89608 Country Economic Memorandum Georgia: Seizing the Opportunity to Prosper Draft 5.indd 1 22/12/14 15:19:03 Draft 5.indd 2 22/12/14 15:19:03 Country Economic Memorandum Georgia: Seizing the Opportunity to Prosper December 2014 Macroeconomic and Fiscal Management Global Practice Europe and Central Asia Region Draft 5.indd 3 22/12/14 15:19:03 Draft 5.indd 4 22/12/14 15:19:03 Georgia - Government Fiscal Year January 1–December 31 Currency Equivalents (Exchange Rate Effective as of December 5, 2014) Currency Unit Georgian Lari US$1.00 1.8984 GEL Weights and Measures: Metric System Vice President: Laura Tuck Country Director: Henry G. Kerali Senior Practice Director: Marcelo Giugale Practice Director Satu Kristiina Kahkonen Practice Manager: Ivailo V. Izvorski Task Team Leader: Rashmi Shankar Draft 5.indd 5 22/12/14 15:19:03 Georgia Country Economic Memorandum | December 2014 Abbreviations & Acronyms B40 Bottom 40 Percent IDA International Development Association BEEPS Business Environment and Enterprise IDPs Internally Displaced Persons Performance Survey BCP Business Continuity Plans IFC International Finance Corporation BIS Bank for International Settlements IFRS International Financial Reporting Standards CAD Current Account Deficit JSC Joint Stock Company CAB Current Account Balance LCs Letters of Credit CASPAR Caspian Shipping Service Providers LFPR Labor Force Participation Rate CAGR Compound Annual Growth Rate MDGs Millennium Development Goals CEM Country Economic Memorandum MFI Microfinance Institution CIS Commonwealth of Independent States MSMEs Micro, Small and Medium Enterprises CPS Country Partnership Strategy NCTS New Computerized Transit System CTC Caucasus Transit Corridor NBFI Non-Bank Financial Institution DCFTA Deep and Comprehensive Free Trade Area NIIP Net International Investment Position EBRD European Bank for Reconstruction and NGO Non-governmental Organization Development ECA Europe and Central Asia NSDC National Skills Development Cooperation ECMT European Conference of Ministers of OECD Organisation for Economic Co-operation and Transport Development EIB European Investment Bank PISA Program of International Student Assessment ETI Enabling Trade Index PPP Purchasing Power Parity EU European Union REER Real Effective Exchange Rate FDI Foreign Direct Investment R&D Research and Development FTA Free Trade Agreement SMEs Small and Medium Enterprises GDP Gross Domestic Product SOE State Owned Enterprises GEM Global Entrepreneurship Monitor TFP Total Factor Productivity GEOSTAT National Statistics Office of Georgia TRACECA Transport Corridor Europe - Caucasus - Asia GIZ German Society for International Cooperation TTOs Technology Transfer Officers GEZI Customs Clearance Zone TSA Targeted Social Assistance GR Georgian Railways UNCTAD United Nations Conference on Trade and Development GVA Gross Value Added USAID United States Agency for International Development IADI International Association of Deposit Insurers VET Vocation Education and Training HHs Households VAT Value Added Tax IBRD International Bank for Reconstruction and WBG World Bank Group Development ICD Inland Container Depot WDI World Development Indicators ICT Information and Communications WITS World Integrated Trade Solution Technologies iv | Abbreviations & Acronyms Draft 5.indd 6 22/12/14 15:19:03 Georgia: Seizing the Opportunity to Prosper Contents Acknowledgementsix Executive Summary x  rowth and Shared Prosperity in Georgia: Why Trade? Chapter 1: G 1 A. Growth Pattern 2 B. Sources of Growth 4 C. The Drivers of and Constraints on Shared Prosperity in Georgia 7  ustaining Export Growth in Georgia Chapter 2: S 19 A. Export Dynamics in Georgia  20 B. Export Firm Survival in Georgia 26 C. Conclusions and Policy Messages 28 Chapter 3: Trade, Labor Outcomes, and Skills 31 A. Employment and Trade Patterns 32 B. The Labor Sophistication of Georgia’s Exports  36 C. Strengthening the Link between Exports and Labor Market Outcomes 42 D. The Skills Context in Georgia 47 E. International Best Practice in Skills Development 51 F. Conclusions and Policy Messages 55 Chapter 4: Logistics and Trade Facilitation 58 A. Transit Potential and Corridor Performance 59 B. Gaps in Services and Infrastructure  65 C. Trade Facilitation and Supply Chain Management 70 D. Conclusions and Policy Messages 74  onstraints on Firm Competitiveness: Beyond Tbilisi Chapter 5: C 77 A. Growth in Georgia: Sub-National Trends 79 B. Firm Competitiveness and Location 83 C. Decentralization and Firm Competitiveness in Georgia 87 D. Conclusions 91  rivers and Sustainability of External Imbalances Chapter 6: D 93 A. Basic Trends: Building Up External Liabilities  94 The Drivers: Brisk Growth in Credit and Imports Supported by High Capital Spending B.  97 C. External Sustainability 100 D. Conclusions 103 References105 | Contents   v Draft 5.indd 7 22/12/14 15:19:03 Georgia Country Economic Memorandum | December 2014 List of Figures Figure 1.1. Real GDP Index: Georgia and ECA Comparators 2 Figure 1.2. Contributions of Capital, Labor and TFP During Sub-Periods 4 Figure 1.3. Macroeconomic Snapshot 5 Figure 1.4. Job Creation and Destruction 7 Figure 1.5. Regional Trends, 2006–12 8 Figure 1.6. Shared Prosperity and Poverty Over Time 8 Figure 1.7. Poverty by Region 9 Decomposing Poverty Using Income Sources; Share of Poverty Reduction Accounted for Figure 1.8.  by Each Income Source 10 Figure 1.9. Quarterly Variation in Poverty Rates in Georgia 11 Figure 1.10. Quarterly Variation in Poverty Rates in Georgia 11 Figure 1.11. The ECA Shared Prosperity Framework 12 Figure 1.12. Dependency Ratios in ECA 12 Figure 1.13. Share of Working Age Population in the b40 12 Figure 1.14. Age Decomposition by Gender and Location 13 Figure 1.15. Human Capital and the b40, 2010 13 Figure 1.16. Borrowing from Banks 13 Figure 1.17. Labor Force Participation Rates: b40 Vs. t60 by Gender 14 Figure 1.18. Unemployment in the B40 Relative to the Overall Population 14 Figure 1.19. Share of Wage Income 15 Figure 1.20. Monthly Earnings by Sector and Gender in Georgia 15 Figure 1.21. Share of Income from Social Transfers 2010 16 Figure 1.22. Share of Income from Social Transfers 2010 16 Figure 2.1. Share of Exporting Firms in Total 20 Figure 2.2. Employment for Exporters and Non-Exporters (2006–12) 21 Figure 2.3. Turnover for Exporters and Non-Exporters (2006–12) 21 Figure 2.4. Key Characteristics of Exporting Firms 21 Figure 2.5. Characteristics of Single and Multi-Product Exporters 23 Figure 2.6. Characteristics of Single and Multi-Destination Firms 24 Figure 2.7. Productivity in Georgia Has Increased Systematically Since 2006 25 Internationalized Firms are More Productive in Georgia too, but their TFP Premium has Figure 2.8.  Fallen Over Time 25 Figure 2.9. Medium to Large Firms are More Productive 25 Figure 2.10. Distribution of the Average Spell Length in Years of a “Product-Destination” Combination 27 Figure 2.11.Average Export Value at Each Spell Length 27 Figure 3.1. Georgia’s Export Growth 33 Figure 3.2. Georgia’s Exports by Destination 35 Figure 3.3. Georgia’s Exports by Destination and Composition 35 Figure 3.4. Labor Sophistication of Georgia’s Exports over Time and by Destination 37 vi | Contents Draft 5.indd 8 22/12/14 15:19:03 Georgia: Seizing the Opportunity to Prosper Figure 3.5. EXPY Indicators Showing Labor Content of Georgia’s Exports Over Time and by Destination 38 Figure 3.6. Comparing Georgian Exports to Export of Other Emerging Markets in ECA 39 Figure 3.7. Cumulative Distribution of Georgia’s Exports to the World in 2000 and 2008 40 Figure 3.8. Cumulative Distribution of Georgia’s Exports to EU27 and ECA in 2012 41 Cumulative Distribution of Georgia’s Exports Compared to Exports from Estonia, Poland Figure 3.9.  and Latvia in 2012 42 Figure 3.10. Labor Force By Education and Location 48 Figure 3.11. Modern Firms are Particularly Affected by Skills Shortage 48 Figure 3.12. D ifferences between Share in Total Employment and Unemployment by Educational Attainment, 2012 49 Figure 3.13. Provision of Formal Training 50 Figure 3.14. Fewer Firms Developed New Products 54 Figure 3.15. Fewer Firms Spent on R&D 54 Figure 4.1. The Enabling Trade Index 2014—Georgia’s Infrastructure Ranking 59 Figure 5.1. Poverty Differentials by Region 77 Figure 5.2. Labor Productivity in 2012 81 Figure 5.3. Capital Intensity Pattern is Similar that of Productivity 81 Figure 5.4. Labor Market Conditions and Constraints on Mobility May Explain Variations in Regional Patterns 82 Figure 5.5. Public Spending by Region 89 Figure 6.1. Georgia’s External Imbalances 95 Figure 6.2. Financing the Deficit 96 Figure 6.3. Snapshot of Savings and Investment Trends in Georgia 97 Figure 6.4. Structural vs. Cyclical Components of the CAB and FDI 98 Figure 6.5. Some of the Drivers of Reliance on External Financing in Georgia 99 Figure 6.6. Key Balance Sheet Indicators 101 List of Boxes Box 2.1. Productivity Dynamics 25 Box 3.1. More is Needed on Innovation 54 Box 4.1. Wine Exports Supply Chain 71 Box 5.1. The Drivers of Regional Differences 84 Box 6.1. Should the Level and the Composition of Georgia’s Net Foreign Liabilities be a Source of Concern 102 | Contents   vii Draft 5.indd 9 22/12/14 15:19:03 Georgia Country Economic Memorandum | December 2014 List of Tables Table ES.1. Suggested Policy Actions xvi Table 1.1. Sectoral Composition of GDP and of Growth 3 Table 1.2. Proposed Policy Framework 17 Table 2.1. Exit from and Entry into Exporting in Georgia, 2006–12 20 Table 2.2. Exporters and Multi-Product Firms in Georgia 22 Table 2.3. Exporters and Multi-Destination Firms in Georgia 22 Table 2.4. Production Functions Estimates 25 Table 2.5. Main Policy Messages 30 Table 3.1. Labor Market Characteristics 32 Table 3.2. Worker Transitions Across Sectors and Into and Out of Labor Force Status in 2011 45 Table 3.3. Mobility Costs 46 Table 3.4. Main Messages 56 Table 4.1. Transit Potential and Composition of Transit Traffic by Product and Transport Mode 60 Table 4.2. Non-Liquid Bulk Cargo: Estimated Transit Potential and Actual Transit via Georgia  61 Table 4.3. Main Policy Messages 75 Table 5.1. Convergence of Regional GNI, 2006–12  79 Table 5.2. Regional GVA Share by Sector, 2012  80 Table 5.3. Labor Productivity in 2012  82 Table 5.4. Labor Productivity Heterogeneity 86 Table 5.5. Main Policy Messages 92 Table 6.1. External Accounts 94 viii | Contents Draft 5.indd 10 22/12/14 15:19:03 Georgia: Seizing the Opportunity to Prosper Acknowledgements The team would like to thank the Government of Georgia, especially the Ministry of Finance, the Ministry of Economy and Sustainable Development, the Ministry of Education and Science, the Georgian Revenue Service, the Central Bank, and GeoStat, for their support and collaboration. The report was prepared by a team led by Rashmi Shankar, Program Leader and Task Team Leader, and a core team including Congyan Tan (Economist, Chapters 1, 6, 7), Gregory Kisunko (Senior Public Sector Specialist, Chapter 1 and 7), Branco Ponomariov (Consultant, Chapter 1 and 7), Nistha Sinha (Senior Economist, Chapter 1), Cesar Cancho (Economist, Chapter 1), Giorgia Dimarchi (JPA, Chapter 1), Gonzalo Varela (Economist, Chapter 2 and 7), Antonio Martuscelli (Consultant, Chapter 2), Daniel Saslavsky (Trade Specialist, Chapter 4), Claire Honore Hollweg (Economist, Chapter 3), Elizabeth N. Ruppert Bulmer (Senior Economist, Chapter 3), Andrés Rodríguez-Pose (Consultant, Chapter 5), Daniel Hardy (Consultant, Chapter 5), Carolina Monsalve (Senior Transport Economist, Chapter 4) and Hiroyuki Tsuzaki (Private Sector Development Specialist, Chapter 2, 3 and 7). Angela Prigozhina (Country Sector Coordinator for Finance and Markets), Leyla Castillo (Financial Sector Specialist), Natalia Tsivadze (Consultant) kindly shared a background report on access to finance that has informed this document, Special thanks are due to our peer reviewers at various stages of preparation: Samuel Freije-Rodriguez (Chapter 3), Jose Guilherme Reis (Concept Stage), Emily Sinnott (Concept Stage), David Michael Gould (Decision Point), Ana Margarida Fernandes (Chapter 2), Thomas Farole (Chapter 6), Ekaterine T. Vashakmadze (Concept Stage and Decision Point), and Jasmin Chakeri (Decision Point). Valuable comments were received from Juan Gaviria (Practice Manager, Transport and ICT), Mona Haddad (Practice Manager, Trade), Aurora Ferrari (Practice Manager, Finance and Markets), Mustapha Benmaamar (Senior Transport Specialist), Mona Prasad (Senior Country Economist), John Gabriel Goddard (Senior Economist) and Mariam Dolidze (Senior Economist). The report was prepared under the overall supervision of Ivailo V. Izvorski (Practice Manager) and Henry G. Kerali (Country Director). Sarah Babirye (Program Assistant), Ewelina Lajch (Program Assistant) and Tamuna Namicheishvili (Program Assistant) supported logistics and the preparation, production and publication of the final version of the report. | Acknowledgements   ix Draft 5.indd 11 22/12/14 15:19:03 Georgia Country Economic Memorandum | December 2014 Executive Summary “Georgia: Seizing the Opportunity to Prosper” suggests a path towards sustainable and shared growth. Georgia’s story is associated with three stylized facts: high growth with persistent unemployment—currently at nearly 15 percent after 10 years of annual growth that averaged above 5.5 percent; a doing business rank of 8 out of 189 countries achieved without recovery to 1990 levels of per capita income—suggesting a relatively difficult transition experience in spite of noteworthy success with several governance and business environment reforms; and obstinate socio-economic vulnerabilities—reflected in Georgia’s status as one of the poorest countries in the Europe and Central Asia (ECA) region of the World Bank1 with a relatively weak performance on reducing poverty and inequality. Georgia is well positioned to achieve its development objectives. Past history is not the ideal lens through which to view a country that experienced multiple crises—domestic, global, and conflict-related—most recently during three successive years starting 2007. Rather, given the country’s strong history of trade liberalization, structural reform, and commitment to market-based growth, the new opportunities offered by the Association Agreement (AA) and Deep and Comprehensive Free Trade Area (DCFTA) with the European Union (EU), and a pronounced policy shift towards inclusion starting around 2010, the future looks bright. The main challenge is persistent joblessness, which must be addressed to establish a sustainable basis for the pro-poor development model outlined in the Government’s Socio-Economic Strategy 2020. Georgia has grappled with persistent and high unemployment—the main driver of poverty—since independence and needs to grow approximately 52,000 jobs net2 over the next three years to meet even a modest target of 12 percent unemployment. Growth in the past decade was driven by capital accumulation with high dependence on public spending—including social transfers that helped reduce poverty and boost shared prosperity after 2010—and external borrowing. Productivity growth and foreign direct investment (FDI) was mainly in the non-tradables sectors, especially real estate and construction and net job growth was weak, even pre-crisis during the double digit growth years. Looking forward, sustaining high growth with poverty reduction will entail investment in human, institutional, and physical capital to ensure productivity growth in the tradables sectors, necessary for this small open economy to strengthen export competitiveness and leverage new market opportunities into higher employment and output, and lower poverty. This report, which is anchored in the Government’s Socio-economic Development Strategy 2020, explores the potential for improved export competitiveness to strengthen employment growth in Georgia and is intended to inform a policy agenda mainly focused on the demand side of the labor market. International evidence suggests that structural reforms need to gestate before labor market outcomes improve, and that far more targeted focus is needed to support productivity growth, especially in younger and more innovative firms 1 To facilitate comparison of poverty across countries, the World Bank computes comparable and harmonized consumption aggregates using latest available household survey data, which differ from the national poverty estimates and are benchmarked against regional poverty lines of $1.25, $2.50 and $5 per person per day. Cross-country comparisons show a high level of poverty in Georgia as compared to other countries. 2 Assuming no increase in the labor force participation rate. x | Executive Summary Draft 5.indd 12 22/12/14 15:19:03 Georgia: Seizing the Opportunity to Prosper that tend to lead on net job creation. Job creation is best served through actions supporting the development of the labor market: both on the demand-side, i.e. the actual expansion of production and jobs through measures targeting firm productivity—and on the supply side, i.e. building the capacity of the workforce to absorb additional demands through improved labor productivity and mobility. This report attempts to inform the policy discussion in this context, mainly by focusing on the demand side of the labor market and on export competitiveness. The challenge While Georgia has improved its performance on poverty reduction and shared prosperity since 2010, more needs to be done to strengthen the income generation capacity of the bottom 40 percent of the population, which will call for a comprehensive policy agenda targeting job growth and job-readiness. By the World Bank’s “shared prosperity” metric, or the relative growth in consumption of the poorest 40 percent of the population (the bottom 40 or b40), the marked improvement in recent years could be attributed largely to strengthened social safety nets. While this is not unusual in itself, what is unique about Georgia is that this is largely non-contributory social assistance. In fact, dependence on fiscal transfers is extraordinarily high relative to the rest of the ECA region, both among the b40 and the top 60 percent. Minorities and Internally Displaced Persons (IDPs) are at the core of the b40 and are especially reliant on state support. Sustainable poverty reduction will therefore depend on strengthening the income generation capacity of the b40. Supporting economic competitiveness—through actions targeting firm and labor productivity—and leveraging strengthened capacity to trade to overcome the limitations imposed by domestic market size will be necessary. Georgia’s high current account deficit reflects challenges that can be addressed through a shift towards new, export-oriented sources of growth. The structure, financing and sustainability of the current account correspond directly to current and future patterns of growth and economic competitiveness. High deficits in Georgia reflect consumption led growth and a high propensity for capital inflows—including FDI—to increase imports rather than generate exports, and have led to high borrowing needs and risk perceptions that inflate borrowing costs. While low savings have been a driver of the current account deficit, a sustained improvement in the current account balance would have to come from an improved trade balance for three main reasons. First, though its impact has been small, growing trade has had a significant and positive impact on Georgia’s current account balance. Second, real exchange rate movements do have an impact on net exports but this is weakened by the high degree of dollarization of the financial system, the reliance of firms on external financing, and the limited development of the tradables sectors. Third, while savings have picked up in recent years, investment needs remain high; and adjustments to consumption could be especially painful for the poor. Export survival is key Georgia’s export growth rates have been impressive, at well above 10 percent per year on average since 2000: however market survival—critical for deepening trading relationships and supporting sustained improvements in competitiveness—remains a challenge. During 2006–2012 Georgia’s exports tripled in nominal value and doubled in volume; at the same time, 75 percent of new export products and 70 percent of all exports products failed to survive past their first year. The average length of an active export product from Georgia is relatively short, though survival for longer than a year significantly increases the probability of sustaining exports. | Executive Summary   xi Draft 5.indd 13 22/12/14 15:19:03 Georgia Country Economic Memorandum | December 2014 Since an important difference between successful and less successful exporters is the former’s ability to maintain export relationships for longer periods, which allows deeper trade, export growth, and job creation, a closer look at the main drivers of export survival in Georgia is important. Greater capital intensity of production, associated with higher productivity as well, increases the probability of export survival. Firms with foreign participation are more likely to survive, possibly due better management, closer-to-the-frontier technologies, greater knowledge of international markets, and more capacity for and access to finance and market research. Network effects appear to be important as well. The more firms there are that export one product, or export to one destination, the more information there is about the specificities of the particular market. That information may spillover to new entrants or to diversifying firms. Increased information may also spill over to the financial sector, which may find it less risky to finance innovative activities of firms trying to diversify. Georgian firms rely mainly on growing exports of the same product or exports to the same destination. Sustainable export growth is typically associated with an expansion into new products and new markets (the extensive margin) and the extension of existing export relationships (the intensive margin). While Georgian exporters are no exception in terms of being larger and more productive than domestic oriented firms and grow faster than non-exporters in terms of employment and output, export growth in Georgia has been largely due to exporting more of the same products to the same destinations. Though there is noteworthy experimentation with new products, it remains largely small scale; new products made up 50 percent of the total number of exported products in 2012, but accounted for only 16 percent of total exports in terms of value. About 75 percent of the new products introduced were by new firms. With high costs of learning about foreign demand conditions, tastes, and product specification requirements, many firms engage directly in small international transactions without having a thorough understanding of the market. The fact that firms with higher initial exports survive longer in export markets seems to support this interpretation. Firms that start exporting at a greater scale are more likely to have undertaken market research. Increased trade and improved skills for more and better jobs Georgia’s export growth has mainly been in the primary sectors, reflecting the structure of the labor market and the relative lack of improvement in export sophistication. Export growth was driven primarily by chemicals, base metals, foodstuffs and vegetable products. This reflects both the nature of the labor market in Georgia and the relative lack of progression in terms of the skills content of exports, a measure of export sophistication that reflects the labor profile of the export basket. While Georgia’s labor force has a much larger share with secondary level education now than a decade earlier, labor outcomes are still dominated by the low wage, low skill primary sector, which absorbs 57 percent of the workforce. There has been limited improvement in skills and value added export sophistication, and it is largely concentrated in goods destined for non-EU non-Russia Europe and Central Asia (ECA*)3 rather than the EU27, where quality requirements may make it more difficult to gain entry for higher value added goods. The skills profile of Georgia’s exports has shown less dynamism than that of regional performers such as Poland or Estonia, which have progressed to higher value 3 Non-EU non-Russia ECA countries or ECA* include Albania, Armenia, Azerbaijan, Bosnia and Herzegovina, Belarus, Georgia, Croatia, Kazakhstan, Kyrgyzstan, Moldova, Macedonia, Montenegro, Serbia, Montenegro, Tajikistan, Turkmenistan, Turkey, Ukraine, and