An IGC (2011) review of the literature on export promotion programs found that services provided by trade promotion organizations (TPOs) had a positive impact on export volumes. The FERDI (2017) study found that increases in income per capita were associated with TPO efforts to market and build the country brand. Evaluations of the effect of trade fairs and trade shows on exports suggest that they have had little impact in developing countries; however, access to market studies and outreach to prospective clients do seem to be effective. Lederman also found evidence that investment in domestic activities, such as export-supporting services, were more productive than off-shore activities, although the finding was not statistically significant. The IGC review, citing Volpe, suggests that offering bundled services through the entire export process has been critical to success. It also points out that credit and export guarantees have been effective in increasing the probability to export, particularly for SMEs. Evidence of the effectiveness of credit financing of working capital and loans for fixed investments is scant but positive. For example, beneficiary satisfaction has been high (Colombia, Macario, et al. 2000, and Malaysia, Hashim, et al. 2008). The IGC review suggests that the effectiveness of export subsidies, in the form of direct transfers, indirect duty drawbacks, and tax exemptions is questionable, and their impact on export performance is country-specific (based on Nogues 1989, Moreira et al. 2002, Helmers et al. 2009). Duty drawbacks seem to be the only instrument with a generally positive track record for promoting exports, particularly when used by SMEs (based on Macario 2000, Agosin 2001, Kate et al. 2000). Data on EPZs is hard to come by, and when available it has shown questionable impact on exports and jobs, and few economic returns. With a few exceptions (e.g., Mauritius, Ghana, Lesotho, and China) the data are generally disappointing. The IGC review suggests that most policy makers from developing countries that rely on EPZs have not seen the results they were expecting. FERDI (2017) also documented that spending a larger share of EPA budgets on established exporters increased exports but that investment in potential first-time exporters showed a positive but short-lived effect on export growth. However, Lederman (2010) found that the share of the budget spent on established exporters was linked to fewer exports. Later, Olarreaga (2015) found that programs that prioritized new exporters, rather than either occasional or established exporters, had higher returns. Clearly, the results are mixed, and contextual factors may be affecting how effects materialize. In terms of the size of exporters, the evidence is also mixed. Olarreaga (2015) argued that programs that target medium-sized firms perform better than those directed to small or large firms. Lederman (2010), however, argued that exports were shown to increase when large firms that are not yet exporting were targeted. This discussion, however, does not address the issue of additionality of impact, and the evidence considered was not experimental. The FERDI (2017) study indicates that support for nontraditional rather than general exports paid off. The finding of Volpe and Carballo (2015) was similar; in surveying Costa Rican exporters, they found that exporters of differentiated goods that participated in the PROVEE program saw their sales rise along the extensive margin with a strong expansion of the number of countries they export to (although the program did not motivate exporters to start selling a variety of products abroad). However, they found no impact on exporters trading reference-priced and homogeneous goods (Martincus and Carballo 2012) Other factors are known to affect export performance: Trade restrictiveness and the volatility of exchange rates have been documented as having negative and statistically significant effects on exports (Lederman et al. 2007). The availability of standards and a system of supporting infrastructure of testing and conformity assessment services to upgrade national quality standards and the general business climate in the exporting country are also critical. Among other factors are the cost structure of the goods, the reputation of the country, and collaboration between public and private agents. A recent World Bank study by (Metz et al. 2017) found that programs that combine capabilities upgrading with market links may be more effective than those that focus only on capabilities (e.g., consulting services, mentoring) or markets (e.g. trade fairs, market studies). 40 References Belloc, M; and Di Maio, M. 2011. “Survey of the literature on successful strategies and practices for export promotion by developing countries” International Growth Centre (IGC). Belloc, M; and Di Maio, M. 2011. “Successful strategies to help developing countries boost Exports” International Growth Centre (IGC). Irwin, D. 2019. “Does Trade Reform Promote Economic Growth? A Review of Recent Evidence” Peterson Institute for International Economics. Lederman, D., Olarreaga, M and Payton, L. 2007. “Export Promotion Agencies: What Works and What Doesn’t” World Bank Policy Research Working Paper 4044, Olarreaga, M. Sperlich, S, and Trachsel, V. 2015. “Export Promotion: what works?” Foundation for studies and Research on International Development (FERDI); JEL CODES: F13, C14. Schmidt, F. 2015. “Export Assistance: A literature Review and Challenges for Future Research”. Institute for Applied Economic Research (IPEA). Volpe, M. and Carballo. J. 2012. Export promotion activities in developing countries: What kind of trade do they promote? The Journal of International Trade & Economic Development, 41 Annex 2: Challenges to Future Performance (Current Sectors) In 2016, of all exports, goods constituted the highest share, except for wholesale exporters,. Firms with foreign ownership accounted for 50.2 percent of exports, domestic private firms for 42.1 percent, and State Owned Enterprises SOEs for 7.7 percent). Figure A2.1. Foreign Revenue by Industry and Firm Ownership Type, 2016 Source: SBRA data, World Bank calculations. To get a deeper sense of capability needs and operating issues, we studied the following prominent export sectors: (1) services: transport, creative, programming and information, and professional services; and (2) goods: food and beverage, wood and furniture, computer and electrical machinery, and motor vehicles and transport equipment. The selection was based on the Ministry of Economy selection of priority sectors and was confirmed using the following criteria: importance, tradability, revealed comparative advantages, and degree of foreign ownership. Figure A2.2. Ownership of Exporting Sectors, Ownership 2016 by exporting industries in 2016 (n= 18,875; Total RSD 1,953 Million) 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Retail Motor vehicles, transport man. Education manufacturing (furniture) Professional/technical services Restaurants, cafes, catering Metals, materials, petro, pharma Food, bev., tobacco manufacturing Mining IT, programming, consulting Tour agency Construction Telecommunications Wood, paper and printing Water supply, sewerage, waste Other services Health Administrative and support Publishing and media Agriculture Accommodation Computers, machinery, electrical Transport Other primary (forestry, fishing) Electricity, gas, steam and air Textiles, apparel and leather Wholesale Domestic (de novo) Foreign (de novo) Privatized (foreign) State-owned (20%+) Privatized (domestic) Mixed SOE and Foreign Source: SBRA data, World Bank staff calculations. 42 Since 2006, the products sector grew faster than service sectors, with CAGRs of 18.7 percent for motor vehicles. 3.9 percent for computers and electrical machinery, and 3.1 percent for furniture. The only service sector whose CAGR stood out from the pack was programming and information at 10.5 percent. Figure A2.3. Export Contributions to Output Export contribution in selected Selected Growth, to revenue growth Sectors, sectors vis-à-vis sectors 2008 and 2016 CAGR 2008-2016 (%) 18.7 17 Service sectors 12.0 12 10.5 7 5.7 3.9 3.1 1.0 1.6 1.4 2 0.2 Motor vehicles and Computer & Wood and furniture Food and Beverage Transport services Programming (0.3) and Professional Creative industries All others -3 tranportation electrical machinery information services services (1.2) equipment (0.3) (0.5) (3.6) (1.9) (6.0) -8 Product sectors -13 -18 (16.3) Exports Output Source: SBRA data, World Bank staff calculations. Growth in exports seemed to have driven output growth for motor vehicles, automotive and electrical machinery, and food and beverage sectors. Transport and creative industries growth fell significantly. Figure A2.4. Exports Exports Export and sectors from selected Growth, Selected Sectors, 2006 (2008) to 2016 2006-2016 Real RSD billions; Percentage (%) 400.0 110 350.0 90 300.0 70 250.0 50 200.0 30 150.0 10 100.0 2008 2009 2010 2011 2012 2013 2014 2015 2016 50.0 -10 - -30 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Food and Beverage Wood and furniture Computer & electrical machinery Motor vehicles and tranportation equipment Transport services Creative industries Programming and information services Professional services All sectors Source: SBRA data, World Bank staff calculations. 