Uzbekistan. The EU27 includes Austria, Belgium, Bulgaria, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden and the United Kingdom. xii | Executive Summary Draft 5.indd 14 22/12/14 15:19:03 Georgia: Seizing the Opportunity to Prosper added manufactured exports. In fact, since 2000, the median wage, value-added and skill content of Georgian exports has changed very little. If Georgia is to benefit fully from trade in terms of improved labor market outcomes, policy actions to facilitate the reallocation of resources to higher productivity and value added sectors will be necessary. Attracting FDI and capital investment in the tradables sectors could help Georgia expand its product offerings up the value chain, which could spur dynamism in the export sector, with positive knock-on effects for employment and wages. Policies that reduce labor mobility costs would reduce distortions affecting labor supply decisions and increase labor market flexibility, enabling workers to adjust more quickly to changing market signals. An important pre-requisite for benefiting from improved economic relations with the EU and leveraging the free trade agreement with Turkey and other countries is that resources move quickly and at relatively low cost to more productive and export oriented sectors. The fact that mobility costs are especially high for entry into the manufacturing sector suggests that there could be a transition period over which policies targeting manufacturing job growth would have a smaller-than-anticipated effect in the absence of concerted policy attention. Georgia can draw on a rich policy experience to tackle the issue of skills mismatches, consistently identified as a driver of high unemployment and a constraint on labor mobility. The availability of skilled labor in Georgia has been identified as a “binding” constraint, especially by modern and international firms in Georgia. Simulations suggest that were the supply of jobs to increase sufficiently to absorb the unemployed, skills mismatches would still be consistent with 15 percent unemployment. The enrollment rate in tertiary education has declined in Georgia in the last few years and is now lower than in most ECA or EU countries. There remains considerable scope for improvement in vocational and technical education (VET) as well. The new education strategy (2013–2020) pledges to put more emphasis on quality of education, transferability of qualifications, elimination of “dead-ends” and counseling to help students find jobs. Strengthening the general education requirement for the VET track will also be critical since this determines students’ ability to keep adapting to ever evolving skills-related needs of firms. Both life-long learning and on-the-job training also need to be improved, especially in the context of Georgia’s aging demographic profile and high skills obsolescence in more modern sectors. In 2013, only 11 percent of the surveyed firms in the BEEPS reported offering formal training. A common characteristic of countries that have had successful skills development strategies is a long term vision for education and innovation. Germany, India, China, Korea, and Singapore provide pragmatic examples of how firms were incentivized to invest in training more people than they could hire. In the Netherlands, for instance, firms can claim up to 140 percent of their training expenditures as tax deductions. In 2006, the North Rhine-Westphalia state of Germany introduced training vouchers that cover up to 50 percent of the fees. Since innovative firms account for a large share of the jobs being created, there are high returns from fostering entrepreneurship and strengthened R&D. This is especially true in Georgia, where innovators outperform non- innovators to a greater extent than in the rest of ECA and along multiple dimensions and where development of new sectors with potential for export competitiveness will be as critical as supporting competitiveness of existing sectors for increased job opportunities. Linking academic research to the needs of the productive sector, as in Chile, Singapore, Spain and Brazil, or focusing on encouraging entrepreneurship from an early age as in Sweden are potential strategies to draw on. Ireland developed a focused action plan to support the development of a work force that could feed a young, growing, innovative private sector. Particular emphasis was placed on ensuring that basic skills, such as literacy and numeracy, are imparted to everyone. | Executive Summary   xiii Draft 5.indd 15 22/12/14 15:19:03 Georgia Country Economic Memorandum | December 2014 Reducing the costs of trade An important driver of trade competitiveness is the cost of logistics; in Georgia, the pending agenda is largely in transit given the challenge of realizing the potential of the Caucasus Transit Corridor (CTC). Located at the crossroads of Europe and Central Asia, Georgia is a transit country with the potential to connect several important economic regions in Asia and Europe. The Caucasus Transit Corridor (CTC) is a key transit route between Western Europe and Central Asia for the transportation of oil and gas, as well as dry cargo. CTC is part of the international and regional corridor TRACECA; an alternative to the north corridor running through the Russian Federation and Belarus, and the southern corridor running through Turkey and Iran, since the latter cannot handle cargo originated in Europe and United States. Potential transit flows are large when compared to current volumes handled on the CTC, especially in non-liquid bulk products. The overall value proposition of the CTC relative to other options—including the Iranian ports for non-US and EU originated cargo and the Baltic ports for the rest—has high potential, though better integrated systems along the border with Azerbaijan, where costs are reported as unpredictable, would be necessary. For cargo destined or originated in Central Asia, only 22 percent of the potential dry bulk transit cargo and 10 percent of the potential containerized transit is currently transported through Georgia. However corridor level planning and facilitation would be critical to realize this potential. Despite Georgia’s proactive investment strategy on infrastructure in recent years, there remains a pending agenda to strengthen logistics. Georgia’s transport policy is centered on maximizing private sector involvement in management of and investment in ports and airports, and more strategic prioritization of government sector investments, including the East-West Highway transit corridor and rural roads. Transport services have been liberalized in the railway sector as well. Going forward it will be necessary to take a “big picture” view of the sector and ensure that public investment fully accounts for synergies between different modes of transportation, and that the focus moves away from the preparation and implementation of large single-mode infrastructure projects. The development of a national transport strategy is ongoing and will help address the planning and coordination issues. The supply chain management industry is still underdeveloped and in need of capacity development. Third Party Logistics Providers (3PLs) that integrate operations with warehousing and transport are present in the country, but do not offer a full range of services. A reliance on typical transit operations is evident for the most part. Nonetheless, Georgia’s participation in controlling transit flows is rather limited (only 7 out of 100 containers entering and leaving Georgia are booked in the country). Companies manage logistics assets in house for the most part, and lack awareness of outsourcing and joint ventures opportunities. The most sophisticated segment of the market, comprised of a few firms, shows a high degree of vertical integration. There is a need to develop capacities in the sector through education, coordination, outreach and improved regulatory frameworks for the forwarding sector. No region left behind Over the last decade, Georgia has undergone rapid economic growth, yet Georgian regions have failed to benefit in an equitable manner. Solid growth rates and evolving trade patterns, and the development of large trade deficits have not been accompanied by regional convergence in terms of either income or productivity. A xiv | Executive Summary Draft 5.indd 16 22/12/14 15:19:03 Georgia: Seizing the Opportunity to Prosper case in point is Tbilisi, the capital, which contains approximately one third of the national population and accounts for half of the country’s GDP. The city-region’s per capita output levels are almost twice the national average and more than three times that of the most lagging regions. The ongoing decentralization process in Georgia also has implications for firm competitiveness and for regional economic development in general. The decentralization process does offer an opportunity to strengthen public participation, voice, and promote inclusive governance. Where successfully implemented, this process can support the specific needs of lagging regions and groups in Georgia, and improve the prospects for firms to compete and regions to develop. However, evidence tends to be mixed on whether impacts of decentralization are positive or negative, and suggests that the complexities of design, planning and sequencing are major stumbling blocks that require close consideration. If Georgia is to continue to harness the development potentials of global trade, it will be important to ensure a balanced matching of power, responsibility and resources to provide the regions, along with institutional capacity building. Only when regional government units are able to diagnose and evaluate bottlenecks in the regional environment, including shortcomings in local socio-economic conditions, identifying on the way appropriate actors and stakeholders to instigate reforms and balance the complex needs of the local territory, can real, sustainable development really begin to take place. The globalization process offers unprecedented opportunities for firms to grow and regions to prosper; but heightens the risk that some firms and regions will be left lagging behind unless local public services are strengthened, especially with a view to supporting small firms. Two questions are important in this context; the impact of regional characteristics vis-à-vis firm competitiveness, and how decentralization process will impact existing disparities. While firm specific characteristics matter more, location or place-specific effects are relevant productivity drivers. Local public expenditures, transport infrastructure, and human capital endowments are particularly important for the competitiveness of Georgian firms. As global trade continues to intensify—especially in view of the DCFTA with the EU—building capacities in less favored areas is imperative to ensure that existing regional disparities do not widen further. While the nature of Georgia’s firms, with a large proportion of exports and value added being derived from low-productivity sectors, affects the strength of the findings, the impact of improved public services in the regions is only likely to strengthen as firms evolve to levels that enable better exploitation of the positive externalities associated with knowledge spillovers and trade opportunities. Support for small firms should be a particular priority, given strong potential for job creation and rapid productivity growth. | Executive Summary   xv Draft 5.indd 17 22/12/14 15:19:03 Georgia Country Economic Memorandum | December 2014 Table ES.1. Suggested Policy Actions Policy Required Actions Strengthen export Strengthen export promotion activities to reduce entry costs associated with informational competitiveness asymmetries and contribute to improved export survival. Can be supported by strengthening the to facilitate provision of information about foreign consumers’ preferences, help with the identification of expansion of jobs potential buyers, and assistance in tackling the regulatory complexities associated with serving and output foreign markets. Reforms would benefit from a transparent system of monitoring and evaluation, Ensure all firms based on improved firm level and transactions data, to ensure that scarce public funds are put to have equal their best use. opportunities to Encourage foreign direct investment and facilitate interaction between foreign and domestic firms. build productive Inflows of foreign investment have been targeting services sectors such as construction and and innovative banking, and real estate. Investment promotion activities could be focused on reorienting potential capacity, diversify investors to tradable sectors and more innovative activities. exports, and move Further develop the national quality infrastructure (accreditation, metrology, standardization and up the value chain conformity assessment) according to DCFTA requirements to enable firms to adhere to international standards and strengthen institutional and technical capacity. Newer firms innovate but need support to scale up, survive and grow jobs. Strengthened access to finance, for example through innovation grants, will boost R&D and help firms adapt and adopt new products and technologies. Facilitating university—enterprise linkages would build a stronger national system of innovation to support the dissemination of basic knowledge from academia to the productive sectors of the economy, and from firms to universities. Support labor Strong emphasis on general education is critical in a world in which technologies come and go at a mobility fast pace since these allow workers to better adapt. through skills VET quality and appeal strengthened, ensure that tracking into vocational does not happen too development early, develop general education component of VET to improve the employability of students that are streamed relatively early in their lives, before they get the basic skills. Firm incentives to invest in training, including of older workers; stronger engagement with private sector on VET and life-long learning. Facilitate Strengthen cooperation and institutional framework with Azerbaijan: realization of Operationalize block train from Poti to Baku and beyond; including a transparent single negotiated transit potential rail tariff (ADDY) and accessorials. for both direct Deepen cooperation with CASPAR to address charges transparency and schedule reliability for export of ferry services. transport services and reduction of Operationalize bi-national Corridor Working Groups. overall trade costs Promote pilots of joint-ventures between shippers, forwarders and (alternative) companies providing shipping services in the Caspian Ensure sustainability of financing of infrastructure investments in the transport sector xvi | Executive Summary Draft 5.indd 18 22/12/14 15:19:03 Georgia: Seizing the Opportunity to Prosper GROWTH AND SHARED PROSPERITY IN CHAPTER 1:  GEORGIA: WHY TRADE? Georgia’s socio-economic indicators reflect that it was hit harder by the transition to independence than any other country in the region and that its reform history has been short and interrupted. Following independence, Georgia’s per capita income plummeted to 28 percent of its 1990 level by 2005, before recovering to about 78 percent in 2013. Growth has averaged over 5.5 percent since 2004, mainly because of the structural reforms of 2004–08. However weak initial conditions and the “interruption” provided by a triumvirate of shocks over three years—namely protests in 2006 and 2007, the closing of the Russian market in 2007 and the conflict with Russia, and the global crisis in 2008—manifested in sluggish net job creation, continued over-reliance on “aggregate demand” to generate growth, and challenging vulnerabilities in terms of poverty and inequality. Georgia remains one of the poorest countries in the Europe and Central Asia (ECA) region4. At a Gini of 0.39, inequality is also relatively high by ECA standards. Increases in total factor productivity (TFP) and capital accumulation, mainly in the non-tradables sectors, have been the main growth drivers over the past decade. TFP growth has been the largest determinant of overall gross domestic product (GDP) growth in Georgia, and appears to have been largely concentrated in non- tradables—including services, construction, and manufacturing—though some of these investments, particularly in infrastructure, have no doubt contributed to the economy’s productive capacity. After the crisis, brisk public spending supported a broad-based recovery, and the decline in agriculture started to slow. While jobs were created, significant shedding of the work force in the public sector largely offset private sector employment growth, and labor accumulation played a relatively small role. Strengthening export competitiveness has been a challenge, which reflects both the underlying structure of the economy and the current sources of growth, and will call for a comprehensive policy effort. Faster growth in the non-tradables relative to the tradables sectors reflects the pending challenge of strengthening Georgia’s export competitiveness and underpins the large and persistent current account deficit. Higher productivity growth in non-tradables has contributed to real effective exchange rate (REER) appreciation, with adverse implications for competitiveness. These trends highlight the need for faster productivity growth and resource accumulation—of both labor and capital—in the tradables sectors, to support economic competitiveness and allow trade policy and closer ties with the European Union (EU) to translate into higher incomes, more jobs and improved development outcomes for the people of Georgia. 4 To facilitate comparison of poverty across countries, the World Bank computes comparable and harmonized consumption aggregates using latest available household survey data, which differ from the national poverty estimates and are benchmarked against regional poverty lines of $1.25, $2.50 and $5 per person per day. Cross-country comparisons show a high level of poverty in Georgia as compared to other countries. | Chapter 1: Growth and Shared Prosperity in Georgia: Why Trade?   1 Draft 5.indd 19 22/12/14 15:19:03 Georgia Country Economic Memorandum | December 2014 The World Bank Group has indicated as its twin goals the elimination of extreme poverty and boosting shared prosperity to highlight that both growth and inclusion are essential for development. Benchmarking absolute progress on poverty reduction and relative improvement in inclusion of the population at the lower end of the distribution in the growth process allows zooming in on the income generation capacity of the poor and the policy channels to strengthen this capacity. The WBG’s measure of shared prosperity is either the growth in consumption or in income of the bottom 40 percent (b40) of the population. This measure is relevant for comparisons across all income levels. Georgia’s track record on these key development goals has improved since 2010, but accelerating and sustaining progress remains a challenge. Performance on both poverty reduction and boosting shared prosperity has improved since 2010, following the fiscal stimulus that drove post-crisis recovery and the increase in pension and social assistance benefit levels and coverage. However net job creation, the basis of sustainable growth and inclusion, remains sluggish. Improving labor market outcomes remains a key challenge. The empirical evidence points towards improved trade prospects—the main focus of this report—through strengthened competitiveness as being a potential avenue not only to job creation and growth but also to greater shared prosperity. As demonstrated later in the report, the productivity growth associated with an enhanced ability to export is critical to job growth—the main channel to boosting shared prosperity and reducing poverty—in Georgia as well as to real wage increases. International experience also suggests the conditions under which trade is beneficial for poverty reduction and growth. This report therefore focuses on trade and trade facilitation and on the key elements necessary to support strengthened competitiveness and job growth. A. Growth Pattern An impressive track record of growth since Figure 1.1. Real GDP Index: Georgia and ECA 2004 has still left the country worse-off than at Comparators independence in 1990. As noted in World Bank 1990=100 250 (2013)5, growth in Georgia over the last twenty years has been characterized by four phases: collapse, 200 stabilization, acceleration, and finally, crisis and rebound. Following the breakup of the former Soviet 150 Union, Georgia experienced one of the sharpest contractions in output among transition economies. 100 By 1994, GDP collapsed to a mere 27 percent of its 1990 level (Figure 1.1), as widespread economic 50 disorder and civil conflict took hold. From 1996, a brief period of macroeconomic stability and intermittent 0 structural reforms enabled the economy to rebound ▬▬TUR 1990 92 94 ▬▬CIS-RR 96 98 ▬▬EU-10 2000 02 ▬▬SEE 04 06 ▬▬CIS-NR 08 10 ▬▬GEO 2013 and stabilize from highly depressed levels. Growth Source: World Bank staff calculations using GDP (PPP, 2011 international$) from the World Bank’s World Development Indicators. 5 Georgia Rising (2013), World Bank, Washington DC,. 2 | Chapter 1: Growth and Shared Prosperity in Georgia: Why Trade? Draft 5.indd 20 22/12/14 15:19:03 Georgia: Seizing the Opportunity to Prosper averaged 5.2 percent during 1999–2003, and then, following the Rose Revolution at end-2003 and far-reaching and broad-based reforms, accelerated to an average rate of 9.3 percent during 2004–07. This acceleration was halted by the twin shocks of the August 2008 conflict (preceded in 2007 by the closing of the Russian market) and the global financial crisis. The economy rebounded in 2010–13, with growth averaging just over 5.5 percent. As of 2013, Georgia’s GDP was at 80.5 percent of 1990 levels (Figure 1.1). The structure of the economy shifted towards services over the past decade, led by public administration, real estate, and financial services  (Table 1.1). In terms of sectoral contributions to growth, between 2004 and mid-2008, the main drivers were services, construction, and manufacturing, financed with substantial inflows of foreign direct investments of about 16 percent of GDP in 2007 (World Bank 2013a and b). The global financial crisis and August 2008 conflict led to a sharp fall in foreign direct investments, exports, and remittances fell. In 2009, the economy contracted by 3.8 percent. The government introduced a fiscal stimulus, which supported recovery starting 2010, underpinned by robust public spending. The recovery was more broad-based, with the decline in agriculture also starting to gradually slow and even reverse. Services, especially tourism and transport, and manufacturing were the main contributors to growth over the past three years. In terms of the share of GDP, manufacturing has stagnated while services are now at over two-thirds of the economy. In particular, the contribution to GDP of the public sector, real estate and financial services has increased steadily during the past 10 years. Table 1.1. Sectoral Composition of GDP and of Growth In percent of GDP and contribution to growth Share of GDP (%) Contributions to Growth (%, Period Avg.) Sectors: 1997 2003 2007 2010 2013 1998–03 2004–07 2008–10 2011–13 Agriculture 29.2 20.6 10.7 8.4 9.3 0.19 -0.33 -0.68 0.48 Manufacturing 10.7 9.3 9.6 9.2 10.6 0.31 1.46 0.33 1.47 Product Processing by HHs 5.4 4.6 3.2 3.0 2.8 0.06 0.48 -0.17 0.13 Mining and Energy 3.7 5.0 3.8 4.0 3.8 0.28 0.23 0.22 0.13 Construction 3.8 6.8 7.8 6.1 6.7 0.76 1.29 -0.25 0.25 Services 47.1 53.8 65.0 69.4 66.8 3.21 6.13 1.94 3.49 Of which: Trade, Hotel, & Restaurant 13.9 17.2 17.2 19.1 19.7 1.25 2.09 0.53 1.23 Transport and Communication 8.2 14.8 12.1 11.5 10.7 1.30 1.72 0.50 1.05 Financial Services 1.0 1.6 2.5 2.6 3.0 0.36 0.47 0.22 0.59 Real Estate and Rental 9.7 6.4 6.5 8.2 9.1 0.30 0.62 0.19 0.41 Public Administration 4.0 3.8 14.9 13.0 10.1 -0.02 0.16 0.13 0.09 Education, Health, and Social 6.7 7.8 8.5 11.5 11.0 0.11 0.77 0.53 0.19 Other Services 3.7 2.3 3.3 3.4 3.1 -0.09 0.30 -0.15 -0.07 Total GDP 100.0 100.0 100.0 100.0 100.0 4.82 9.27 1.88 5.41 Aggregate Demand Components: HH Consumption 101.5 71.6 70.3 74.3 70.5 -1.24 6.36 2.45 2.69 Government Consumption 10.2 9.8 21.9 21.1 16.7 0.41 4.68 0.06 -0.48 Gross Capital Formation 17.8 31.3 32.1 21.6 24.8 3.68 3.17 -3.00 2.56 Net Exports -26.5 -14.6 -26.7 -17.8 -13.0 1.24 -5.10 2.62 0.62 Total GDP 100.0 100.0 100.0 100.0 100.0 4.82 9.27 1.88 5.41 Source: World Bank Staff Calculations. Note: GDP at basic prices excluding net taxes | Chapter 1: Growth and Shared Prosperity in Georgia: Why Trade?   