43 A dynamic analysis of exports for these sectors between 2006 and 2016 found that there was profound growth for computers and electrical machinery, food and beverages, and motor vehicles, with dramatic spikes in 2012, 2015, and 2016. Exports of transport services hit a low in 2012 but recovered somewhat in the next three years. Creative industry exports have been declining steadily, and furniture and professional service exports have been flat. Figure A2.5. Value-added and TFP Growth, Selected Sectors, 2006 (2008) average 2008-2016 to 2016 Percentage (%) Percentage (%) 60 40 50 30 40 20 30 10 0 20 2008 2009 2010 2011 2012 2013 2014 2015 -10 10 -20 0 2008 2009 2010 2011 2012 2013 2014 2015 2016 -30 -10 -40 -20 Food and Beverage Wood and furniture Computer & electrical machinery Motor vehicles and tranportation equipment Transport services Creative industries Programming and information services Professional services All sectors Source: SBRA data, World Bank staff calculations. Productivity has been strong for firms exporting programming and information services, wood and furniture, and motor vehicles, but for exporters of professional services and creative industries in recent years there has been a drastic decline, raising concerns about the competitiveness of service exports that rely on highly skilled workers. Figure 15 in the text shows that TFP growth plunged in 2014 and 2015. However, the results from business registry data for service exports data must be interpreted carefully, because information for these is not as readily available or reliable as information for goods. The positive performance of manufactured exports has been partly driven by efficiency-seeking FDI; these investors decided to locate in Serbia because of its favorable price-to-quality skills ratio, based on the country’s capabilities, traditional expertise, and good agroclimatic conditions. It is thought that Serbia has been effective so far in attracting foreign firms seeking to expand operations by tapping into medium- level technology capabilities. Many foreign companies have been active in machines and equipment, motor vehicles, and FMP. Such firms were especially prominent in the food industry, textiles and garments, and motor vehicles and machinery. Growth in medium-technology manufacturing has been brisk. A liberal FDI regime and generous tax incentives have also helped. Foreign companies enjoy equal treatment with domestic companies, with no discrimination and no limits on foreign ownership except for defense and a few other sectors. The privatization process has welcomed foreign companies. The law protects against expropriation and guarantees repatriation rights for foreign companies. In promoting investment, tax incentives are highly effective in attracting efficiency-seeking FDI, which is more motivated by low costs than other types of investment. 44 Although incentives have traditionally received considerable attention from policymakers, there is global evidence (OECD 2003; World Bank 2017) that they are not a major driver of investment decisions, though they often have influence at the margins when investors compare similar options. Serbian traditions of engineering and other technical skills, especially in mechanical design and construction and metal processing, have been instrumental in attracting foreign exporters of parts for rotating electric plants and electric motors, generators, and generating sets. The growth of food and beverages has benefited from favorable climatic conditions, rich and fertile land, and a recent injection of capital through privatization of crucial processing facilities. The Main Exporter Segments by Ownership The general pattern of cumulative exports revealed significant concentration, which is typical, with a few exporters accounting for most exports. Among all exporters, 85 percent were new domestic firms (16,111 firms), 11.6 percent were new foreign firms (2,187), and 1.6 percent were newly privatized but domestically owned firms (305). Table A2.1. Profile of Exporters by Ownership, 2016 Source: SBRA data, World Bank staff calculations. Although fewer in number, state-owned and newly privatized and state-owned exporters, were on average larger than domestic de novo firms. In 2016 about 11.3 percent of firms, 2,126 companies, exported more than RSD 105 million (~US$1million), with most being new domestic firms (1,313). However, the rate of firms exporting above this threshold within the ownership group was the lowest, at 8.1 percent; all others ranged from 20.3 percent to 63.9 percent. Exporters tend to invest intensively on R&D, as the proportion of firms in investing on R&D, among all firms, but when it comes to the intensive margin—the share of investment as a percent of total sales, the proportion is flat. Among all exporter groups, the share of innovative firms (those investing in R&D) in new domestic firms is the lowest among all groups (Figure A2.6). 