3 Draft 5.indd 21 22/12/14 15:19:04 Georgia Country Economic Memorandum | December 2014 Consumption has accounted for the largest share of GDP on the demand-side. After peaking at 106 percent of GDP in 2009, consumption—private and public—has settled back to about 87 percent of GDP in 2013 (Table 1.1). The other side of this coin—low domestic savings—has contributed to capital imports, dollarization of the banking system, and a high current account deficit. For the economy to generate the supply response necessary for sustained job growth and improved living standards, a further re-orientation towards investment in tradables and improving export competitiveness is called for. B. Sources of Growth Productivity increases and capital accumulation, Figure 1.2. Contributions of Capital, Labor and TFP concentrated mainly in the non-tradables sectors, During Sub-Periods have been the main growth drivers, with labor In percent 10 contributing relatively little. Over 1999–2012, TFP and capital accumulation were the main sources of 8 growth. While the contribution of labor has been smaller, reflecting fairly flat net job creation, it increased 6 6.32 post-crisis following the fiscal stimulus and growth 4 3.86 recovery. The concentration of productivity increases 3.65 in the non-tradables sectors is clear from the pattern 2 2.25 of TFP over time. High TFP growth during 2004–07 1.48 0.67 1.56 0 accounted for 6.3 percent out of an overall average -2.02 growth of 9 percent. This period saw rapid growth -2 in construction and services, with manufacturing JJTFP growth 1999–2003 JJLabor force 2004–07 2008–09 JJCapital stock 2010–12 retaining its share in GDP but not increasing it. The fall Source: Georgia Rising (2013), World Bank, Washington DC. in TFP over 2008–09—the main source of the decline in GDP after the onset of the global crisis—accounted for 2.02 percent out of the overall contraction of GDP by 0.78 percent over 2008–09. This period saw a shrinking of the construction sector and a sharp fall in services growth. Post crisis growth recovery was also mainly accounted for by TFP growth, which contributed 3.9 percent out of overall GDP growth of 6.5 percent over 2010–12 (Figure 1.2). The macroeconomic fundamentals reflect the underlying economic structure and current sources of growth and highlight the importance—and challenge—of strengthening export competitiveness. Low domestic savings relative to investment and high imports, especially of consumer goods, relative to exports, have led to persistent current account deficits, which have been financed—and sustained—by large capital inflows supported by high domestic interest rates. These inflows, coupled with weak capital market development and limited lari financing instruments, have led to high dollarization in the banking system. An appreciating real exchange rate, reflecting productivity growth and capital accumulation in the non-tradables sectors, has further constrained export competitiveness. While the persistent current account deficit suggests that the currency is over-valued, significant forex borrowing and lending would limit the positive impact of depreciation on export competitiveness while risking losses to firms that have currency exposure. Strengthening export competitiveness to contain and ultimately reverse the trade deficit will contain external imbalances, limit the need for a painful 4 | Chapter 1: Growth and Shared Prosperity in Georgia: Why Trade? Draft 5.indd 22 22/12/14 15:19:04 Georgia: Seizing the Opportunity to Prosper Figure 1.3. Macroeconomic Snapshot GDP Growth & Unemployment Inflation & Average Lending Rate In percent In percent In percent 14 18 35 12 16 30 10 14 25 8 12 6 20 10 4 15 8 2 10 0 6 4 5 -2 -4 2 0 -6 0 -5 2000 01 02 03 04 05 06 07 08 09 10 11 12 2013 Jun-02 Jun-03 Jun-04 Jun-05 Jun-06 Jun-07 Jun-08 Jun-09 Jun-10 Jun-11 Jun-12 Jun-13 Mar-14 JJGDP growth, lhs ▬▬Unemployment rate, rhs ▬▬Ave. lending rate in national currency ▬▬Ave. lending rate in foreign currency ▬▬Inflation rate Real Effective Exchange Rate CAB, Trade Balance, Saving & Investment LCU per US$, end-period Dec. 1995=100 In percent of GDP 1.0 140 30 1.2 120 20 1.4 100 10 1.6 80 0 1.8 60 -10 2.0 40 -20 2.2 20 -30 2.4 0 -40 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08 Jan-09 Jan-10 Jan-11 Jan-12 Jan-13 Mar-14 2001 02 03 04 05 06 07 08 09 10 11 12 2013 ▬▬Nominal ER ▬▬REER, rhs JJGross national saving JJGross domestic investments ▬▬Current account balance ▬▬Trade balance Foreign Direct Investment Dollarization of Loans & Deposits In percent of GDP In percent of foreign currency 18 90 16 85 14 80 12 75 10 70 8 65 6 4 60 2 55 0 50 2007 2008 2009 2010 2011 2012 2013 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08 Jan-09 Jan-10 Jan-11 Jan-12 Jan-13 Apr-14 JJAgriculture & fishing JJManufacturing JJEnergy & mining JJConstruction JJFinancial sector ▬▬Loans ▬▬Deposits JJReal estate JJHealth & social work JJTransports & comm. JJHotels & restaurants JJOther sectors Exports (2010–13) Imports (2010–13) In percent In percent Plastics, rubbers 4 Others 12 Others 2 Service exports 40 Chemicals, & Textiles, footwear, headgear 1 allied industries 7 Metals 7 Service imports 16 Machinery, electronics 1 Stone, glass 3 Mineral products 5 Transportation 10 Chemicals & Transportation 15 allied industries 6 Mineral products 16 Metals 13 Machinery & Food & drinks 14 electronics 14 Food 14 Source: World Bank staff calculations based on WITs and Geostat data. Chapter 1: Growth and Shared Prosperity in Georgia: Why Trade?   5 | Draft 5.indd 23 22/12/14 15:19:04 Georgia Country Economic Memorandum | December 2014 adjustment to consumption and/or investment in the event of a shock, and support the expansion in the tradables sectors necessary for net job creation. Re-orienting investment towards the tradables sectors will be especially important. Capital in Georgia has mainly flowed to non-tradables and has been import rather than export-oriented. For Georgia to be able to strengthen export competitiveness—and take advantage of closer economic ties with the EU—the tradables sectors must expand and this will take higher investment than available only from domestic savings. In many countries, FDI has helped develop export capacity by supporting market integration, inter-industry linkages, updating skills and technological modernization. This has not yet happened in Georgia to the extent needed, but can potentially be a source of high job growth (World Bank 2013c). Successful job creation in the ECA region has depended on there being a consistent track record of structural reforms, skills, and an enabling business environment to support firm productivity growth. The World Bank’s flagship report on jobs6 asks two questions: how do countries create jobs and which policies help workers access these jobs? The main findings of this study suggest that successful job creators in ECA made an early start on structural reforms, supported the development of an enabling business environment for entrepreneurship by facilitating firm entry, productivity growth, survival and expansion (or firm exit by allowing fast and cheap failure), and had incentivized, mobile and prepared workforce. Looking forward, Georgia needs net job creation in the tradables sectors to be the basis of shared and sustainable growth. So far, while Georgia has created jobs, the pace has been insufficient given the normal process of job destruction in declining sectors (Figure 1.4). For a small open economy like Georgia, export growth, especially of higher value added but labor intensive manufactured goods, is an important potential means to achieving brisker net job creation. Supporting firm productivity is a pre-requisite for employment growth and will call for renewed commitment to and deepening of structural reforms. New, innovative and modern firms are the main source of job growth in the ECA region. The business environment must enable these firms to emerge, survive or exit, and grow. This also means supportive macroeconomic fundamentals, in particular policies to encourage domestic savings and local currency financing for young firms. Supporting education reform and skills development, strengthening incentives to work and supporting additional growth centers outside of Tbilisi will facilitate labor mobility as well as higher labor force participation. This resource mobilization to higher productivity sectors will complement and leverage the market access gained by Georgia through the Association Agreement and the DCFTA with the European Union. 6 Back to Work: Growing with Jobs in Europe and Central Asia (2013), World Bank, Washington DC. 6 | Chapter 1: Growth and Shared Prosperity in Georgia: Why Trade? Draft 5.indd 24 22/12/14 15:19:04 Georgia: Seizing the Opportunity to Prosper Figure 1.4. Job Creation and Destruction Job Creation and Destruction in Georgia Net Job Creation by Sectors In percent of previous year’s employment In percent of previous year’s employment 20 30 21.9 15 20 16.6 14.8 14.8 12.0 12.7 10 11.2 10 8.9 7.2 11.0 10.8 9.2 4.5 8.7 2.6 2.11.3 1.7 1.0 0.2 5 5.6 0 4.9 -2.3 -0.7 -0.9 -0.7 3.8 -2.6 -3.1 -5.0 -4.9 0 -10 -2.3 -3.0 -0.2 -11.8 -10.2 -12.1 -13.4 -5.4 -16.1 -16.2 -5 -7.0 -20 -19.8 -9.9 -9.4 -10 -11.7 -30 -13.5 -31.8 -15 -40 2007 2008 2009 2010 2011 2012 2007 2008 2009 2010 2011 2012 JJJob creation JJJob destruction JJNet job creation JJAgriculture, forestry and fishing JJMining and quarrying JJManufacturing JJEnergy JJWater supply, sewerage and waste management Annualized Employment Growth by Sectors In percent 30 25.3 20 14.0 10.3 12.3 12.5 9.9 10.6 10.0 10 6.8 8.2 7.0 4.2 3.9 2.6 2.4 2.7 0.9 0.9 0.8 0.8 0 -0.5 -0.1 -3.0 -2.4 -4.7 -5.1 -5.3 -4.0 -6.2 -10 -7.3 -15.6 -20 -25.2 -30 1999–2003 2004–07 2008–09 2010–13 JJAgriculture JJIndustry JJConstruction JJTrade and repairs JJHotels and restaurants JJTransport and communications JJReal estate JJEducation, health, social, and personal services Source: World Bank Staff Calculations based on Geostat data. Note: Data include firms in non-service sectors only. C. The Drivers of and Constraints on Shared Prosperity in Georgia Georgia’s performance on shared prosperity relative to the rest of the region has not been stellar. Georgia has fared relatively poorly in terms of growth of consumption of the b40 relative to the total population over 2006–12 and is among the weakest performers in the ECA region (Figure 1.5). However while the global shock of the 2008–09 crisis affected all countries, Georgia did experience additional and idiosyncratic shocks—protests in 2007 and 2008 conflict with Russia, which interrupted structural reforms and led to diversion of resources towards rehabilitation and repair, and also led to a need to transition to new export markets. Chapter 1: Growth and Shared Prosperity in Georgia: Why Trade?   7 | Draft 5.indd 25 22/12/14 15:19:05 Georgia Country Economic Memorandum | December 2014 Figure 1.5. Regional Trends, 2006–12 In percent 14 12 10 8 6 4 2 0 -2 -4 RUS BLR MDA UKR ARM GEO ALB KSV MKD MNE SRB BGR HRV CZE EST HUN LVA LTU POL ROU SVK SVN TUR KAZ KGZ TJK JJGRW bottom 40% JJGRW mean all Source: World Bank Staff Calculations. However, unbundling the data reveals a marked improvement in Georgia’s shared prosperity performance in recent years largely because of the fiscal stimulus and higher social transfers. A closer look at the growth rates for the bottom 40 and top 20 suggests significant variation over time, with the period immediately following the 2008 crisis largely driving Georgia’s poor shared prosperity performance. During 2006–07, as the economy restructured and job destruction outweighed job creation, average household consumption expenditure barely recorded any growth for the population overall, but while the bottom 40 experienced a contraction, the top 20 grew by 2 percent (Figure 1.6). After 2007, there was a brief resumption in growth enjoyed by all (including the population in the b40) until the crisis hit in late 2008. In 2009, when the economy contracted, the average consumption expenditure contracted for all, although this contraction became visible only during 2009–10. The period between 2008 and 2009 saw the most unequal growth pattern where average household consumption expenditure declined only for the bottom 40, while the top 20 enjoyed 5 percent growth. When consumption contracted throughout the distribution over 2009–10, the (b40) were affected the most, with their average Figure 1.6. Shared Prosperity and Poverty Over Time Consumption growth, in percent In percent 2005 US$ 6 35 2,500 5.4 2,077 4 30 1,969 1,839 1,851 2,000 3.6 1,795 1,758 25 1,595 2 2.0 20.1 21.0 1,500 1.2 0.3 20 18.1 17.7 17.4 17.7 0.7 0 14.8 15 1,000 -2.0 -2 10 6.0 6.7 5.2 5.2 5.1 5.5 500 -4 5 3.7 -4.9 -6 0 0 2006–12 2006–08 2008–10 2010–12 2006 2007 2008 2009 2010 2011 2012 JJAll JJBottom 40% ▬▬Poverty headcount ▬▬Extreme poverty headcount ▬▬GDP per capita, rhs Source: World Bank Staff calculations using Integrated Household Survey data, several years. GDP per capita from WDI. Note: The national poverty line used here is an absolute poverty line to facilitate comparisons of poverty performance over time in Georgia. The 2012 values of this line were GEL 91.2 per adult equivalent per month and GEL 52.9 per adult equivalent per month (extreme or food poverty line). The differences between the national and ECA-harmonized poverty rates lies in the difference consumption aggregates used as well as the application of adult equivalence in the national measure as compared to per capita in the ECA-harmonized measure. 8 | Chapter 1: Growth and Shared Prosperity in Georgia: Why Trade? Draft 5.indd 26 22/12/14 15:19:05 Georgia: Seizing the Opportunity to Prosper consumption expenditure contracting by 9 percent (compared to 5 percent for the top 20). The fiscal stimulus and broad-based growth recovery in the post crisis period appears to have especially benefited the bottom 40, who experienced 4 percent growth (compared to 0.6 percent for the top 20). These trends are mirrored in the poverty data. Progress on reducing poverty and inequality has been slow, largely due to same factors that affected the poor shared prosperity performance: low net job-creation, limited urbanization, and high dependence on low productivity, subsistence agriculture. Poverty rates have stagnated between 18 and 21 percent between 2004 and 2011, peaking during the crisis. 2012 saw a noteworthy dip in poverty, reflecting higher social assistance levels and falling food and energy prices. The Gini coefficient has also remained fairly high by ECA standards, fluctuating around 39 over this period. The structure or pattern of growth in Georgia has tended to favor urban areas more than rural areas, and there are persistent regional disparities. Poverty is more responsive to growth as well as to changes in inequality in urban than in rural Georgia.7 Agricultural growth has lagged (making almost no contribution to overall growth) as most growth has come from services and industry (which favor urban areas). In 2011, 22 percent of rural residents and 13 percent of urban residents were poor, with 8 percent of rural residents facing extreme poverty as compared to 3 percent of urban residents. As with overall poverty, the depth of poverty is greater in rural areas (poverty gap of 7.4 percent of the poverty line and squared poverty gap of 3.6) than in urban areas (3.7 percent of the poverty line).8 Not only is rural poverty higher but nearly two-thirds of the poor are residents of rural areas. This distribution of the poor differs from the overall population shares residing in rural and urban areas (about 50 percent each). The probability of being poor, transient poor, or Figure 1.7. Poverty by Region among the b40 is also associated strongly with In percent labor market status and gender of the household 30 head. The poor and bottom 40 are also more likely: 25 24.2 24.7 (i) to live in larger households with a greater number 21.9 20.6 21.9 19.6 of dependents; (ii) to live in households headed by 20 18.8 someone with less than secondary education; (iii) to 15 16.7 17.1 be unemployed or economically inactive; (iv) to have 16.2 14.2 13.4 13.3 household heads who are less likely to be in paid work 10 10.5 and more likely to be self-employed (which is largely how subsistence farmers are classified); and (v) to 5 live in households headed by women. Among those 0 households where the head is unemployed, poverty 2006 2007 2008 2009 2010 2011 2012 ▬▬National ▬▬Urban ▬▬Rural rate is 24 percent as compared to 14 percent among Source: World Bank Staff calculations using Integrated Household Survey data, several years. households whose head is employed. Transient poverty and economic mobility are also very strongly linked to employment status, in particular unemployment. 7 Poverty-growth elasticity was -2.06 in urban areas and 1.52 in rural areas. 8 If transfers could be made efficiently then this pattern in the poverty gap suggests that the amount of transfers needed to bring rural poor above the poverty line is larger than that needed to bring urban poor above the poverty line. Chapter 1: Growth and Shared Prosperity in Georgia: Why Trade?   9 | Draft 5.indd 27 22/12/14 15:19:05 Georgia Country Economic Memorandum | December 2014 The main source of changes in poverty observed in the last few years were accounted for mostly by social transfers (pensions and social assistance) and to a much lesser extent by labor market indicators (earnings, employment)  (Figure 1.8). Using a poverty measure based on household income, simulations based on sources of income micro-decomposition show that most of the poverty reduction is explained by the old-age pension or targeted social assistance (TSA). Together social transfers account for 50 percent of the decline in the income-based poverty observed between 2006 and 2012, and 80 percent of the decline observed between 2010 and 2012. The second most important source of changes in the period 2006–12 is the category “other sources”, which includes income from sale or rental of property, loans, non-cash income and transfers from relatives. Figure 1.8. Decomposing Poverty Using Income Sources; Share of Poverty Reduction Accounted for by Each Income Source 2006–12 2006–08 Adj. PC/PAE 0.7 Adj. PC/PAE -0.5 Share of adults -2.7 Share of adults 3.0 Share of employed 6.9 Share of employed 5.3 Wages 12.8 Wages 11.4 Share self-employed -0.1 Share self-employed 2.4 Self-empl. earnings -2.2 Self-empl. earnings -2.7 Agric. income 2.5 Agric. income -2.7 Pensions & soc. ass. 51.3 Pensions & soc. ass. 76.0 Other sources* 30.6 Other sources* 6.8 Remittances 0.1 Remittances 1.0 -10 0 10 20 30 40 50 60 -20 0 20 40 60 80 In percent In percent 2008–10 2010–12 Adj. PC/PAE -1.7 Adj. PC/PAE 5.6 Share of adults -8.7 Share of adults 6.3 Share of employed -5.1 Share of employed -1.6 Wages -8.1 Wages 3.8 Share self-employed -19.9 Share self-employed -3.3 Self-empl. earnings -38.5 Self-empl. earnings -6.9 Agric. income 9.0 Agric. income 1.3 Pensions & soc. ass. 77.7 Pensions & soc. ass. 80.0 Other sources* 113.5 Other sources* 16.3 Remittances -18.1 Remittances -1.4 -40 -20 0 20 40 60 80 100 120 -10 0 10 20 30 40 50 60 70 80 90 In percent In percent Source: World Bank Staff calculations using Integrated Household Survey data. Note: (*) Other Sources: income from sale or rental of property, loans, non-cash income and local transfers from relatives. More households move out of poverty than fall into poverty over time (Figure 1.9 and 1.10). Quarterly poverty changes as well as movements in/out of poverty between 2007 and 2012 indicate that there is “churning” under the relatively unchanging overall poverty rates. For example, poverty rose from 17 percent in the fourth quarter of 2009 to 23 percent in the second quarter of 2010. Generally, poverty rates are the highest during the second and third quarter of any given year. This pattern suggests seasonal employment patterns. About 32 percent of 10 | Chapter 1: Growth and Shared Prosperity in Georgia: Why Trade? Draft 5.indd 28 22/12/14 15:19:05 Georgia: Seizing the Opportunity to Prosper those who were poor in 2009 were still poor two years later in 2011, while the rest of the 2009 poor moved out of poverty. Conversely, only 13 percent of those who were not poor in 2009 fell into poverty by 2011. Figure 1.9. Quarterly Variation in Poverty Rates in Figure 1.10. Quarterly Variation in Poverty Rates Georgia in Georgia In percent In percent 25 100 22.3 22.6 20.7 21.2 90 19.6 33.1 20 19.0 19.2 19.1 18.6 19.0 80 20.5 17.6 18.5 18.8 19.2 70 17.6 17.7 17.0 17.3 15 16.1 16.1 60 14.1 86.5 50 14.8 10 40 7.3 6.9 6.4 6.7 8.7 66.9 6.1 5.4 5.9 5.9 6.3 30 5.3 5.1 5.4 5 5.3 5.7 6.4 5.9 5.6 5.0 5.0 20 4.3 3.7 4.4 3.0 10 1.4 13.5 0 I II III IV I II III IV I II III IV I II III IV I II III IV I II III IV 0 2007 2008 2009 2010 2011 2012 Poor in 2009 Non-poor in 2009 ▬▬Poverty ▬▬Annual poverty ▬▬Food poverty ▬▬Annual food poverty JJStill poor JJLeaves property JJNever poor JJFalls into poverty Source: World Bank Staff calculations using Integrated Household Survey data, various years. Source: World Bank Staff calculations using Welfare Monitoring Survey (2009 and 2011 rounds). The Drivers of Shared Prosperity in Georgia In the interest of sustainability, the emphasis has to shift from state support through transfers to strengthening the income generation capacity of the b40 in Georgia. The Government of Georgia’s new Socio-economic Development Strategy 2020 emphasizes the need to support faster and more inclusive growth. To achieve this objective there is a need for sustainable sources of growth, backed by net job creation and strengthened economic opportunities to exit poverty and dependence on fiscal transfers. Creating these conditions, which would allow the b40—more likely to be inactive or unemployed- to use their relatively strong endowments productively, would boost progress towards shared prosperity. The b40 in Georgia, in fact, is rather well endowed in terms of assets (in particular human capital assets) but is not able to maximize their productive use through work. While Georgia is on track to achieving several of the Millennium Development Goals (MDGs) by 2015, important challenges remain, which limit the growth potential of the country and which could undermine the sustainability of the observed trends. There has been significant progress in recent years with declining infant and maternal mortality; full enrolment in primary education, with higher ratio of girls to boys in primary and secondary education; and increased proportion of the population with access to safe water sources. However, significant inequalities remain in terms of access to basic services and economic opportunities between the poor and marginalized population groups and the rest of the population (Georgia CPS 2014–17). While growth is not a sufficient condition for boosting shared prosperity it is a necessary one. While focusing only on overall GDP growth is not enough to ensure income growth for the poor—as self-evident in the Georgian case—the relationship between growth and reducing poverty and inequality is a positive one overall— as is clear from the developing world’s pre-crisis experience. Chapter 1: Growth and Shared Prosperity in Georgia: Why Trade?   11 | Draft 5.indd 29 22/12/14 15:19:05 Georgia Country Economic Memorandum | December 2014 An assets based approach that marries macroeconomic drivers with microeconomic characteristics is therefore appropriate for understanding the policy levers available for boosting shared prosperity. The approach is outlined in detail in Bussolo and Lopez-Calva (2014) and is summarized in Figure 1.11. The idea is to take neither a top down nor bottoms up approach, but rather acknowledge that income growth is simultaneously driven by individual characteristics and macroeconomic drivers. Figure 1.11. The ECA Shared Prosperity The central idea behind this approach is that Framework Georgia’s success at boosting shared prosperity Growth will depend on strengthening the b40’s endowment External conditions Sector composition Key prices of assets and the long-run productive capacity of the poor. The framework captures the idea that the household’s income depends on four main factors: (i) the stock of its assets, including human, physical, Sustainability Sustainability Income Intensity Net assets of use Prices Transfers generation natural, social and financial capital; (ii) asset use, capacity including reflection of skills or entrepreneurship in work, land use, and being economically active; (iii) asset returns, including wages, interest; and Distribution Change in income (iv) transfers, including public transfers and generation capacity remittances. Within this context, growth is seen as the Source: Bussolo and Lopez-Calva (2014). aggregate of the individual income generation capacity of all individuals and households in the economy. Figure 1.12. Dependency Ratios in ECA Figure 1.13. Share of Working Age Population in the b40 Ratio In percent 80 70 66.3 66.2 66.3 70 60 60 50 50 40 40 30 30 20 20 20.4 18.1 16.6 17.1 15.6 10 10 13.3 0 0 RUS SVK ARMMKDMNE KAZ UKR GEO ROU LTU POL EST HUN CZE SRB LVA SVN BGR MDA ALB KGZ KSV TJK TUR 0–14 15–64 65+ JJBottom 40% QQTop 60% JJAll JJBottom 40% JJUpper 60% Source: Bussolo and Lopez-Calva (2014). Source: World Bank staff calculations using 2012 IHS. Demographic trends in Georgia point to a diminishing share of the working age population relative to dependents, including among the b40. Dependency ratios are slightly higher among the b40. Overall the share of the working age population is projected to decline dramatically, with implications both for social safety nets and for the ability of the poor to enhance productive capacity. The main concern is in fact old age dependency. 12 | Chapter 1: Growth and Shared Prosperity in Georgia: Why Trade? Draft 5.indd 30 22/12/14 15:19:06 Georgia: Seizing the Opportunity to Prosper Figure 1.14. Age Decomposition by Gender and Location In percent In percent 70 70 66.3 65.6 67.1 66.3 66.8 65.8 60 60 50 50 40 40 30 30 20 20 20.2 18.1 18.0 18.1 19.4 17.0 17.2 16.3 15.6 15.6 10 12.8 10 13.8 0 0 0–14 15–64 65+ 0–14 15–64 65+ JJAll JJFemale JJMale JJAll JJUrban JJRural Source: World Bank staff calculations using 2012 IHS. Human capital endowment as measured by years of schooling varies across deciles but is relatively high for the b40 in Georgia, mainly because of a “cohort” effect. The share of the population in Georgia with only primary school enrollment is among the lowest in the region. This likely reflects a cohort effect, with older cohorts with more education having lost jobs in the SOE restructuring. But taking tertiary education as a measure of higher human capital endowments, generally resulting in higher likelihood to participate in the labor market and higher returns to work, the gap between the top 60 and bottom 40 appears larger. There is significant variation in Georgia’s tertiary educated 25+ populations between the top 60 and b40, though this gap is comparable with top shared prosperity performers such as Latvia, Poland, Bulgaria and Slovenia. In terms of financial capital endowments, which can be illustrated by indicators on access to credit, there is also unevenness in distribution. Individual bank borrowing in Georgia suggests differences between the top 60 and the b40, with the latter at a disadvantage in financial capital accumulation. High interest rates and collateral requirements may be a deterrent to small borrowers. Figure 1.15. Human Capital and the b40, 2010 Figure 1.16. Borrowing from Banks Share of population 25+ with tertiary education In percent 60 60 50 50 40 40 30 30 20 20 10 10 0 0 TUR ROUMKD ALB SRBMDAKSV BGR TJK MNESVN CZE POL HUN KGZ HRV ARMSVK UKR GEO KAZ LVA RUS EST LTU UZBKGZALBMDAUKRLTU LVA KAZMKDTJKGEOESTPOL AZEBLRARMTURRUSSRBKSV BIH CZESVKROUHRVBGRHUNSVNMNE JJBottom 40% QQTop 60% JJBottom 40% QQTop 60% Source: Bussolo and Lopez-Calva (2014). Source: Bussolo and Lopez-Calva (2014). Chapter 1: Growth and Shared Prosperity in Georgia: Why Trade?   13 | Draft 5.indd 31 22/12/14 15:19:06 Georgia Country Economic Memorandum | December 2014 Despite a relatively high level of human capital endowments, the b40 is under-utilizing its capacity to generate income through work and displays low labor force participation rates. Labor force participation among the b40 is lower than that for the top 60 and also exhibits significant gender disparities. The gap between the b40 and the top 60 percent of the population in terms of LFPRs is about 2 percentage points while the gap between male and female LFPRs among the b40 is about 22 percentage points, slightly higher than the overall gender gap by this metric Figure 1.17. Labor Force Participation Rates: Figure 1.18. Unemployment in the B40 Relative to b40 Vs. t60 by Gender the Overall Population In percent In percent 100 25 90 91.2 88.1 90.0 80 20 20.1 70 69.6 60 65.9 68.1 15 16.5 14.2 50 40 10 30 20 5 10 0 0 Female Male All Top 60 Bottom 40 JJTop 60% JJBottom 40% JJTotal Source: World Bank staff calculations using 2012 IHS. Source: World Bank staff calculations using 2012 IHS. Unemployment rates are significantly higher among the bottom 40. As noted earlier the poor are more likely to be unemployed or to depend on seasonal employment. Overall official unemployment—which understates rural unemployment—is over 15 percent, while unemployment among the bottom 40 is over 20 percent. Agriculture is largely of the subsistence type, which is reflected in high rural poverty rates. Nearly half the population depends on subsistence agriculture, which contributed about 9 percent of GDP on average between 2010 and 2013. Land use patterns suggest fragmentation of holding, though efforts are underway to establish a framework for consolidation and cooperation that may support productivity increases. Rapid growth in the wine sector is another positive development. Investment in agriculture has been low historically with some improvement in 2013 following a concerted policy effort by the government that is largely based on credit subsidies. This suggests that there is limited borrowing against agricultural assets for investment and that the agri-business environment needs far more focus. Returns to assets—be they the relatively high human capital assets or physical assets such as land— are generally low, given great reliance on informal and self-employment, and on subsistent agriculture, which reflects in the low share of wages in total income. Figure 1.19 suggests that the share of wages in total income is relatively low in Georgia for both the b40 and the top 60 percent of the population. Overall wages are low, especially in the primary sectors that generate the most employment, in spite of recent wage growth in agriculture. These trends are consistent with the high dependence on subsistence agriculture among the poor and with the relatively high dependence on social assistance in the overall population. Formal employment in Georgia is dominated by the public sector. 14 | Chapter 1: Growth and Shared Prosperity in Georgia: Why Trade? Draft 5.indd 32 22/12/14 15:19:06 Georgia: Seizing the Opportunity to Prosper Figure 1.19. Share of Wage Income Figure 1.20. Monthly Earnings by Sector and Gender in Georgia In percent Nominal GEL 80 800 70 700 600 729 697 60 500 50 558 400 504 481 40 300 404 376 365 342 334 303 293 280 30 200 269 235 215 20 100 139 0 organizations Financial Public adm., & defense Mining & quarrying Electricity, gas & water supply R&D, computer & related activities & communications Total Wholesale & retail trade, repairs Manufacturing Community & social services Hotels & restaurants Health & social work Education Private house- holds as employers Agriculture, forestry, fishing Extra-territorial intermediation Transport, storage Construciton 10 0 ALB ARM GEO HRV KAZ KSV MDA MKD SRB TJK TUR UKR JJBottom 40% QQTop 60% JJAll QQFemale ‹‹Male Source: Bussolo and Lopez-Calva (2014). Source: World Bank staff calculations using 2012 IHS. The relative importance of fiscal transfers in financing the consumption of the poor and boosting shared prosperity outcomes in Georgia is high relative to the rest of the region, though the picture is different if both social assistance and social insurance/labor benefits are included. The Georgian population overall— rather than just the b40—appears to depend significantly on non-contributory social assistance. Contributory social insurance on the other hand is non-existent in Georgia and relatively high in other countries. The bulk of social assistance in Georgia—80 percent—consists of pensions, which reach households across the whole distribution. Pensions therefore help explain the high share of consumption (nearly 30 percent) of the top 60 percent that is financed by transfers. Private remittances are also significant, at about 5 percent of GDP. Our findings are mainly that the b40 in Georgia own fewer assets, make less use of them by engaging in productive activities, and that returns to their endowments are low resulting in dependence on social transfers and subsistence agriculture. These findings are consistent with those reported in Bussolo and Lopez- Calva (2014). High dependency ratios push down returns to assets, while reliance on small holding subsistence agriculture limits asset use. Low wages, skills gaps, and weak access to credit limit income growth opportunities. Reliance on social transfers is very high compared to the rest of the region, while the share of wages in income is relatively low, reflecting lower labor force participation rates among the bottom 40 and reliance on informal, seasonal and self-employment—particularly in the agricultural sector. The policy levers available to strengthen the stake of the b40 in the Georgian economy are closely linked to the interventions needed to strengthen productivity, facilitate resource reallocation towards more competitive sectors, and thereby support export and job growth. This report contributes to our understanding of some of these policy channels (bolded) summarized in Table 1.2, which highlights the close links between the interventions needed to promote export competitiveness, job growth and shared prosperity, particularly in Georgia where poverty is largely driven by high unemployment. While country case studies suggest some common patterns in terms of poverty and inequality across trade liberalization episodes, evidence is mixed. However there is broad consensus that the benefits of trade liberalization for growth and poverty ultimately depend on productivity increases in some sectors and the corresponding capacity of the economy to reallocate Chapter 1: Growth and Shared Prosperity in Georgia: Why Trade?   15 | Draft 5.indd 33 22/12/14 15:19:07 Georgia Country Economic Memorandum | December 2014 Figure 1.21. Share of Income from Social Transfers Figure 1.22. Share of Income from Social Transfers 2010 2010 In percent 45 TJK 11 KSV 13 40 KAZ 12 AZE 11 35 ARM 12 GEO 13 ALB 13 30 BIH 09 MKD 11 25 LVA 12 KGZ 11 20 BLR 12 MDA 10 BGR 13 15 EST 11 HRV 12 10 LTU 09 MNE 13 5 SRB 10 ROU 10 0 UKR 11 ALB ARM HRV KAZ KSV MDA MKD SRB TJK TUR UKR GEO 0 5 10 15 20 25 In percent JJBottom 40% QQTop 60% JJSocial assistance JJLabor market JJSocial insurance Source: Bussolo and Lopez-Calva (2014) and ECA SPeeD as reported in the Public Expenditure Source: Bussolo and Lopez-Calva (2014) and ECA SPeeD as reported in the Public Expenditure Review, World Bank (2014). Review, World Bank (2014). Note: Year for which data is shown is indicated next to country label. resources—both labor and capital—to more productive sectors. This is also consistent with the findings for Georgia of a recent trade sustainability impact analysis9 commissioned by the EU. This report focuses on jobs and trade, by taking a closer look at the key drivers of export competitiveness in Georgia, which is critical to scaling up production and employment. Chapter 2 takes a closer look at export dynamics and firm survival. Chapter 3 explores the links between trade and labor in Georgia and discusses skills and skills development as a means to relieving at least partially the constraints on labor mobility. This is especially important if trade and trade policy (including closer ties with the EU) are to impact job growth as desired since resources will have to move from less to more productive sectors. Chapter 4 digs into the pending agenda on trade and transit facilitation. Chapter 5 discusses the constraints on firm competitiveness in the context of regional disparities and contributes to our understanding of the horizontal constraints on competitiveness at the sub- national level. Chapter 6 assesses the role of financial access in promoting MSME growth and capacity to generate much needed employment. Chapter 7 updates our understanding of Georgia’s external imbalances in the light of trade patterns and its drivers, going deeper into determinants of the current account deficit and its sustainability. 9 See Trade Sustainability Impact Assessment in Support of Negotiations of a DCFTA between the EU and Georgia and the Republic of Moldova, October 2012. 16 | Chapter 1: Growth and Shared Prosperity in Georgia: Why Trade? Draft 5.indd 34 22/12/14 15:19:07 Georgia: Seizing the Opportunity to Prosper Table 1.2. Proposed Policy Framework Policy Area Asset Accumulation Intensity of Use Prices Transfers Sustainability Macroeconomic Encourage capital Create an Lower costs Strengthen Savings fundamentals and physical asset economic of borrowing, targeting to the environment and fiscal accumulation, structure that especially chronic poor and Financial Stability systems especially beneficial generates more beneficial to distinction from Macroeconomic to small borrowers/ employment small borrowers/ transitory poor. sustainability investors though opportunities Investors. stable macro and incentives Address effectively environment. to formalize issues of food and employment. energy inflation. Ensure no Address the disincentives to impact of fiscal work generated systems on by social prices. transfers/taxes for low wage earners. Public spending, especially investment spending, needs to be in line with development priorities, including the concern for inclusiveness of growth. Service Delivery Emphasis on quality Back to work Focus on skills for Transfers Monitoring and of and access to all incentives to productivity and transparently, evaluation to levels of education, support LFPRs, real wage growth, efficiently and support quality ensuring equality in particular of also with a view effectively of public services, of opportunities women and IDPs. to decrease the managed. Policies especially to the to gain quality Strengthening gender wage gap. need to ensure b40. education for b40 connectivity Returns affected that the b40 has Public and t60. in terms of by quality of the necessary expenditures Strengthening development public goods. information and should be quality and of markets and easiness to access efficient and not access of health logistics, ensuring social assistance threaten fiscal services for the that the necessary and other sustainability. entire population, infrastructure and services. including the b40. markets access Special training, life- reach rural areas long learning, and and other less skills development, well-off regions. quality and access (b40). | Chapter 1: Growth and Shared Prosperity in Georgia: Why Trade?   17 Draft 5.indd 35 22/12/14 15:19:07 Georgia Country Economic Memorandum | December 2014 Table 1.2. Proposed Policy Framework Policy Area Asset Accumulation Intensity of Use Prices Transfers Sustainability Well-functioning Support credit access Removing Constraints Public subsidies State capture of markets/ to new and small constraints on imposed by to firms that regulation business and mid-sized firms. firm productivity markets or by distort returns Business environment Are markets growth that the business and competitive environment competitive for limit economic environment conditions should generating MSMEs? participation of on productivity be carefully inequality traps the b40. growth need to be evaluated. and threatening Create a favorable reflected in real social cohesion business wage levels. environment in areas where b40 are overrepresented, so as to foster inclusion and economic opportunities, including in entrepreneurship, for the b40. Allocative efficiency. Risk Provision for Inability to Prices reflect Mechanisms Longer run management cushioning assets manage risks risks generating systemic risks. from shocks, generating lower adverse especially b40. asset use. incentives/Moral hazard. Source: Adapted from Bussolo and Lopez-Calva (2014). Note: In italics if this report discusses proposed focus area. 18 | Chapter 1: Growth and Shared Prosperity in Georgia: Why Trade? Draft 5.indd 36 22/12/14 15:19:07 Georgia: Seizing the Opportunity to Prosper SUSTAINING EXPORT GROWTH IN CHAPTER 2:  GEORGIA As a small and open economy, Georgia’s growth prospects are inevitably linked to its ability to compete, and to grow and sustain exports. Georgia’s export growth rates have been impressive, at well above 10 percent per year on average since 2000. Over 2006–12, Georgia’s exports tripled in nominal value and doubled in volume. A closer look at the sources of export growth reveal that firms mainly relied on growing exports of the same product or exports to the same destination. As well, survival has been a challenge. The average length of an active export product is 2.15 years, compared to 2.84 in Lithuania, 3.35 in Slovakia, or 3.5 in Czech Republic, with the same period considered across all countries. Understanding the main challenges to export survival is crucial from a policy perspective if exports are to grow and if this growth is to be sustained. Sustainable export growth is typically associated with an expansion into new products and new markets (the extensive margin), the extension of existing export relationships (the intensive margin) and the survival of these relationships across time (the sustainability margin). Exporting is a risky activity characterized by a high degree of uncertainty resulting in flows of short duration.10 Low survival rates can entail welfare losses for the economy as a whole when sunk costs of entry and exit are high. An important difference between successful and less successful exporters is the former’s ability to maintain export relationships for longer periods, which allows deeper trade, export growth, and job creation. A detailed firm level dataset is matched with export transaction data for the period 2006–12 to explore export dynamics in depth. This unique dataset allows us to assess export dynamics in detail. Productivity and product diversification, foreign partnerships and networks matter for export survival. Firms that are more diversified at the product level show better chances of survival relative to those that have a concentrated export bundle. However, firms that are more diversified at the destination level show lower survival rates than those with export bundles concentrated in fewer destinations. Third, it is production efficiency, rather than size, that boosts export survival chances. The rest of this chapter is organized as follows. Section A analyzes the characteristics of exporting Georgian firms in terms of size, ownership, productivity and technology, and diversification patterns. Section B reviews trends in the survival of Georgian export flows and its determinants. Finally section C derives some policy implications and concludes. 10 See, for example, Besedes and Prusa (2004, 2006), Brenton et al (2010), Cadot et al (2013). | Chapter 2: Sustaining Export Growth in Georgia   19 Draft 5.indd 37 22/12/14 15:19:07 Georgia Country Economic Memorandum | December 2014 A. Export Dynamics in Georgia11 The share of exporting firms is low relative to Figure 2.1. Share of Exporting Firms in Total other countries in the region and movements in EST and out of export markets are quite large. The share LTU LVA of exporting firms ranges from 10 to 16 percent of all POL firms in the sample, lower than the top performers in GEO ALB the ECA region (Figure 2.1). However movements in ROU and out of export markets are large relative to other KAZ countries. Table 2.1 shows the dynamics of entry and BGR MEX exit into the export market by tracking firms that PAN exported the year before and ceased to export the PER ARM following year and vice-versa (firms that exported but AZE were not exporting the year before). In-sample exit 0 5 10 15 20 25 30 35 40 rates out of exports were between 15 and 20 percent In percent Source: World Bank staff calculations. while entry rates into exports were slightly lower. These figures are likely to be underestimated because larger firms are over-represented in the matched dataset. Computing entry and exit rates for the entire universe of Georgian exporters from the full custom data gives rates close to 50 percent, high by international standards. Table 2.1. Exit from and Entry into Exporting in Georgia, 2006–12 Year Firms Exporting Exit from New Total Exported Total Exporting Firms (%) Exporting Exporters Products (HS6) Destinations 2006 2,117 234 (11.1) 45 265 50 2007 1,581 166 (10.5) 29 30 286 55 2008 1,606 168 (10.5) 25 29 317 51 2009 2,415 237 (9.8) 46 28 397 70 2010 2,220 211 (9.5) 34 37 479 69 2011 1,848 292 (15.8) 46 36 665 76 2012 2,029 255 (12.6) … 42 588 71 Source: Authors’ Calculations based on Geostat—Industrial Survey. Exporters are larger both in terms of turnover and employment. In 2006, exporters’ turnover and employment were four times bigger than those of non-exporters. The difference increases over the years with exporters having on average five times the turnover and employment of non-exporting firms in 2012, reflecting higher growth. Exporters are more capital intensive, more productive, and show a higher share of foreign ownership than firms oriented to the domestic market. TFP, an indicator of firm efficiency that measures the amount of 11 The analysis is based on export transaction data merged with firm level data obtained from Geostat, Georgia’s National Statistics Office. The firm level data consists of a panel of Georgian firms spanning the period 2006–12. This is matched with Customs data recording all export transactions occurred in the same period using a common firm identifier. Large firms (those with more than 100 employees) are all included in the dataset while small and medium enterprises have been randomly sampled. There are 13,816 firm-year observations with 6,745 firms surveyed at least once in the panel (Table 2.1). On average firms have been surveyed around twice from 2006–12. 