45 Figure A2.6. Intensive and Extensive R&D Margins by Ownership Group, 2017 Source: SBRA data, World Bank staff calculations. For export margins, domestic new firms have larger intensive margins, especially those that invest in R&D. The new domestic firms that invest in R&D also had a higher proportion of new destinations. New foreign firms showed a similar pattern, but among those that invest in R&D, a larger proportion of their exports come from new products and new destinations. Privatized exporters show a higher proportion of new entrants, particularly if they invested in R&D. Figure A2.7. Intensive and Extensive R&D Margins by Ownership, Exporters, 2017, Percent Source: SBRA data, World Bank staff calculations. [[Delete percent signs on the y axis.]] 46 Exporter Constraints and Upgrading Opportunities This section relies on content from sector studies commissioned by the Government of Serbia as part of the Competitiveness and Jobs project. These studies shed light on issues affecting within-firm performance—issues rooted in either low-level firm capabilities or few incentives for firms to invest in upgrading. According to the business registry, domestic exporters tend to be represented in agribusiness, food and beverages, wood and wooden furniture, tourism and accommodation. and the creative industries.  In agribusiness and foods and beverages, land compartmentalization has contributed to fragmentation of agricultural production and lack of scale. Farms in Serbia average about 6 hectares (ha), far less than the EU average at 16 ha; small farms make it hard to benefit from increases in scale and productivity.  For the same sectors, lack of post-harvest and supply chain infrastructure prevent local firms from meeting the quality standards of European markets or penetrating attractive consumer niches. Limited agro-industrial facilities due to land scarcity is divided into even smaller agricultural plots, and institutional burdens make use of state-owned land suboptimal. Current post-harvest techniques and facilities, logistics, and market links make it impossible for the highest-quality fruits to reach international markets. Instead, most fruit is exported frozen, which lowers its value. All processing stages in local supply chains have losses due to obsolete machines and poor management. Additional investment is urgently needed to upgrade production equipment, transportation vehicles, and technology.  The unavailability of skills in design, craftmanship and branding prevents Serbian furniture exporters from participating effectively in downstream functions like product development and marketing that have higher value and are closer to the final customer. The dearth of workers with the required technical education and craftsmanship makes it impossible to increase production in the short term. The relative absence of experienced, successful designers and marketing experts prevents domestic firms from bridging the quality gap by tightening relationships with retailers and local showrooms.  Most furniture companies are SMEs with a limited track record and limited production skills. These skills are believed to be different from the skills needed downstream in the value chain. An integrated approach requires not only furniture-making skills but also skills in analyzing a market before entering and selecting an effective ratio of price-quality-design.  Size and absorptive capacity make the EU-28 the most important market for future growth. Serbian firms have potential to grow fruit exports and scale to new stages of added value by moving to retail- ready packaging and new processing technologies, such as freeze-drying and thermal processing. Organic production can be considered an opportunity to take advantage of fragmented production. EU consumers are increasingly concerned about health, ecology, and food safety, and are also turning more often to local products. Consequently, it has become vital for producers to master EU standards for control and quality, product labeling, and branding. Agribusinesses will need to position themselves as high-quality to penetrate high-value niche markets for fresh produce.  This implies a need to make available export support assistance and market intelligence services—to expand producer markets through links with foreign operators and customer-matching services; promote demand for Serbian produce abroad; empower Serbian producer associations; tighten up phytosanitary protocols; and work to meet the quality requirements of primary EU destination markets. It may also be necessary to reduce the fragmentation that is crippling productivity and support smaller domestic companies that find it difficult to achieve the required standards, have lower capacity and less cooperation between companies; and find it difficult to access markets because they have lower visibility, less negotiating power, and little access to information and contacts.  It will be important to facilitate expansion of production through capital deepening. Manufacturers need to invest in improving their equipment and in professionalizing their management so that they can increase their production capacity, productivity, and efficiency. 47  If cooperation and coordination between companies, perhaps through associations, were easier to achieve—which will require greater trust between companies and more confidence in contract enforcement—companies would be more likely to specialize in producing certain products.  