20 | Chapter 2: Sustaining Export Growth in Georgia Draft 5.indd 38 22/12/14 15:19:07 Georgia: Seizing the Opportunity to Prosper Figure 2.2. Employment for Exporters and Non- Figure 2.3. Turnover for Exporters and Non- Exporters (2006–12) Exporters (2006–12) In thousands Turnover, GELs, in millions 140 12 120 10 100 8 80 6 60 4 40 20 2 0 0 2006 2007 2008 2009 2010 2011 2012 2006 2007 2008 2009 2010 2011 2012 JJDomestic JJExporting firms JJDomestic JJExporting firms Source: Calculations based on GeoStat firm level data. Source: Calculations based on GeoStat firm level data. Figure 2.4. Key Characteristics of Exporting Firms Total Factor Productivity by Exporter Status Productivity Growth In logs In logs 3.8 3.8 3.7 3.7 3.6 3.6 3.5 3.5 3.4 3.4 3.3 3.3 3.2 3.2 3.1 3.1 3.0 3.0 2.9 2.9 2.8 2.8 2006 2007 2008 2009 2010 2011 2012 2006 2007 2008 2009 2010 2011 2012 JJNon-exporting firms JJExporting firms ▬▬Non-exporters ▬▬Exporters ▬▬Total Foreign Ownership Labor Intensity of Production In percent Ratio 35 0.0025 30 0.0020 25 0.0015 20 15 0.0010 10 0.0005 5 0 0 2006 2007 2008 2009 2010 2011 2012 2006 2007 2008 2009 2010 2011 2012 JJDomestic JJExporting firms JJNon-exporting firms JJExporting firms Source: Calculations based on GeoStat firm level data. Chapter 2: Sustaining Export Growth in Georgia   21| Draft 5.indd 39 22/12/14 15:19:07 Georgia Country Economic Memorandum | December 2014 output a firm can produce with a given amount of inputs, is on average, 24 percent greater for exporters than for non-exporters (see Box 2.1 for a spotlight on productivity dynamics in Georgian exporters). While all firms exhibit growing capital intensity in Georgia, this trend is more pronounced for exporters. The share of foreign ownership among exporters is substantially higher than among non-exporters, and has grown for both groups over time, increasing from 1.8 to 6.7 percent for non-exporters over 2006–12 and from 14 to 32 percent for exporters over the same period. Multi-Product firms have increased their share in exporting firms and outperform single-product firms. Multi-Product firms represent a sizable share of exporting firms and this share has increased steadily over the years going from 42 percent to 58 percent between 2006 and 2012 (Table 2.2). However when we only consider significant products (more than 1 percent of total revenues), the average number of products exported per firm falls somewhat and stays more or less stable, though the share of multi-product exporters as a proportion of exporting firms still shows impressive growth. Multi-Product firms have higher foreign participation and are also more diversified in terms of export destinations. These firms also create more jobs and are 6.5 percent more productive than single-product firms. Table 2.2. Exporters and Multi-Product Firms in Georgia Year Firms Exporting firms Multiple product Multiple product exporting firms exporting firms (% of exporting firms) excluding products less 1% of export revenues (% of exporting firms) 2006 2,117 234 98 (41.9) 87 (37.2) 2007 1,581 166 88 (53.0) 75 (45.2) 2008 1,606 168 88 (52.4) 76 (45.2) 2009 2,415 237 122 (51.5) 104 (43.9) 2010 2,220 211 124 (58.8) 108 (51.2) 2011 1,848 292 160 (54.8) 140 (47.9) 2012 2,029 255 147 (57.6) 128 (50.2) Source: Calculations based on GeoStat firm level data. Table 2.3. Exporters and Multi-Destination Firms in Georgia Year Firms Exporting firms Multiple destinations exporting firms (% of exporting firms) 2006 2,117 234 104 (44.4) 2007 1,581 166 87 (52.4) 2008 1,606 168 88 (52.4) 2009 2,415 237 114 (48.1) 2010 2,220 211 112 (53.1) 2011 1,848 292 167 (57.2) 2012 2,029 255 138 (54.1) Source: Calculations based on GeoStat firm level data. Multi-Destination firms are also larger, more diversified, and have higher export revenues. While there are no major differences with respect to the labor intensity of technology, TFP is 23 percent greater in multi- destination firms than in firms exporting to a single destination over the 2006–12 period. The size and the 22 | Chapter 2: Sustaining Export Growth in Georgia Draft 5.indd 40 22/12/14 15:19:07 Georgia: Seizing the Opportunity to Prosper Figure 2.5. Characteristics of Single and Multi-Product Exporters Export Revenues Employment US$ thousands In thousands 3,000 180 160 2,500 140 2,000 120 100 1,500 80 1,000 60 40 500 20 0 0 2006 2007 2008 2009 2010 2011 2012 2006 2007 2008 2009 2010 2011 2012 JJSingle-Product exporting firms JJMulti-Product exporting firms JJNon-exporting firms JJSingle-Product exporting firms JJMulti-Product exporting firms Foreign Ownership TFP In percent In logs 40 4.0 35 3.5 30 3.0 25 2.5 20 2.0 15 1.5 10 1.0 5 0.5 0 0 2006 2007 2008 2009 2010 2011 2012 2006 2007 2008 2009 2010 2011 2012 JJNon-exporting firms JJSingle-Product exporting firms JJMulti-Product exporting firms JJNon-exporting firms JJSingle-Product exporting firms JJMulti-Product exporting firms Source: Calculations based on Geostat firm-level data. employment premium of multi-destination firms appear to be slightly lower than the one observed for Multi- Product firms. Moreover, while Multi-Product firms’ employment premium increased between 2006 and 2012, it declined for multi-destination firms over the same period. In fact, both employment and turnover were more than three times those of single destination firms in 2006 but less than 50 percent higher in 2012. Export growth is largely accounted for by more of the same products to the same destinations. An export growth decomposition into the intensive and extensive margins shows that about 80 percent of the export growth is due to the intensive margin and 20 percent is due to the extensive margin i.e. due to the addition of new products to the export mix. Product diversification in Georgia is largely because of new firms. We decompose the extensive margin into a “within-firm” component, which represents the contribution of products exported by continuing firms (firms exporting both in 2006 and 2012) and an “entry/exit” component, which represents the net effect of new products exported by entrant firms minus the products dropped by exiting exporters. Between 2006 and 2012, 75 percent Chapter 2: Sustaining Export Growth in Georgia   23 | Draft 5.indd 41 22/12/14 15:19:08 Georgia Country Economic Memorandum | December 2014 Figure 2.6. Characteristics of Single and Multi-Destination Firms Employment TFP In thousands In logs 180 4.5 160 4.0 140 3.5 120 3.0 100 2.5 80 2.0 60 1.5 40 1.0 20 0.5 0 0 2006 2007 2008 2009 2010 2011 2012 2006 2007 2008 2009 2010 2011 2012 JJNon-exporting firms JJSingle-Destination exporting firms JJMulti-Destination exporting firms JJNon-exporting firms JJSingle-Destination exporting firms JJMulti-Destination exporting firms Foreign Ownership Export Revenues In percent US$ thousands 35 3,500 30 3,000 25 2,500 20 2,000 15 1,500 10 1,000 5 500 0 0 2006 2007 2008 2009 2010 2011 2012 2006 2007 2008 2009 2010 2011 2012 JJNon-exporting firms JJSingle-Destination exporting firms JJMulti-Destination exporting firms JJSingle-Destination exporting firms JJMulti-Destination exporting firms Source: Calculations based on Geostat firm-level data. of the export growth due to product diversification was within new firms. 30 percent of the growth in the number of product exported is due to within firm diversification while 70 percent is due to the entry of new firms. There is substantial small scale experimentation with new products, at low survival rates. Attempts to add new products are typically at a small scale and “experimental.” These products, however, struggle to contribute substantially to export growth probably because of low survival rates which impedes their consolidation. The number of export products increased from 1497 in 2006 to 2024 in 2012. In 2012 the product mix is equally split between continuing products (products already exported in 2006) and new products. However, new products only account for 16 percent of total exports in terms of value. From the perspective of destinations, export growth is obtained almost entirely from increased exports to the same destinations. The extensive margin accounts for a mere 2.8 percent of total export growth in the period. This implies that at the firm level destination diversification has been minimal and much lower than product diversification. Contrary to what it is seen for product diversification, the major part of the destination 24 | Chapter 2: Sustaining Export Growth in Georgia Draft 5.indd 42 22/12/14 15:19:08 Georgia: Seizing the Opportunity to Prosper Box 2.1. Productivity Dynamics Firm productivity is defined here as the amount of output produced for a given level of inputs used. Productive efficiency of firms, that is the amount of output produced with a given amount of inputs, measured by the (log) total factor productivity, has been increasing since 2006 in Georgia, at a rate of 8.2 percent per annum. While TFP was relatively stagnant during 2006–09, it increased almost exponentially since 2010. Interestingly, exporting firms’ TFP is higher than non-exporting firms’ TFP, but the rate of growth of TFP is lower for the former than for the latter group (6.5 percent per annum versus 8.7 percent per annum, respectively). Figure 2.7. Productivity in Georgia Has Table 2.4. Production Functions Estimates Increased Systematically Since 2006 Average log TFP 2006–12 3.8 Variables (LP) lny (OLS) lny (FE) Lny 3.7 Log employment 0.858*** 0.853*** 0.791*** 3.6 (0.0356) (0.0191) (0.0247) 3.5 Log capital stock 0.191*** 0.113*** 0.0512*** 3.4 (0.0730) (0.00969) (0.0125) 3.3 Log energy 0.165 0.258*** 0.192*** 3.2 (0.116) (0.0105) (0.0116) 3.1 Constant 3.455*** 4.353*** 3.0 (0.0431) (0.0994) 2.9 Observations 6,209 6,209 6,209 2.8 R-squared 0.724 0.359 2006 2007 2008 2009 2010 2011 2012 Waldcrs 3.849 ▬▬Non-exporters ▬▬Exporters ▬▬Total Source: World Bank staff calculations. Source: World Bank staff calculations. Note: Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 Figure 2.8. Internationalized Firms are More Figure 2.9. Medium to Large Firms are More Productive in Georgia too, but their TFP Productive Premium has Fallen Over Time Exporters and foreign owned log TFP premium over non-exporters and domestic owned Log TFP 2006–12 by quartiles of employment firms 2006–12 0.5 4.5 4.0 0.4 3.5 0.3 3.0 2.5 0.2 2.0 0.1 1.5 1.0 0 0.5 -0.1 0 2006 2007 2008 2009 2010 2011 2012 2006 2007 2008 2009 2010 2011 2012 JJExporter TFP premium JJForeign ownership premium JJSmall JJMedium JJMedium–large JJLarge Source: World Bank staff calculations. Source: World Bank staff calculations. Chapter 2: Sustaining Export Growth in Georgia   25 | Draft 5.indd 43 22/12/14 15:19:08 Georgia Country Economic Memorandum | December 2014 Internationalized firms and larger firms are more productive. As mentioned above, exporting firms are more productive than non-exporters, displaying a productivity premium of 17 percent, on average, above non-exporters over the sample period. However, this premium has been falling due to TFP growing faster for domestic oriented firms than for exporting firms. Similarly, foreign owned firms display a 15 percent productivity premium, on average, above domestically owned firms. Larger firms are more productive than smaller firms. Firms in the first quartiles of the employment distribution, which are very small (one or two employees), have the lowest productivity, and productivity tends to increase with size although not monotonically. For example, medium-large firms are 34 percent more productive than small firms and 7 percent more productive than medium-small firms, while large firms are 8 percent less productive than medium-large firms. Source: World Bank Staff Calculations. diversification happens within continuing firms (62.2 percent) while new destinations introduced by entrants account for 37.8 percent of the extensive margin. B. Export Firm Survival in Georgia The survival rate of Georgian export is low compared to other countries in the region. A firm-product- destination export spell is defined as a period in which a firm continuously exports a particular product (HS 6 digits) to a given destination. We use a three-year spell definition meaning that if a product is not exported for three consecutive years the spell is completed. If the firm starts exporting that product again following the three year period it is considered a new spell. The mean spell length over which an export flow is active for Georgia is on average slightly more than two years, while for Lithuania it is 2.84 years, for Slovakia 3.35 years, and for Czech Republic 3.5 years. A high percentage of spells are of short duration, though longer spells are associated with higher exports. Around 70 percent of product spells last one year, while only about 18 percent exceed two years. A higher duration of export relationships is associated with higher exports, that is, when the export relationship survives beyond the first two years, it grows in value over time, suggesting that as exporters and clients face imperfect information, they experiment with small export transactions, which grow in value as additional information is obtained by both parties. With imperfect information firms face a high degree of uncertainty—and sunk costs—when entering export markets, which vary by sector. Firms need to obtain information about export costs, the demand profile of potential markets, the requirements they need to meet to enter these markets, etc. They also need to judge in advance, their own ability to survive in the export markets given market conditions. There is need for significant research, and the willingness and ability to invest in establishing presence in a new market. There is substantial sectoral heterogeneity in survival probabilities. The probability of survival is higher for traditional firms producing vegetables, milling, cereals, organic agents, textiles and footwear products. 26 | Chapter 2: Sustaining Export Growth in Georgia Draft 5.indd 44 22/12/14 15:19:08 Georgia: Seizing the Opportunity to Prosper Figure 2.10. Distribution of the Average Spell Figure 2.11.Average Export Value at Each Spell Length in Years of a “Product-Destination” Length Combination Product-destination spell length distribution Length 7 7 6 6 5 5 4 4 3 3 2 2 1 1 0 10 20 30 40 50 60 70 80 0 100 200 300 400 500 600 In percent US$ thousands JJAverage final value JJAverage initial value JJAverage value Source: World Bank staff calculations. Source: World Bank staff calculations. Survival in export markets is for the fittest, rather than for the largest. More productive firms perform better at keeping the export flow active, displaying longer trade relationships. The effect is not very strong though as doubling the average firm total factor productivity would increase the survival probability by 0.55 percentage points. Instead, larger firms (with size measured in term of number of employees) tend to have lower probabilities of survival in export markets after controlling for productivity. Foreign owned firms have a lower probability of dropping a product. The effect is substantial even after controlling for their higher productivity: products exported by foreign owned firms have an average survival probability of 2.3 percentage points higher than domestic owned firms. This result is in line with the theoretical expectations that see information and network barriers as important obstacles to export success. Foreign firms are likely to have more established links with foreign partners and better knowledge of foreign markets that can help in identifying successful export products. Product and destination diversification interact with export survival in a complex way. While product diversification increases the chances of survival, destination diversification decreases them. Firms with higher product diversification have higher survival probabilities and are less likely to drop an export product. Adding one product to the firms’ export mix from the average of 3.3 products per firm increases the probability of survival by 0.32 percentage points. The number of export destinations, instead, increases the hazard of ending the transaction. Firms reaching a larger number of markets tend to have a higher probability of dropping products form their export mix. The average number of destinations for firms in the sample is three. Adding one destination would increase the likelihood of dropping a product by 0.83 percentage points. It is possible that the information required to export to a particular market is high but not easily transferrable across destinations. This would imply that the challenges of surviving increase with the number of destinations (or that the chances of survival decrease with them). Instead, the information required to exporting specific products is transferrable across products. Therefore, given that a firm is exporting one product already, it faces less costs of exporting an additional variety, and then, survival chances increase with the firms’ product scope. | Chapter 2: Sustaining Export Growth in Georgia   27 Draft 5.indd 45 22/12/14 15:19:08 Georgia Country Economic Memorandum | December 2014 There is strong evidence of network and information spillover effects. Having many firms producing and exporting a given product in one market increases survival probability, even after controlling for sector comparative advantage. Doubling the number of firms exporting the same product from the average of 1.8 would increase the survival probability by 1.43 percentage points. Although the effect does not appear to be large, it is important, considering that around 30 percent of product spells survive the first year this would account for a proportional increase of 5 percent (from 30 percent to 31.5 percent)12. International evidence suggests two channels of influence for networking: improved information sharing and strengthened financial access as lenders risk perception is reduced existing firm presence in the market. Interestingly, survival is not higher in markets with which Georgia has a free trade agreement (FTA). This effect, however, is likely captured by the network effects. It may also imply the need for a stronger push to leverage the market access opportunities offered by the FTAs. Cultural affinities and distance also increase export survival, reflecting long standing trade relationships and demand for traditional products. Regional coefficients are statistically significant and are associated with a higher probability of survival in export markets. Products to traditional markets in ECA countries and to Europe are about 21.5 and 20 percentage points respectively more likely to survive than those that go to South Asia (the baseline). Firms with more capital intensive production technologies have higher survival probability for their export products. The capital to labor ratio tends to be higher in firms that survive. Looking at the sectoral patterns of labor intensity, we see that there is no sector which is labor intensive. Instead labor intensity seems to depend more on firms’ characteristics within sectors (domestic vs. exporting firms), with exporters tending to employ more labor but relatively more capital13. C. Conclusions and Policy Messages Georgian exporters are no exception in terms of being larger and more productive than domestic oriented firms. As is generally the case internationally, exporting firms in Georgia are far stronger performers than their domestic-oriented counterparts. They are more productive and therefore grow faster than non-exporters in terms of employment and output. They also represented about 9–16 percent of the total number of firms in the economy (excluding services), which is in line with other countries in the region, over the 2006–12 period. During this period, exports doubled in volume and tripled in nominal value. Export growth in Georgia has been largely due to more of the same products to the same destinations. Export diversification has been modest, and though there is noteworthy experimentation with new products, especially by new firms, this remains largely small scale. New products were 50 percent of the total number of 12 The firm-product couple instead of the firm-product-destination triplet gives similar results. The network variable is again statistically significant and positively related to the survival probability. The effect is much stronger in this case: doubling the number of firms exporting the same product (independently of the destination market) from the average of 3.9 would increase the survival probability by 2.6 percentage points. This would account for a proportional increase of 8.7 percent (from 30 percent to 32.6 percent) of the first year survival rate. 13 Positive terms of trade shocks are not significant. Increases in relative prices increase the profit margins of exporters and should increase the probability of survival. The product real exchange rate is however not statistically significant. Export survival is not affected by the product’s degree of differentiation. 28 | Chapter 2: Sustaining Export Growth in Georgia Draft 5.indd 46 22/12/14 15:19:09 Georgia: Seizing the Opportunity to Prosper exported products in 2012 but accounted for only 16 percent of total exports in terms of value. 75 percent of the new products introduced were by new firms. With high costs of learning about foreign demand conditions, tastes, and product specification requirements, many firms engage directly in small international transactions without having a thorough understanding of the market. The fact that firms with higher initial values survive longer in export markets seems to support this interpretation. Firms that start exporting at a greater scale are more likely to have undertaken prior thorough market research, and to be more confident about the survival of its market relationship with foreign buyers. While export survival is crucial for deepening trade links and growing exports, Georgian export spells tend to be of short duration. 75 percent of new export products and 70 percent of all exports products fail to survive past their first year. If they do survive, the probability of being dropped reduces significantly. Understanding what factors influence the duration of export flows in global marketplace is important for policy purposes, given the important potential role of exports in growing jobs and boosting shared prosperity. The main drivers are: a. Productivity: Survival is for the fittest, not for the largest. While the impact is small, it dominates the size effect, so that large firms appear to be negatively associated with survival once productivity levels are accounted for. Since large firms are more likely to perform well, the partial impact of size should not be over-interpreted but only seen as evidence of the primary importance of productivity growth in sustaining exports. b. Firms with foreign participation are more likely to survive. This may reflect better management, closer- to-the-frontier technologies, greater knowledge of international markets, and more capacity for and access to finance and market research. c. Network effects are important. The chances of surviving active in export markets increase with the number of firms exporting the same product and to the same destination. The effect is unchanged when controlling for sectoral comparative advantage and sector fixed effects. There are two channels through which network or spillover effects may improve export survival. Both are related to informational spillovers. First, the more firms exporting one product, or to one destination, the more information available there is about the specificities of exporting that product or exporting to that destination. That information may spillover to new entrants or to diversifying firms, which may benefit from it, since it reduces the risk of entering in a completely new market. Second, increased information also spills over to, for example, the financial sector, which may now find it less risky to finance innovative activities of firms trying to diversify. d. Diversification matters. Product diversification is positively correlated with survival with market diversification is associated with a higher risk of dropping products. This finding may reflect the strong correlation between export survival and deeper trading relationships, networking, and traditional ties, and is likely to change with more investment in export promotion and financing, and market research. | Chapter 2: Sustaining Export Growth in Georgia   29 Draft 5.indd 47 22/12/14 15:19:09 Georgia Country Economic Memorandum | December 2014 Table 2.5. Main Policy Messages Findings Policy Suggestions Strengthening export competitiveness is critical for job Improving access to finance will support export growth diversification both in terms of products and destinations given that these are costly and risky activities for firms. Exporters are better performers, including on job growth, This will also support R&D, adoption and adaptation of and are more productive than domestic-oriented firms. new products and technologies and encourage smaller and newer firms to innovate, scale up, survive and grow jobs. Product diversification is associated with better productivity, growth, job creation, and size but is costly Facilitating university–enterprise linkages to build a stronger and risky national system of innovation to support the dissemination of basic knowledge from academia to the productive Survival depends on productivity growth sectors of the economy, and from firms to universities will Ability to benefit from networks and foreign partners, and help reduce the costs of diversification and will encourage ability to product diversify; foreign participation is also innovation for productivity growth beneficial for productivity growth and therefore affects Strengthening export promotion activities to reduce survival through multiple channels entry costs associated with informational asymmetries Larger initial transactions are associated with larger, more Informational spillovers contribute to improved export productive firms, and with higher survival probabilities survival and can be supported by strengthening the provision of information about foreign consumers’ preferences, help with the identification of potential buyers, and assist in tackling the regulatory complexities associated with serving foreign markets. Any reform process in this area would benefit from a transparent system of monitoring and evaluation, based on improved firm level and transactions data, to ensure that scarce public funds are at their best use. Encouraging foreign direct investment and facilitating interaction between foreign and domestic firms Inflows of foreign investment have been targeting services sectors such as construction and banking, and real estate. Investment promotion activities could be focused on reorienting potential investors to tradable sectors and more innovative activities 30 | Chapter 2: Sustaining Export Growth in Georgia Draft 5.indd 48 22/12/14 15:19:09 Georgia: Seizing the Opportunity to Prosper CHAPTER 3: TRADE, LABOR OUTCOMES, AND SKILLS Export growth has been high in Georgia, albeit mainly driven by commodities of unprocessed goods, which also dominate the labor market. The labor market in Georgia is dominated by the low skilled primary sector. The average growth rate of goods exports between 2000 and 2013 was 20 percent. This growth was driven primarily by chemicals and base metals, but other export sectors also exhibited strong export growth. Foodstuffs and vegetables products, two export sectors that are important for Georgia, have been less dynamic in recent years but nevertheless more than doubled in value since 2000. Both sectors contributed more to export growth in the earlier part of the last decade, but also witnessed strong growth in 2009. While Georgia’s export sophistication in terms of labor content has changed only modestly over the past decade, exporting firms outperform domestic market-oriented firms in terms of job creation. Export sophistication in terms of labor content—in particular the skills content of exports—has declined for non-mineral and base metal exports to the EU, which are increasingly dominated by primary goods including vegetable products that embody lower skills and lower per worker value added and capital. However, the skills content of exports to the ECA region has increased, reflecting mainly wine and processed food. Georgia has not kept pace with the region’s more dynamic economies, namely Poland, Estonia, and Latvia in terms of export sophistication. However it outperforms the rest of ECA in terms of employment generation of exporting firms. Strengthening the link between labor market outcomes and trade will depend on how smoothly workers can reallocate to higher productivity sectors. Exporters hire more workers, including more women than domestic firms even though they employ a lower share of female workers depending on sector and destination. While higher exports to the EU favor female employment, trade with ECA favors male workers. Strengthening the impact of trade on labor market outcomes depends largely on the ability of workers to redeploy towards more productive sectors. In Georgia, it is relatively easy for workers to find jobs in agriculture, and more difficult to move into manufacturing or public sector employment. Employer surveys and simulations of the labor market suggest that shortages of skills limit labor mobility into non-agricultural sectors. The skills shortage is consistent with a simulated lower bound unemployment rate of 15 percent under specific assumptions. Innovative and foreign-owned firms in Georgia report a significant skills constraint. In fact the bulk of employment in Georgia remains in traditional sectors, where the skills constraint is viewed as less of an obstacle. Even were jobs to grow sufficiently to absorb existing excess supply of labor at each skill level, there would be a skills mismatch because of insufficient availability of workers with the right skills, even broadly defined, i.e., without digging deeper into specific professional qualifications. This skills mismatch induced unemployment rate is simulated to be as high as 15 percent. | Chapter 3: Trade, Labor Outcomes, and Skills   31 Draft 5.indd 49 22/12/14 15:19:09 Georgia Country Economic Memorandum | December 2014 The rest of this chapter is organized as follows. Section A discusses employment and trade patterns and Section B looks at export sophistication in terms of labor content. Section C examines labor mobility in Georgia and Sections D and E assess the skills context in the country and review international experience. Section F concludes. A. Employment and Trade Patterns The labor market in Georgia is dominated by the low wage, low skill primary sector. Based on household survey data, 57 percent of the workforce is in the primary sector, mainly in agriculture. Two-thirds of these workers are unskilled. The average wage earned in agriculture was only one third of that in manufacturing, which employs only 4.4 percent of the workforce. Out of those employed in manufacturing, 62 percent are skilled and 63 percent are male. The highest wages are in mining and quarrying—89 percent male—and construction—97 percent male—and services, which are also sectors that have recorded high capital investments and productivity growth. Skilled manufacturing workers earn 20 percent more that their unskilled counterparts.14 Interestingly, the skill wage gap in agriculture is negligible, supporting the idea that agriculture is an employer of last resort. Gender gaps are large in the labor market. The female labor force participation rate (LFPR) has remained relatively stable at around 59 percent. This is comparable to the trends observed in Poland. Estonia and Latvia have shown far more dynamism in attracting women into the labor force, with the female LFPR reaching nearly 72 percent in both countries in 2011, compared to 77 percent for their male counterparts. The best employment opportunities for females tend to be concentrated in public sector service jobs, which accounted for a third of skilled women. Females account for 57 percent of public sector jobs, mainly due to the preponderance of women in low paid teaching jobs, which accounted for 19 percent of total female employment. Agriculture accounted for 60 percent of female employment. In the manufacturing sector, men earn 92 percent more than women on average (Table 3.1). Table 3.1. Labor Market Characteristics In percent unless otherwise indicated Total Male Female Unskilled Skilled Employment Share Agriculture, hunting, forestry, fishing 55.8 49.6 50.4 64.7 35.3 Mining, quarrying 1.1 88.9 11.1 45.3 54.7 Manufacturing 4.4 63.1 36.9 37.6 62.4 Construction 3.3 96.8 3.2 48.1 51.9 Public services jobs 15.6 43.0 57.0 17.2 82.8 Private services jobs 19.8 59.1 40.9 34.7 65.3 Median Monthly Wage (current Georgian lari) Agriculture, hunting, forestry, fishing 100.0 111.6 72.0 100.0 100.0 Mining, quarrying 500.0 500.0 325.0 600.0 330.0 Manufacturing 300.0 383.3 200.0 250.0 301.6 14 The sectoral wage gaps in the GeoStat wage data are significantly higher. 32 | Chapter 3: Trade, Labor Outcomes, and Skills Draft 5.indd 50 22/12/14 15:19:09 Georgia: Seizing the Opportunity to Prosper Table 3.1. Labor Market Characteristics In percent unless otherwise indicated Total Male Female Unskilled Skilled Construction 333.3 333.3 - 283.3 400.0 Public services jobs 320.0 595.0 265.0 260.0 330.0 Private services jobs 300.0 378.3 238.3 230.0 333.2 Employment Share of Skilled Workers Agriculture, hunting, forestry, fishing 35.3 35.0 35.7 n.a. n.a. Mining, quarrying 54.7 52.3 73.7 n.a. n.a. Manufacturing 62.4 62.2 62.8 n.a. n.a. Construction 51.9 50.8 86.7 n.a. n.a. Public services jobs 82.8 73.3 90.5 n.a. n.a. Private services jobs 65.3 61.9 70.6 n.a. n.a. Source: Georgia Integrated Household Survey 2011. Notes: Skilled is defined as college, technical school, or professional degree, a Bachelor s degree, or higher. Wage data may vary from that published by GeoStat, in particular for the agriculture sector. The Integrated Household Survey only published household-level income data, so industry-level wages are calculated using household that work in the same industry. Households with zero reported income are dropped. Georgia’s labor force has undergone a fundamental shift in terms of education attainment composition since 2002, with a much larger share of the population having secondary level education. Whereas in 2002 nearly half of the labor force had only a primary education, this share fell to under 10 percent by 2007, while those with a secondary education increased to 60 percent. But those with a secondary education also account for the largest share of those unemployed, namely 56 percent (compared to only 5 percent for those with primary level education) in 2011. Export growth in Georgia has been driven by the primary sector, especially chemicals and base metals. Georgia’s overall export growth has been strong, increasing steadily over the last decade at an average growth rate of over 20 percent between 2000 and 2013. Overall, Georgian goods exports peaked in 2013 at US$2.9 billion, up from US$324 million in 2000. This growth was driven primarily by base metals and chemicals, but other export sectors also exhibited strong export growth (Figure 3.1 right panel). Foodstuffs and vegetables products, two Figure 3.1. Georgia’s Export Growth Total exports, export value in US$ millions By sector, export growth in percent 1,800 50 50 1,600 40 40 1,400 30 30 1,200 20 20 1,000 10 10 800 0 0 600 -10 -10 400 -20 -20 200 -30 -30 0 -40 -40 2000 01 02 03 04 05 06 07 08 09 10 11 2012 2001 02 03 04 05 06 07 08 09 10 11 2012 JJExport growth ▬▬Export value JJBase metals JJChemicals JJMineral products JJPrecious metals JJFoodstuffs JJVegetable products JJOthers Source: World Bank staff calculations. Chapter 3: Trade, Labor Outcomes, and Skills   33 | Draft 5.indd 51 22/12/14 15:19:09 Georgia Country Economic Memorandum | December 2014 export sectors that are important for Georgia, have been less dynamic in recent years but nevertheless more than doubled in value since 2000. Both sectors contributed more to export growth in the earlier part of the last decade, but also witnessed strong growth in 2009. Exporting firms tend to be larger on average compared to the rest of the ECA region, and hire more skilled workers and women in absolute terms. Georgian exporting firms tend to be larger than non-exporting firms by more than a factor of three, compared to two for ECA15. This difference may reflect a higher fixed cost of exporting in Georgia than other ECA countries, which only larger, more productive firms can overcome. Or it may reflect a larger productivity gap between the largest firms and SMEs in Georgia than in other ECA countries. Georgian exporters hire more female workers than non-exporters though the share of female employees is smaller in exporting firms, with 37 percent of the workforce being women in exporting firms compared to 50 percent in non-exporting firms. Exporting firms also hire more skilled workers than non-exporting firms though the share of skilled workers is smaller in exporting firms, with 40 percent being skilled in exporting firms compared to 56 percent in non-exporting firms. This trend contrasts that of ECA where the share of skilled workers is slightly higher in exporting firms than non-exporting firms, 52 percent compared to 49 percent. The share of exporting firms in Georgia is small and declining, though they export more intensively. In Georgia, 25 percent of the firms in the sample were exporters in 2002, falling to 18 percent in 2005, 13 percent in 2009, and less than 9 percent in 2013. Some industries are more integrated with global markets than others. At the industry level a higher share of Georgian food and beverage firms were exporting compared to the ECA average but the share is lower for manufacturing, construction and services, including both wholesale and transport. The share of exports in total sales of exporting firms is higher than in ECA in all significant exporting categories, including food and beverages, metals and minerals, and chemicals. The European Union and ECA have been Georgia’s most important export partners over the last decade. Figure 3.2 depicts the value (left panel) and share (right panel) of Georgian exports destined to the EU27 countries, ECA*16 and Russia since 2000, illustrating a fairly stable trend that EU27 and ECA* consume about 20 and 50 percent of Georgia’s total exports respectively. ECA’s share of Georgia’s exports fell slightly after the global financial crisis of 2008–09, while the European Union’s share changed little. Russia lost importance as a trading partner starting 2005, with the country’s share of Georgia’s exports falling from 20 percent in 2000 to 2 percent by 2010, before increasing again to about 6 percent mainly because of a recent spurt in wine, water and used car exports in 2013. The composition of Georgia’s exports to the EU, ECA and Russia differs significantly. We compare the sectoral composition of Georgia’s exports across its main trading partners. Whereas Georgia’s exports to ECA* and EU27 countries have experienced similar growth in value terms since 2000, the composition of exports differs significantly. Exports to the EU27 are more concentrated in mineral products, especially petroleum oils and copper 15 The latest available year from the World Bank Business Environment and Enterprise Performance Surveys was used (2009) for all comparisons with ECA. 16 Note that ECA* denotes the following non-EU non-Russia ECA countries: Albania, Armenia, Azerbaijan, Bosnia and Herzegovina, Belarus, Georgia, Croatia, Kazakhstan, Kyrgyzstan, Moldova, Macedonia, Montenegro, Serbia, Montenegro, Tajikistan, Turkmenistan, Turkey, Ukraine, and Uzbekistan. EU27 includes Austria, Belgium, Bulgaria, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden, and United Kingdom. 34 | Chapter 3: Trade, Labor Outcomes, and Skills Draft 5.indd 52 22/12/14 15:19:09 Georgia: Seizing the Opportunity to Prosper ores and concentrates, as well as base chemicals. Foodstuffs have increased significantly however in importance for exports to the EU27 market, from 6.5 million in 2000 to 42 million in 2012. Base metal and mineral products are also important export sectors to ECA, but have been overtaken by agro-based goods, most notably exports of alcoholic and non-alcoholic beverages, which are actually at the higher value end of Georgia’s export spectrum. Figure 3.2. Georgia’s Exports by Destination Export values, in US$ millions Export shares, in percent 900 60 800 50 700 600 40 500 30 400 300 20 200 10 100 0 0 2000 01 02 03 04 05 06 07 08 09 10 11 2012 2000 01 02 03 04 05 06 07 08 09 10 11 2012 ▬▬EU27 ▬▬ECA* ▬▬Russia ▬▬EU27 ▬▬ECA* ▬▬Russia Source: World Bank staff calculations. Figure 3.3. Georgia’s Exports by Destination and Composition In percent EU27, 2000 Russia, 2000 ECA, 2000 EU27, 2012 Russia, 2012 ECA, 2012 JJBase metals JJChemicals JJFoodstuffs JJMineral products JJPrecious metals JJVegetable products JJOthers Source: World Bank staff calculations. Chapter 3: Trade, Labor Outcomes, and Skills   35 | Draft 5.indd 53 22/12/14 15:19:09 Georgia Country Economic Memorandum | December 2014 B. The Labor Sophistication of Georgia’s Exports Empirical studies show that the structure of a country’s exports—that is, what types of products it exports, in which sectors, with what levels of value-added, and to whom—is linked to the types of labor needed to produce these exports. Understanding the evolution of Georgia’s export basket in terms of product mix, product sophistication and destination markets therefore helps understand better potential policy channels that could support improved labor market outcomes, which in turn supports improved export outcomes. We therefore examine trends in export sophistication, by export destination, to infer potential implications for labor market outcomes. We construct and benchmark export sophistication indicators, in terms of labor content, to better understand the potential impact of export growth on labor market outcomes. We construct four indices or “EXPYs17“ that measure the following aspects of labor sophistication of Georgia’s exports: (i) median wage; (ii) median value added per worker; (iii) ratio of skilled to total workers; and (iv) capital stock per worker. For each of these labor-related EXPYs for Georgia, we: (i) look at how their levels have evolved over time since 2000; (ii) consider variations in labor-content across different destination markets, comparing the average for all Georgian exports (i.e., destined to the “world”) to those destined to the European Union (EU27), ECA*, and Russia; (iii) make bilateral comparisons of labor content with a range of top-performing ECA exporters including Estonia, Poland and Latvia (advanced “modernizers” in the region); and (iv) repeat each of the above analyses across the entire distribution to see what products are driving the observed differences. Since 2000, the median wage, value-added and skill content of Georgian exports has changed very little, though there is variation in these trends depending on destination. The solid red lines in Figure 3.4— depicting the labor sophistication for Georgia’s total world exports—suggest stagnating trends with respect to all four measures. Georgia’s exports to Russia behaved differently compared to its exports to the EU27 and ECA* countries between 2000 and 2006, embodying lower wage, value-added and skill content and much higher physical capital content. But in 2007, the wage and the skill ratio spiked upwards, and physical capital content experienced sharp declines, approaching average levels by 2010 and beyond. This trend was driven primarily by a sharp decline in exports of wine (and to a lesser extent base metals) to Russia. Georgia’s non-mineral exports have not gained in labor sophistication. Figure 3.5 shows that the indices for Georgia’s global exports of non-mineral goods (the solid bold line) have been mostly flat, with only marginal increases in median wage, value-added and skill ratio, and a small decline in physical capital content. This observation is consistent with the relatively modest compositional shifts in Georgia’s export basket over the past decade. The average ratio of skilled to total workers hovered around 64 percent. 17 Without access to detailed employment and wage data, we cannot derive precise measures of the labor-market impact of Georgia’s export performance. Adapting the Hausmann, Hwang and Rodrick (2007) measure of export sophistication, we develop indirect measures of the labor-related sophistication of Georgia’s export basket. EXPYs can be interpreted as the export share-weighted average labor market outcome associated with Georgia’s export basket, and are calculated using two-steps. The first associates to each product the global trade-weighted average wage, value-added, skill ratio, or physical capital covering all countries in the world that export the product, denoted PRODY. The second step then weights the PRODYs appearing in Georgia’s export basket by the share of each product in Georgia’s total exports, denoted EXPY. These resulting EXPYs reflect the global average labor content of goods that appear in Georgia’s export basket. 36 | Chapter 3: Trade, Labor Outcomes, and Skills Draft 5.indd 54 22/12/14 15:19:09 Georgia: Seizing the Opportunity to Prosper Figure 3.4. Labor Sophistication of Georgia’s Exports over Time and by Destination EXPY median wage, PPP thousands EXPY median value added, PPP thousands 30 100 90 25 80 70 20 60 15 50 40 10 30 20 5 10 0 0 2000 01 02 03 04 05 06 07 08 09 10 11 2012 2000 01 02 03 04 05 06 07 08 09 10 11 2012 EXPY skill ratio EXPY capital stock per worker, US$ thousands 80 120 70 100 60 80 50 40 60 30 40 20 20 10 0 0 2000 01 02 03 04 05 06 07 08 09 10 11 2012 2000 01 02 03 04 05 06 07 08 09 10 11 2012 ▬▬EU27 ▬▬Russia ▬▬ECA* ▬▬World Source: World Bank Staff Calculations. Georgia’s exports to the EU have become less sophisticated in skills content in the past decade, but exports to ECA have increased in sophistication, though mainly because of growth in wine and processed food exports to ECA. Disaggregating non-mineral exports into the EU27, Russia, and the non-EU- and non-Russia ECA destination markets in Figure 3.5 shows a minor decline in median wage and value-added content of exports destined to EU27 from 2007 onward, and a more marked decline in terms of physical capital content, contrary to the trend for non-Russia ECA. ECA in particular absorbed Georgia’s wine and processed food exports after the Russian market closed, which embody relatively higher physical content compared to Georgia’s other exports. This suggests that diversifying Georgia’s export basket by targeting barriers to entry—particularly in terms of the national quality infrastructure, which needs to be upgraded to international standards—toward the EU market is essential to leverage the DCFTA with the EU. Georgia’s export development lags behind regional performers such as Estonia, Poland and Latvia: especially noteworthy is its relative inability to progress up the product chain. Based on the same EXPY indicators of embedded labor sophistication, Figure 3.5 shows that Georgia’s non-mineral global exports (the bold solid line) are characterized by relatively higher wages and value-added content compared to Estonia, Poland and Latvia, and these have increased albeit marginally since 2000. On the other hand, the skills and physical capital content of Georgia’s exports have slipped to the bottom in relative terms. What is most notable is that Georgia’s | Chapter 3: Trade, Labor Outcomes, and Skills   37 Draft 5.indd 55 22/12/14 15:19:10 Georgia Country Economic Memorandum | December 2014 Figure 3.5. EXPY Indicators Showing Labor Content of Georgia’s Exports Over Time and by Destination EXPY median wage, PPP thousands EXPY median value added, PPP thousands 25 70 60 20 50 15 40 30 10 20 5 10 0 0 2000 01 02 03 04 05 06 07 08 09 10 11 2012 2000 01 02 03 04 05 06 07 08 09 10 11 2012 EXPY skill ratio EXPY capital stock per worker, US$ thousands 70 140 60 120 50 100 40 80 30 60 20 40 10 20 0 0 2000 01 02 03 04 05 06 07 08 09 10 11 2012 2000 01 02 03 04 05 06 07 08 09 10 11 2012 ▬▬EU27 ▬▬Russia ▬▬ECA* ▬▬World Source: World Bank Staff Calculations. Note: Excluding mineral and base metal exports. trends are relatively subtle and fairly stagnant, whereas the regional performers show more dynamic progress up the product chain. The largest differences emerge with respect to physical capital, with the performing countries increasingly exporting products exported by countries with high capital stocks per worker. Indeed, Georgia’s export sectors have not benefited to the same degree from capital inflows over the past decade, especially the very high levels of FDI attracted by Poland and to a lesser extent Estonia. Georgia’s exports to ECA are less sophisticated compared to ECA-destined exports from Estonia, Poland and Latvia reflecting differences in the composition of the export basket. To explore whether the changing composition of Georgian exports to the ECA* market may have implications for labor demand, we compare the labor content of Georgia’s ECA*-destined exports with ECA*-destined exports from Estonia, Poland and Latvia. Whereas Georgia is broadly in line with regional competitors with respect to wage and value-added content of exports, its exports lag significantly in terms of skill content (the ratio of skilled workers), and capital content (the capital stock per worker). But the trends since 2000 vis-à-vis skill and physical capital are positive, suggesting that expanding Georgian exports to ECA* has been accompanied by increased labor sophistication in these indicators, differences in the implied skill ratio and capital stock per worker of Georgia’s export basket remain significant and are likely to be persistent given that they would require considerable time and resources to close. 