It is particularly important to intensify investments, both private and public, in expansion of the knowledge economy and education about ICT, starting early in primary school. Programs should be created to support gifted pupils and foster their ICT abilities by offering specialized programming and gaming courses, so that they can prepare for the higher education of the future. Since there are social enterprises (privately owned) that offer such programs, the state could create public-private partnerships, identify gifted pupils with potential, and subsidize the cost of their education. The pay- off will be higher exports of ICT services.  It would also be useful to identify and penetrate niches and focus more on design and marketing. This would allow Serbian producers to conquer higher-price segments and have a stronger presence for their own brands in more demanding markets like Belgium, Italy, Denmark, and the Netherlands. A Problematic Operating Environment for Exporters We have synthesized some of the issues affecting exporter performance that relate to the business environment of Serbia as the home of domestic exporters and the host of foreign investors. These issues typically affect not only firm performance but also the allocative efficiency of capital and labor in the economy. These issues are discussed in three categories: poor knowledge transfer mechanisms (knowledge supply); absence of complementary conditions (enabling environment); and economic distortions, such as laws and market imperfections.  Poor governance arrangements in upstream agriculture and forestry supply has led to allocative inefficiencies and lack of control for critical export grade lumber inputs. For example, inadequate forest management by state-owned entities does not encourage sale of wood to domestic producers who can add value before it is exported. Much of the wood is exported as lumber, to the financial benefit of state-owned forests but the detriment of domestic companies that could use it. It appears that this limitation makes the supply of raw wood unreliable. There is no disclosure of quantities sold, to which buyers, or the logic used to allocate raw wood to different buyers. Producers emphasize that the biggest issue in acquiring raw wood is unpredictability, which makes planning production impossible.  Serbian policy makers should emphasize management of forest resources and enable firms producing wooden furniture to source raw material more efficiently.  Creative industries are undercapitalized and lack a unified strategy to articulate stakeholder priorities. Incentives mainly go to audiovisual services (filming and post-production), even though only the broader creative industry, where ICT, e-commerce, and the digital and sharing economy intersect, could catalyze change and build a more inclusive, connected society. Serbia needs the Strategy and the Action Plan for Creative Industry that the Prime Minister pledged a year ago. Even more pressing is the need for smooth implementation of measures to foster creative initiatives and inclusiveness.  Skills shortages and the low mobility of labor threaten the long-term viability of today’s Serbian economic development model. Low mobility not only limits the availability of labor to firms but it also means that skilled people continue to be un- or underemployed. This gap between the ability to produce and the difficulty in retaining talent, which depletes the number of skilled employees available to the private sector, is less extreme in neighboring countries. Having highly skilled labor available at very low cost creates a strong competitive advantage, especially in technically demanding industries that require a high degree of customization and service.  Minimal ICT infrastructure deters both the growth of key exporting service sectors and the opportunity to “servicify” manufacturing. Internet bandwidth in Serbia is below par, and without the appropriate fundamental infrastructure, investors will likely locate elsewhere. 48  According to the latest data on Gross Expenditure on Research and Development GERD per capita US (PPP) Serbia was one of the 3 countries of 35 in Europe that were least involved in R&D; its per capita spending barely reached US$100—Switzerland’s spending was 27 times higher.  Minimal private investment in R&D can reduce the potential for adaptation to international competitive trends, which demand the latest technological capacity. GERD as a percent of GDP in Serbia was just 0.28 percent; Slovenian companies invested 1.29 percent of GDP in R&D. However, Serbia is the top European country for GERD financed by higher education, which means that only academia has access to lucrative projects. Finally, in Serbia government-financed GERD is above the European average; 50 percent of all GERD is financed by the Serbian government, compared to 19 percent in Slovenia.  