38 | Chapter 3: Trade, Labor Outcomes, and Skills Draft 5.indd 56 22/12/14 15:19:10 Georgia: Seizing the Opportunity to Prosper Figure 3.6. Comparing Georgian Exports to Export of Other Emerging Markets in ECA EXPY median wage, PPP thousands EXPY median value added, PPP thousands 25 70 60 20 50 15 40 30 10 20 5 10 0 0 2000 01 02 03 04 05 06 07 08 09 10 11 2012 2000 01 02 03 04 05 06 07 08 09 10 11 2012 EXPY skill ratio EXPY capital stock per worker, US$ thousands 70 110 68 100 66 64 90 62 80 60 58 70 56 60 54 52 50 2000 01 02 03 04 05 06 07 08 09 10 11 2012 2000 01 02 03 04 05 06 07 08 09 10 11 2012 ▬▬LVA ▬▬EST ▬▬POL ▬▬GEO Source: World Bank Staff Calculations. Note: Excluding mineral and base metal exports. We examine distributions across export product ranges to point to particular products that have the potential to spur Georgian labor demand. Instead of comparing the weighted average values of these export- labor indicators, we can examine the distribution of these same indicators across the range of products that Georgia exports. The cumulative distributions plotted show the cumulative share of products (on the vertical axis) at each value of the variable on the horizontal axis: (i) log median wage; (ii) log value-added; (iii) share of employees with a secondary degree; and (iv) log capital stock per worker. Georgia has increased the labor sophistication of some export products. We compare the cumulative distributions of each labor content indicator from 2000, 2008 and 2012 to consider this question. We observe some increase in labor sophistication across the entire distribution of export products, denoted by a shift to the right between 2000 and 2008, particularly with respect to wages and value-added. The skills content of exports saw an increase in the lower half of the distribution, reflecting an increased concentration of exports associated with a skill ratio around 55–65 percent, but a diminished share of exports with a skills ratio above 80 percent. There has been little change between 2008 and 2012, and therefore we do not present those results. | Chapter 3: Trade, Labor Outcomes, and Skills   39 Draft 5.indd 57 22/12/14 15:19:10 Georgia Country Economic Memorandum | December 2014 Figure 3.7. Cumulative Distribution of Georgia’s Exports to the World in 2000 and 2008 Cumulative distribution; kernel=epanechnikov, degree=0, bandwidth=.14 Cumulative distribution; kernel=epanechnikov, degree=0, bandwidth=.19 1.0 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 8 9 10 11 12 13 8 9 10 11 12 13 Median wage Median value added Cumulative distribution; kernel=epanechnikov, degree=0, bandwidth=.04 Cumulative distribution; kernel=epanechnikov, degree=0, bandwidth=.12 1.0 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 0 0.2 0.4 0.6 0.8 1.0 8 9 10 11 12 13 Skill ratio Capital stock per worker ▬▬2008 ▬▬2000 Source: World Bank Staff Calculations. Note: Excluding mineral and base metal exports. Different destination markets spur different trends in the labor content of exports. We detect notable differences for Georgian goods exported to the EU27 compared to ECA at the product level. As Figure 3.8 illustrates, exports to ECA have a higher wage, value-added and skills content across all or most of the product distribution curves compared to exports to the EU. In contrast, the difference in the physical capital content is driven by a greater density of exports to ECA* of products at the higher end of the distribution while exports to the EU dominate the lower end. This finding—that the labor sophistication of exports across various indicators to the EU is lower than other regional markets—is consistent with observed trends in other countries including Turkey (with the ECA market) and South Africa (with the Sub-Saharan African market). Enhanced access to the EU market under the Association Agreement, which entails significant institutional reforms and improvements in the national quality infrastructure and enforcement standards, may help reverse this trend. Georgian manufacturing exports are more concentrated in products with lower skill content, mainly metals and spirits, compared to regional performers, whose export baskets reflect higher shares of more sophisticated manufacturing. Figure 3.8 illustrates that the skill content is more concentrated at the lower end of the distribution in Georgia, whereas exports from Estonia and Poland, and to a lesser extent Latvia, 40 | Chapter 3: Trade, Labor Outcomes, and Skills Draft 5.indd 58 22/12/14 15:19:11 Georgia: Seizing the Opportunity to Prosper Figure 3.8. Cumulative Distribution of Georgia’s Exports to EU27 and ECA in 2012 Cumulative distribution; kernel=epanechnikov, degree=0, bandwidth=.17 Cumulative distribution; kernel=epanechnikov, degree=0, bandwidth=.15 1.0 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 8 9 10 11 12 13 8 9 10 11 12 13 Median wage Median value added Cumulative distribution; kernel=epanechnikov, degree=0, bandwidth=.06 Cumulative distribution; kernel=epanechnikov, degree=0, bandwidth=.11 1.0 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 0 0.2 0.4 0.6 0.8 1.0 8 9 10 11 12 13 Skill ratio Capital stock per worker ▬▬EU27 ▬▬ECA Source: World Bank Staff Calculations. Note: Excluding mineral and base metal exports. embody higher levels of sophistication both in terms of physical capital and skills. This is essentially explained by product heterogeneity and the compositional mix of exports. Georgia’s sharper increase in skill level around the point where 50–65 percent of workers have a secondary degree is driven by manufactured base metal products (using iron and steel) and beverages (wine, spirits, non-alcoholic beverages), after which the skill level flattens. In Poland and Estonia, by contrast, the skill level keeps increasing, and the driving products are more sophisticated manufacturing products such as television, radio, telephone line apparatuses, electric motors, generators, transformers, electric distribution control apparatus (in Poland), and television and radio receivers, sound or video recording devices, plastics, and rubbers (in Estonia). Similarly for the embedded average wage for global exports, Estonia and Latvia experience a jump driven by their relatively large share of wood products and furniture manufacturing, as well as apparel exports. For Poland, the jump is explained by furniture manufacturing, plastics and food products. | Chapter 3: Trade, Labor Outcomes, and Skills   41 Draft 5.indd 59 22/12/14 15:19:11 Georgia Country Economic Memorandum | December 2014 Figure 3.9. Cumulative Distribution of Georgia’s Exports Compared to Exports from Estonia, Poland and Latvia in 2012 Cumulative distribution; kernel=epanechnikov, degree=0, bandwidth=.14 Cumulative distribution; kernel=epanechnikov, degree=0, bandwidth=.17 1.0 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 8 9 10 11 12 13 8 9 10 11 12 13 Median wage Median value added Cumulative distribution; kernel=epanechnikov, degree=0, bandwidth=.04 Cumulative distribution; kernel=epanechnikov, degree=0, bandwidth=.11 1.0 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 0 0.2 0.4 0.6 0.8 1.0 8 9 10 11 12 13 Skill ratio Capital stock per worker ▬▬GEO ▬▬EST ▬▬POL ▬▬LVA Source: World Bank Staff Calculations. Note: Excluding mineral and base metal exports. C. Strengthening the Link between Exports and Labor Market Outcomes Exporting firms tend to be larger and more productive compared to non-exporting firms and are largely associated with better labor market outcomes, depending on the ease with which jobs can be created. Exporters are larger and more productive than non-exporting firms18. There is also evidence that exporters achieve better labor market outcomes than non-exporters, both in terms of employment and wages.19 Typically the quality of trade-related policy and institutions matters. In countries where it is more difficult to create new jobs, a smaller fraction of firms participate in export markets. And among firms that export, those that find it more 18 Bernard and Jensen 1995, Bernard and Jensen 1999, Bernard, Jensen, Redding and Schott 2007, Clerides, Lach and Tybout 1998. 19 Bambrilla, Lederman and Porto 2012. 42 | Chapter 3: Trade, Labor Outcomes, and Skills Draft 5.indd 60 22/12/14 15:19:11 Georgia: Seizing the Opportunity to Prosper costly to create new jobs export less intensively (Seker 2010). If exporting versus selling to the domestic market is causally linked to greater job creation and higher wages, then improving trade logistics, facilitating labor mobility or strengthening export promotion policies can strengthen the impact of trade on labor market outcomes. Georgian exporters create more jobs than firms that do not export. This can be verified more rigorously with our firm-level survey data. Following the methodology of Cebeci, Lederman and Rojas (2013), we test for a causal effect of firms’ export growth (measured by the change in a firm’s share of exports in total sales) on their labor market outcomes, namely employment level (total and female employment) and average wages (average for total firm employment, and for female employees).20 In addition, we assess whether exporting to different destination markets such as the EU or high-income countries has different impacts on labor-market outcomes.21 Within Georgia, raising the export share leads to higher employment. There is a strong causal relationship between exports and job creation in Georgia, with an increase by one percentage point in the share of sales that are exported causes employment levels to increase by 0.30 percent. The impact on wages is much stronger though less reliably predictable. The positive impact on employment of overall exports appears to be driven by males entering and replacing females, as female employment is lower in exporting firms. A one percentage point increase in the share of sales that are exported causes female employment levels to decrease by 0.19 percent. What is driving this shift, however, is unclear. While the increasing wages may be a reflection of greater skilled versus unskilled employment, the labor force with a secondary education in 2010 is similar between females and males (58.9 percent and 61.3 percent, respectively). Destination matters for female employment in the export sectors. Exporting to the EU and other high- income countries is correlated with higher female employment levels, possibly because of the rapid growth in recent years of agriculture and services based exports to this market22. Exports to the EU that are concentrated in mineral products and chemicals, two sectors that are capital intensive, do not generate substantive labor-market opportunities for women. Mining and quarrying accounted for only 1 percent of total employment in 2011 and 89 percent of those workers were male. Female employment is instead concentrated in agriculture (60 of female employment) and services (36 percent). These sectors that have experienced increasing exports shares—albeit from a low base—are likely behind the positive results in terms of female employment. Agriculture exports in particular witnessed strong growth to the EU market over the last decade. Even as Georgia implements policies to support firm productivity and facilitate trade, the impact of growing exports on labor market outcomes will depend on how resources—including workers—move to expanding sectors. Trade patterns and trade policies affect labor outcomes through various channels, especially relative prices, changes in which call for adjustments in consumption and production, and therefore in the demand 20 Whereas an OLS regression can identify statistically significant correlations between changes in the structure of firms’ sales and employment and wage growth, we employ an instrumental variables approach to identify causal relationships. We use exogenous fluctuations in exchange rates combined with firms’ initial exposure to various markets as instrumental variables (see Cebeci, Lederman and Rojas 2013, and Hollweg and Ruppert Bulmer 2014). 21 Because a panel dimension is necessary to conduct the analysis of changes in export shares and changes in labor market outcomes, the selection of firms was limited to those observed in the data across multiple years. We use the National Statistics Office of Georgia (GEOSTAT) Statistics Survey of Enterprises for 2007–11. 22 We test to see if different destination markets have different employment effects and consider the EU, ECA* and Russia, stratifying destinations by income level, namely high-, upper-middle, lower-middle and low-income countries. | Chapter 3: Trade, Labor Outcomes, and Skills   43 Draft 5.indd 61 22/12/14 15:19:11 Georgia Country Economic Memorandum | December 2014 for labor and wages. An important transmission channel through which trade policies can affect labor outcomes is through relative prices. Trading incentives are guided by relative prices of domestic versus foreign production, and when a trade parameter changes—such as through a tariff change or a broader persistent price shock in international markets—the change is transmitted to the relative price of affected products. In theory, producers and consumers respond to these new prices by altering their production and consumption of these goods, which in turn affects the demand for labor in the production of these goods. In the case of an increase in labor demand, workers will respond by increasing their labor supply to that sector, drawing labor away from other sectors. In practice labor adjusts slowly to changing incentives. Whereas we typically assume that markets clear instantaneously—that is, that supply adjusts to price signals (i.e., demand), in reality, labor supply does not adjust immediately. Changing jobs is costly. These costs stem from various sources, and vary across individuals and industries. Labor mobility costs might include periods of unemployment and job search, re-training for a new sector or technology, geographical relocation, or family ties that increase the cost of moving. Evidence from Mexico indicates that non-wage factors are more important than wages in determining job transitions, and 86 percent of voluntary job exits are in fact driven by changes in family circumstances such as marriage or family care (Kaplan, Lederman and Robertson 2013). When labor mobility costs are large, workers delay or avoid adjustment, with consequences for productivity growth and job creation. Mobility costs not only slow the pace of adjustment, but also reduce the degree of labor reallocation, which in turn affects equilibrium wages. In the event of costly labor reallocation across industries and employers, the gains from trade might be reduced. Idle labor or labor stuck in lower productivity firms and industries implies lost incomes and diminished gains to growth. For policymakers considering trade and/or labor policy changes, it is important to understand the relative magnitudes of “labor-mobility costs”, or the costs faced by workers when searching for alternative employment in response to changing labor demand and relative wages. The Georgian labor market has high mobility costs, not atypical of countries where agriculture is the largest employer.23 We estimate transition costs across sectors for all workers and for differentially skilled workers using quarterly data from the 2009 and 2011 household surveys. Table 3.2 shows the transition matrix for Georgian workers for 2011. Each cell contains the transition statistic for the average share of workers transitioning from each origin sector (in each row) to all other destination sectors (in each column). The cells on the diagonal indicate the shares remaining in its current work/sector status; these “stayers” represent the largest shares. These transition statistics give a sense of the fluidity of the Georgian labor market, even during a relatively short timeframe. Take the unemployed, for example. Sixty-two percent of those unemployed at the beginning of the period remained unemployed, while 15 percent exited the labor force, 11 percent entered the agriculture or other primary sectors, and the remaining 13 percent found employment in various other sectors. Among those outside the labor force, 11 percent entered into agricultural and mining employment, providing additional evidence that agriculture acts as the residual or “last resort” sector, absorbing workers unable to find alternative employment. We observe the largest number of transitions out of construction work, a sector typically characterized by significant demand- driven volatility. Public sector service jobs, by contrast, have the lowest share of exits, suggesting that these jobs 23 Larger mobility costs would lead to higher wage differentials between sectors without corresponding changes in labor allocations, with the converse true if mobility costs are small. The framework is based on an emerging literature that relies on structural models of sectoral employment choices with costly labor adjustment (see, for example, Artuç, Chaudhuri, and McLaren 2010). The total cost is interpreted as the cost a worker would have to incur to move between sectors for a given wage differential. 44 | Chapter 3: Trade, Labor Outcomes, and Skills Draft 5.indd 62 22/12/14 15:19:11 Georgia: Seizing the Opportunity to Prosper are highly desirable, and the lowest share of entry from other sectors, suggesting that these jobs are difficult to obtain. Table 3.2. Worker Transitions Across Sectors and Into and Out of Labor Force Status in 2011 In percent Unemployed Out of Agriculture, Manufacturing Construction Public Private Labor Hunting, Services Services Force Forestry, Jobs Jobs Fishing, Mining and Quarrying Unemployed 62 15 11 2 3 2 6 Out of labor force 9 77 11 1 1 1 2 Primary* 3 6 87 1 1 1 2 Manufacturing 5 2 7 75 3 1 7 Construction 13 3 13 3 60 2 6 Public services jobs 2 1 3 0 0 89 6 Private services jobs 5 3 5 1 1 4 81 Source: Authors’ estimates based on the Georgia Integrated Household Surveys 2009 and 2011. Note: Includes agriculture, hunting, forestry, fishing, mining & quarrying. Combining the transition data with the observed wage gaps between sectors, we are able to estimate the labor mobility costs as a ratio of the annual average sectoral wage  (Table 3.3). We observe the following interesting trends: •• The agriculture sector is the least costly sector to enter for all groups, and particularly for the unskilled. Entering workers incur a cost equivalent to 2.6 times the annual average agriculture sector wage, and unskilled workers incur cost equivalent to 2.4 times the annual average sector wage. •• Construction jobs and service jobs in the private sector represent the next most accessible sectors following agriculture. •• Mobility costs increased slightly between 2009 and 2011 (with the exception of the agriculture sector). •• Public sector service jobs have the highest mobility costs, especially among unskilled workers, and this is consistent with the very low observed turnover in public sector jobs. •• Entering manufacturing employment is relatively costly, for both skilled and unskilled workers. This result suggests a certain level of sector-specific knowledge is needed to access manufacturing jobs, in contrast to agriculture and to a lesser extent construction and private services. The estimated labor mobility cost for finding a manufacturing sector job was equivalent to 4.65 times the average manufacturing wage in 2011, up from 4.35 in 2009. The increased cost was even larger for unskilled workers, reaching 5.23 times the annual average sector wage. Note that the average sector wage is used as the denominator; for unskilled workers with relatively lower wages, the mobility cost represents an even larger ratio of the average unskilled wage. | Chapter 3: Trade, Labor Outcomes, and Skills   45 Draft 5.indd 63 22/12/14 15:19:11 Georgia Country Economic Memorandum | December 2014 •• Leaving employment either by becoming unemployed or exiting the labor force involves the lowest transition cost, which suggests that there may be support systems that facilitate “dropping out.” Table 3.3. Mobility Costs As a ratio of the average annual sectoral wage 2009 2011 Total Unskilled Skilled Total Unskilled Skilled Unemployed/Out of labor force 1.94 1.90 2.09 2.10 2.05 2.24 Agriculture, hunting, forestry, fishing, mining and 2.66 2.40 2.88 2.60 2.39 2.76 quarrying Manufacturing 4.35 4.64 4.09 4.65 5.23 4.32 Construction 3.34 3.21 3.50 3.64 3.59 3.73 Public services jobs 4.60 5.08 4.15 4.84 5.02 4.46 Private services jobs 3.14 3.54 2.76 3.27 3.31 3.18 Source: Authors’ estimates based on the Georgia Integrated Household Surveys 2009 and 2011, individuals aged 16–64. Unskilled is primary or secondary-level education. Skilled is technical school, college, or secondary-professional; bachelor’s degree; higher education or master’s degree; or doctor’s degree. An annual discount factor of 0.9 is assumed. Labor mobility costs in Georgia range from 3–5 times the annual average wage and are comparable to other markets dominated by the primary sector.24 The methodology used here is sensitive to the level of disaggregation: the higher the degree of sectoral disaggregation, the fewer the observed transitions and the higher the estimated mobility cost. It is therefore crucial to exercise caution when interpreting the results.25 This type of exercise is most useful for comparing the relative mobility costs across different sectors of the Georgian economy, particularly those most closely affected by trade rather than for international benchmarking. There is however evidence that such high costs are typical of countries with a large share of primary employment and lower share of tertiary education. The benefits to Georgia of trade-oriented policies, including closer ties with the EU, will depend on the ease and speed with which resources, including labor, can reallocate to higher productivity sectors. As is typical of countries with large shares of employment in agriculture and primary sectors, labor mobility costs are high, especially as a proportion of the unskilled wage. Clearly policy efforts are needed to tackle the constraints that drive these trends if the market access afforded by closer ties with the EU is to have the desired impact on trade and growth. Benefiting from the DCFTA will require that resources—including labor—be able to move to sectors with higher productivity levels and growth potential. 24 Note that the Integrated Household Survey data show quarterly transitions, which we average for each period (i.e., for 2009, and 2011). By using quarterly rather than annual transitions, the resulting lower number of observed transitions biases the mobility costs upward. 25 For example, when we repeat the estimation disaggregating across 14 sectors (instead of the six shown above), the mobility costs for the smaller sectors such as mining, utilities, and financial intermediation are quite high, but the estimates for agriculture, manufacturing and construction remain very close to the estimates reported in Table 3.3. 46 | Chapter 3: Trade, Labor Outcomes, and Skills Draft 5.indd 64 22/12/14 15:19:11 Georgia: Seizing the Opportunity to Prosper D. The Skills Context in Georgia Georgia performs relatively poorly on assessments of the functional literacy of 15-year-olds by these measures. Results from the Programme for International Student Assessment (PISA) 2009+ show that only between 30 and 40 percent of Georgian 15 year-olds score Level 2 or above, which is a threshold usually referred to as functional literacy. Memorization and rote learning are still prevalent. Despite the financing reform and the declining school age population, the education sector continues to be underfunded and faces significant obstacles to attract the best students into the teaching profession. These are serious concerns: research by OECD (2010) has estimated that the impact of improvements on academic quality—as measured by PISA on countries’ future economic growth as being significant. According to the study, a 50 points increase in PISA scores (half standard deviation or one year and a quarter of education) is associated with 0.9 percentage points higher growth rates in the long-term (that is, the impact will be felt over a period of 50 years). An accumulating body of evidence26 shows that investment in skills at an early stage is an effective policy strategy, and Georgia lags in terms of access and quality. It has been shown that, due to the nature of the skill formation process, early investments are those with the highest rate of return. Both medical and economics literature agree that gaps in both cognitive and non-cognitive skills emerge early in life and that policies to mitigate these gaps later in life have proven very costly. Moreover, these effects seem to be more pronounced for females and disadvantaged groups. Evidence shows that having been enrolled in at least a year of pre-primary education has a positive impact on PISA performance even after controlling for factors including socio-economic status and parental education. However, at 46 percent, the enrollment rate in pre-primary education in Georgia still lags behind comparator countries such as the new member states, with local government capacity to formulate and enforce standards raising doubts about quality. There are also wide gaps in participation between the rich and the poor, and between urban and rural areas. Skill mismatches are a significant constraint to both productivity growth and labor mobility in Georgia has been identified as a skills mismatch27. The difficulty faced by firms in Georgia both in terms of becoming internationally competitive and progressing along the product chain is mirrored by findings based on an employer survey conducted in Georgia in 2013. World Bank (2013) also reported that skills are perceived as a significant constraint on firm productivity growth. A STEP survey implemented in Georgia over FY12/13 bears out these findings. The survey was stratified by activity and firm size. Five sectors were covered, namely construction, hospitality, ICT, Other Services, Trade, and Transport. A range of job skills, both cognitive and social/behavioral were surveyed at the high end as well as over the medium to low skill range. The main findings were that while there are many highly educated workers in Georgia, there are relatively few highly skilled jobs and therefore high unemployment, especially among the highly educated. Still, employers often can’t find worker with right skills since the education system does not meet 26 See Heckman and Masterov (2007), Heckman (2006), Cunha and Heckman (2006) and Carneiro and Heckman (2003) among others. 