Weaknesses in the court system discourage investment. Courts are slow and inefficient: a civil case can take 635 days and impose costs that can reach 40 percent. Since lower courts lack capacity, issues tend to be solved at the appellate level. On average insolvency cases last more than 2 years. The laws related to competition are satisfactory, but the commission administering them is underfunded, leaving it vulnerable to interest groups. EU accession will put pressure on performance, but this may take quite a long time to materialize. 49 Annex 3: Destinations for Serbia’s Exports The Serbia of 2006 leveraged historical ties and relied on markets and connections that had made sense previously; by 2016 Serbia was beginning to change its strategy: exporters were seeking wealthier markets in richer countries and trade with EU countries was expanding. For example, Slovenia was the only former Yugoslavian nation that saw imports from Serbia go up; the fact that Hungary and Romania also accepted more goods imported from Serbia is probably because they are EU members. Table A3.1. Serbian Exports Beyond the Former Yugoslavia and Immediate Neighbors, 1975–2016, Percent Country 1975 1990 1998 2006 2016 15.1 12.6 14.0 12.2 Italy 7.7 percent percent percent percent percent 16.1 24.9 11.7 11.6 Germany 9.8 percent percent percent percent percent 23.3 14.7 USSR/Russia 4.3 percent 4.6 percent 5.2 percent percent percent Austria 1.8 percent 3.1 percent 1.6 percent 2.9 percent 3.2 percent Noncontiguous France 2.4 percent 4.6 percent 2.5 percent 3.4 percent 2.6 percent Destinations Czechoslovakia 6.2 percent 2.4 percent -- -- -- Czech Republic -- -- 0.6 percent 1.5 percent 2.2 percent Slovakia -- -- 0.6 percent 0.6 percent 2.0 percent Poland 4.2 percent 1.2 percent 1.1 percent 1.2 percent 1.9 percent USA 5.8 percent 4.4 percent 0.4 percent 1.1 percent 1.9 percent Switzerland 0.9 percent 0.8 percent 8.3 percent 0.7 percent 0.8 percent 68.4 71.2 43.7 39.8 43.6 Category Total: percent percent percent percent percent [[Source?]] [[Delete all percent signs throughout.]] Table A3.1 shows that Serbia’s exports went up to EU members Germany, Austria, the Czech Republic, Slovakia, and Poland. Italy, which had long been one of the top three destinations for exports from Yugoslavia. had become the top export destination for Serbia by 2006 and held that position for the next decade. Manufactured goods, which in 2006 had dominated exports to Italy with a 43.82 percent share of all exports to Italy, had by 2013 fallen to 20 percent, a share that was fairly constant through 2016. Opening of the Fiat plant considerably changed this dynamic, and the machinery and vehicles SITC 1- digit code rose from only 4 percent in 2006 to become Serbia’s top exports to Italy, making up 52 percent of all exports. Since 2013 the SITC 4-digit code for passenger motor vehicles has constituted 40–55 percent of Serbia’s exports to Italy and 7–12 percent of its exports of goods to all destinations. Figure A3.1. Serbian Exports Within Europe – a Shift to Wealthier Markets, 2006 and 2016 2006 2016 50 Germany, which had been the primary destination for Yugoslavian exports several times in the 1970s and again toward the end of the 1980s, in 2016 rose to Serbia’s second-highest export destination (Figure A3.1). In 2006, Serbia sent $600 million in goods to Germany, and by 2016 that had tripled to $1.8 billion. The export basket meanwhile evolved from being primarily comprised of the SITC 1-digit category of manufactured materials (33.15 percent), which included iron and steel (17.61 percent) and nonferrous metals (7.51 percent). to the 1-digit category of machinery and vehicles (52.93 percent), including electric machinery, apparatus and appliances, and parts (17.94 percent) and power-generating machinery and equipment (16.25 percent). The growth in value of exports in the machinery and vehicles SITC 1-digit category has been massive, from just under $100 million in 2006 to $947 million in 2016. This is very interesting not just because of the magnitude (a 10-fold expansion in 10 years is certainly impressive), but also because of the diversity of products that bulked up this category. Unlike in other markets where this category saw significant growth, in Germany the growth was led by machinery rather than motor vehicles. The diversity of the German export basket suggests that growth of this portfolio is sustainable and saleable, with room for considerable expansion into similar goods. This search for wealthier markets has even affected Serbia’s troubled ties with Russia, which as the USSR had topped Yugoslavia’s export destination charts in much of the 1970s and 1980s. Serbia’s exports to Russia now rose from $300 million in 2006 to just over $1 billion in 2013 before dropping to $800 million in 2016. Early goods were primarily manufactured materials (36.59 percent), including paper and pulp articles (19.88 percent) and chemical products (13.32 percent), such as medicinal and pharmaceutical products (11.81 percent). By 2016, manufactured goods (18.19 percent), such as apparel and clothing accessories (11.51 percent), and agricultural goods (35.17 percent), such as vegetables and fruit (28.34 percent), had been added to the export basket. 51