27 Georgia has an insufficient number of graduates from science and engineering disciplines. The majority of teaching programs at Georgian universities do not meet international standards, and their graduates are not well prepared to work in high-technology oriented companies. Reform of tertiary teaching and public research is already being targeted by a State Commission on Education and Science Reforms. Recently in collaboration with the U.S. Millennium Challenge Corporation, the Georgian government initiated new policy reforms of the educational and research sectors at public universities and research institutes. | Chapter 3: Trade, Labor Outcomes, and Skills   47 Draft 5.indd 65 22/12/14 15:19:11 Georgia Country Economic Memorandum | December 2014 Figure 3.10. Labor Force By Education and employers’ needs. Modern firms perceive the skills Location shortage as being especially severe in spite of the high In percent proportion of educated workers in the country. 100 90 18 At the same time new technologies, globalization, 80 50 18 the information revolution, and labor market 70 60 changes have affected the world economy on an 50 unprecedented scale and will continue to do so in 40 19 48 the foreseeable future. The key policy challenge for 30 Georgia therefore is to ensure that its emerging and 20 25 existing workforces have the skills needed to escape 10 6 poverty and embrace new opportunities. Currently, 10 0 3 2 there is potential to strengthen the effectiveness of the JJPrimary Urban JJVocational Rural JJSecondary general JJSecondary technical JJTertiary school system in providing basic skills such as literacy, Source: Rutkowski (2013). numeracy and problem-solving skills that are needed Figure 3.11. Modern Firms are Particularly Affected by Skills Shortage Percentage of Firms Reporting Problem as Major or Severe Percentage of Firms Reporting Problem as Major or Severe In percent In percent 30 30 25 25 20 20 15 15 10 10 5 5 0 0 Technical & vocational education General education of workers Labor availability General education of workers Technical & vocational training Labor availability JJInnovative JJTraditional JJInternational JJLocal Source: Rutkowski (2013). in almost any job. Georgia’s education system lags behind in imparting this basic set of skills (share of functionally illiterate students in PISA is an indicator of poor performance). Moreover, vocational education (levels I–III) starts too early and does not teach general skills that would enable the youth to adapt to changing economic circumstances. Rather, in most cases, VET is focused on specific occupations, which end up tying youth and further increasing labor mobility costs. In addition, school to work links need to be strengthened in both directions: employer incentives and capacity to provide on-the-job training needs to be taken into account when designing a skills development strategy. Georgia does not yet have a skills development strategy. Global trends relating to increased use of technology has led to a debate emerging around the mismatch between skills supplied by recent graduates and skills demanded in the labor market. In several countries, employers demand a sophisticated set of skills—including socio-emotional and higher-order skills, such as problem-solving—and education systems have not yet adapted 48 | Chapter 3: Trade, Labor Outcomes, and Skills Draft 5.indd 66 22/12/14 15:19:12 Georgia: Seizing the Opportunity to Prosper in order to impart these skills to students (Arias et al., 2014). High youth unemployment rates, particularly in Europe, (including among those with advanced degrees and even before the global financial crisis) and the increasing demand for and higher wages of employees with the right skills mix suggest a skills mismatch. Georgia faces a similar challenge. The skills mismatch in Georgia is consistent with a lower bound “simulated” 15 percent unemployment rate. Even if the economy creates additional jobs so that labor demand (jobs) = labor supply (employment + unemployment) there would still be unemployment due to an educational/skills mismatches. Under certain assumptions explained below, about 15 percent of the unemployed would not find jobs because their educational attainment is lower than that required in the newly created jobs (this is a lower bound estimate). There will be an excess supply (unemployment) of workers with secondary general education, and excess demand (shortage) of workers with university education even assuming away the issue of relevance of college degree or lack thereof. There are four main assumptions: an equilibrium condition is imposed on job growth so that the number of newly created jobs = the number of unemployed and labor demand (jobs) = labor supply (employment + unemployment); the educational structure of newly created jobs is the same as the structure of existing jobs, so the educational/skills structure of employment does not change; there is no mobility between occupational groups, i.e. workers Figure 3.12. Differences between Share in Total with secondary general education cannot take jobs Employment and Unemployment by Educational Attainment, 2012 requiring secondary technical education and workers with college degrees will not take jobs requiring University secondary education; and there is no horizontal skills College mismatch, i.e. there is no mismatch within educational groups (that might require matching of occupation Secondary technical with specific college degrees). Secondary general Various studies pointed to the lack of labor Vocational skill as a key obstacle that has impeded the Primary business development in Georgia. More than half Less than primary of the employees have a university degree or higher (Enterprise Survey 2013) and there is a large pool of -8 -6 -4 -2 0 2 4 In percent unemployed workers with higher education. However, Source: Rutkowski, 2014. the 2014–15 Global Competitive Index ranked Georgia as 92nd out of 144 countries for “higher education and training”. On one hand, traditional enterprises that engage in low-productivity activities find a lack of specific technical skill in the workforce; on the other hand, the modern enterprises that engage more innovative and high- productivity work complain the lack of high-skilled workers who can adapt to the changing technologies. The 2008 EBRD-World Bank Business Environment and Enterprise Performance Survey show that close to 30 percent of Georgian employers see inadequate workforce skills as a major obstacle to the operation and growth of their firms. Importantly, innovative and growing firms suffer from this skill shortage the most. The relevance of higher education programs needs to be strengthened. The Government has undertaken several steps towards improving higher education system since 2005. Concerted efforts have also been launched to introduce STEM (Science, Technology and Engineering) education in Georgia through MCC support. Nonetheless, the quality and relevance of higher education programs remains a concern. According to the World Bank’s “Skills | Chapter 3: Trade, Labor Outcomes, and Skills   49 Draft 5.indd 67 22/12/14 15:19:12 Georgia Country Economic Memorandum | December 2014 Mismatch and Unemployment Report (2013)”, over 50 percent of all unemployed have a secondary school diploma and as many as 40 percent have higher education. Although limited net job creation and scarcity of jobs requiring higher education is an important cause for unemployment, the skills mismatch is a contributing factor (Rutkowski 2008). Students continue to pursue degrees in business, humanities and law instead of sciences and engineering. Lack of career centers, advising centers, and orientation to help choose a career path is a factor here. Regarding the relevance of programs, examples of partnerships between higher education institutions and the private sector are rare. Recently, policy makers have turned to vocational education as a tool for addressing high unemployment among the youth. This is a valid alternative as it facilitates transition from school to employment in some cases, provided that tracking into vocational streams should not happen too early, i.e. not before students acquire strong generic knowledge. For instance, Poland has delayed tracking of its students by one year and Jakubowski et al. (2010) have shown that the country made significant progress as evidenced by improved PISA results partly due to the reform. Another concern with vocational education is the adaptability of graduates to an ever changing economy. Hanushek et al. (2011) approach this debate from a short vs. long term trade off perspective. They find evidence that while vocational education leads to a smoother transition immediately after school, it may hinder future employability as technical skills become obsolete. Even in Germany, whose vocational education system is considered a model, having general education has been advantageous in the last decades. In ECA, vocational graduates have lower youth unemployment rates but lower employment rates after age 25 and older28. In Georgia as well, the last few years have seen reforms to the vocational education and training system (VET) though there remains considerable scope for improvement, especially in engagement with the private sector, and strengthening mobility. Reforms include: (i) optimization and consolidation of vocational education institutions; (ii) introduction of a new financing system, which foresees three different types of financing, differentiated vouchers, program budget and targeted programs ; (iii) significant capital investments on new school infrastructure; (iv) the adoption of a National Qualifications Framework; and (v) a whole new strategy (2013–2020) that pledges to put more emphasis on quality of education, transferability of qualifications, elimination of “dead-ends” and counseling to help students find jobs (World Bank, 2014). Despite these Figure 3.13. Provision of Formal Training actions, quality remains a concern. In addition, the VET track provides very little or no general education. Not 15 only does this create difficulties for students that later Permanent employees 18 decide to pursue higher education, but more generally 2008 35 it also leads to a workforce that will have very little mobility given that only specific occupation is taught. 11 There are shortages of on-the-job and vocational Permanent employees 15 trainings, which are crucial for workers’ 2013 35 adaptability to their jobs. Only 11 percent of the surveyed firms reported to offer formal training in the 0 10 20 30 40 50 2013 Enterprise/BEEPS Survey. The 2014–15 Global Percentage of firms offering training for employees JJGEO JJCAU JJECA Competitiveness Index ranked Georgia at 114th out Source: World Bank staff estimates. Note: CAU denotes the South Caucasus. 28 See Arias et al. (2014), p. 238. 50 | Chapter 3: Trade, Labor Outcomes, and Skills Draft 5.indd 68 22/12/14 15:19:12 Georgia: Seizing the Opportunity to Prosper of 144 countries under the “on-the-job training category”. On the availability of research and training services Georgia was ranked as 116th. Over 90 percent of the respondents to a non-representative survey29 said they would like to be self- employed suggesting Georgians have an entrepreneurial spirit. In fact, the share of self-employed is already high among workers older than 40. A World Bank study suggests measures to promote entrepreneurship and job creation through new, more innovative companies including through; (i) increased access to finance to small and medium enterprises (SMEs); (ii) a simplified the tax system; (iii) facilitation of firm exit and restructuring; and (iv) outreach to raise awareness of the possibility of taking on entrepreneurial activities and opportunities for capacity-building. E. International Best Practice in Skills Development A long term vision is a common characteristic of countries that have successful skills development strategies. Policy makers identify a growth strategy and the sectors of the economy more likely to lead the country to the envisioned goal. Afterward, the needs of these sectors—in terms of workforce, financing and institutional framework—are addressed. For that purpose, manpower needs are estimated and schemes to incentivize firms to train their workforce introduced. Also, successful countries have been flexible enough to adjust their strategies to the changing economic circumstances. In the late 1960s and 1970s, the Government of Singapore set up a program to attract foreign companies and engage them actively in training. This model was pioneered by Tata (India), Phillips (Netherlands) and Rollei (Germany). The training model was based on the German apprenticeship scheme and incentives to the firms included tax free remittances, low interest loans, reasonably priced land, cost-sharing operational costs between firms and government, and right of refusal (in the case of Rollei), potential competitors could only operate in Singapore if Rollei decided not to produce that particular good). Additionally, while firms had priority to hire their centers’ graduates, they also needed to train more people than they planned to hire. Since the focus on individual companies would not be enough to address the country’s needs, the model evolved to embrace partnerships between Singapore and countries such as Germany, France and Japan. Heavy foreign direct investment followed and led Singapore to develop expertise in innovative industries (Kuruvilla et al., 2002; Kuruvilla and Chua, 2000). Korea relied on a similar strategy of choosing pioneers and helping them meet their needs. The government played the dominant role in planning, financing and regulating training in the 1960s and 1970s. However, the responsibility for service provision was delegated to the private sector. Vocational education and training was at the heart of Korea’s strategy. Not only was there demand for technical skills, but it also allowed for people that would otherwise drop out to continue education, i.e. it did not compete with tertiary education. Gradually, vocational education as an end point lost ground to higher education (Ra and Shim, 2009). In 2005, Ireland called on a group of experts, the Expert Group on Future Skills Needs, to design a strategy for the country to become competitive, innovation-driven, knowledge-based, participative and inclusive 29 See Georgia Rising (2013), World Bank, Washington DC, p. 19. Original source is Natsvlishvili (2011). | Chapter 3: Trade, Labor Outcomes, and Skills   51 Draft 5.indd 69 22/12/14 15:19:12 Georgia Country Economic Memorandum | December 2014 economy by 2020. As in East Asia, skills needs were identified and measures to address them proposed. Particular emphasis has been put in ensuring basic skills, such as literacy and numeracy, are imparted to everyone. Low skilled and disadvantaged populations, including immigrants, are targeted under the new strategy. Ireland’s strategy is shaped around the identification of individuals’ and employers’ needs, provision of flexible and responsive training, national media awareness campaign, accreditation/quality assurance system, and adequate funding. To that end, the experts group devised a strategy to increase both employees’ and employers’ incentives to engage in training. Moreover and very importantly, the push toward skills development is a comprehensive policy, i.e. it spans several government departments to assure coherence (Expert Group on Future Skills Needs, 2007). In an attempt to address the complaints by the business community that the government was not providing good public education to its population, India has established the National Skill Development Corporation (NSDC) through a public-private partnership. NSDC is a non-profit owned by the private (51 percent) and public sectors (49 percent). However, it is designed and run by the private sector, which has autonomy over its resources. The government has provided seed capital for the program and the NSDC is expected to train 150 million workers by 2022, with focus on over 20 selected sectors of the economy. Among NSDC’s activities are funding vocational training initiatives through loans or grants, setting up standards and accreditation systems, student placement mechanisms, identifying and helping in the development of workers’ skills30. Innovative firms already account for a large share of the jobs created world-wide: therefore, the skills demanded by these firms cannot be neglected31. Moreover, fostering entrepreneurial activity will translate into more job creation in the future. Söderham, a small city in Sweden, offers another form of long term vision: teaching entrepreneurship from an early stage. In the past, the city observed big companies closures, demographic shift towards a shrinking working age population, high unemployment rate and increasing immigration. Local authorities decided to develop around innovation in the service sector in order to face the city’s challenges. The first component of the new strategy was the inclusion of entrepreneurship as an integral part of the basic education system from an early stage. This was an overarching task that involved a paradigm shift. In school, children were stimulated to develop their ideas and not to fear failure. Among other activities, students 16–20 years old come up with ideas and develop business plans in “Business Labs” with the help of a Swedish NGO and also participate in summer entrepreneurship programs. Moreover, partnerships between schools and companies were established so that ideas with potential could be taken to the next level. Between 2007 and 2010, the number of student beneficiaries of the NGO program has increased by almost 100 percent and covers not only business students, but also other fields (Högberg, 2012). Information can play a critical role in mitigating the skills gap as demonstrated by the experience of many countries. In Chile, the Ministry of Education has set up a portal on the internet that contains detailed information about the employability and earnings of graduates. Through a unique ID, data from the tax payments are crossed with the education records. The portal targets prospective tertiary education students and provides them with details of employment rates and earnings by income quintiles, fields of work, and type of institution attended (in 30 See http://nsdcindia.org/index.aspx. 31 Georgia’s innovative firms demonstrate significantly higher level of performance than non-innovative firms, and this difference is bigger than in the rest of the ECA region. The World Bank Report (Fostering Entrepreneurship in Georgia 2013) shows that significant differences exist in Georgia between firms that innovate and firms that do not. The annual real sales growth of innovative firms in Georgia was twice as high as in ECA10 and was significantly higher than that of Georgian non-innovative firms. Similarly, employment growth rates were significantly higher for innovative firms in Georgia than for innovative firms in ECA10. 52 | Chapter 3: Trade, Labor Outcomes, and Skills Draft 5.indd 70 22/12/14 15:19:12 Georgia: Seizing the Opportunity to Prosper some cases the actual institution can be tracked). Moreover, students can see how earnings evolve until 10 years after graduation32. Colombia’s Graduados provides information on labor market’s supply and demand, earnings, and average time to find the first job33. Initiatives like these have a great potential to prevent an oversupply of workers in a particular sector and shortages in another. In addition, they promote a more informed discussion about career paths among secondary education students. Established in 1994, Italy’s AlmaLaurea is a database of graduates’ résumés. Run by a consortium of universities, it gathers information on approximately 80 percent of the country’s graduates. Its main objectives are to facilitate and improve access of young people to the labor market and monitor university courses and their quality. The portal provides advice for jobseekers, a large bank of job offers and also information about companies. Prospective students can also learn more about post-graduate programs’ characteristics. As technological and demographic trends leads to the obsolescence of skills that were once important, continuous education (on-the-job, second chance and for the unemployed) is increasingly more important. Many EU countries provide some type—which may vary—of incentive for firms to invest in the human capital of its employees. In the Netherlands, for instance, firms can claim up to 140 percent of their training expenditures as tax deductions. In 2006, the North Rhine-Westphalia state of Germany introduced training vouchers that cover up to 50 percent of the fees. The remaining is funded either by firms or employees and candidates need to consult approved agencies before being granted the incentive. Training funds are also popular. Firms, according to their size, pay a share of their payroll (usually around 1.0 or 1.5 percent) and this fund is used for training of employees. In many cases, firms can be granted deductions based on their workforce development expenditures. In the case of training funds, evidence suggests that sectoral levy schemes are in place. This reduces cross-subsidization across industries and makes employers more likely to actively participate. State employment agencies can have an important role in helping unemployed find jobs. Not only do these agencies provide a whole host of services and counseling, but they also channel unemployed to skill-enhancing activities. Denmark has a particularly developed system in which the unemployed are divided into different profiles according to factors such as health, length of unemployment, social problems, lack of social competences and motivation. Depending on the profile, the unemployed have to have frequent personal contact with counselors and participate in employment-directed programs. As a result of active labor market policies, Copenhagen has managed to keep unemployment rates low for a country that provides substantial unemployment benefits. The strategic focus on educational training, practical work training and interactions with enterprises has helped keep youth unemployment even lower than the overall unemployment rate (Müller and Behringer, 2012). Several countries have strived to link academic research to the needs of the employers. In Latin America, countries such as Chile, Uruguay, Brazil and Argentina have had low Research & Development investments (and very low private participation in it). While in some cases Technology Transfer Offices (TTOs) are set up at the university level, many countries have recently organized National TTOs, which facilitate the transfer and adaptation of knowledge from universities and research centers to industries. In ECA, Macedonia is working towards creating its own TTO. These offices can provide technical solutions to firms, help manage intellectual property and even support the creation of start-ups. In the US, some universities have their own TTOs, whose mission is to bridge research and industry and make knowledge available for public use and benefit. Not only do 32 See http://www.mifuturo.cl/. 33 See http://www.graduadoscolombia.edu.co/html/1732/channel.html. | Chapter 3: Trade, Labor Outcomes, and Skills   53 Draft 5.indd 71 22/12/14 15:19:12 Georgia Country Economic Memorandum | December 2014 Box 3.1. More is Needed on Innovation Fostering new and innovative firms is critical to improve job creation in ECA (World Bank 2014). In the region, about 10–15 percent of the firms—most of them young—account for approximately two thirds of net job creation. Innovation, especially technology adoption and adaptation, is critical for Georgian firms to increase productivity. How does Georgia fare? The 2014–15 Global Competitiveness Index ranked Georgia’s capacity for innovation at the 121th out of 144 countries. The Global Innovation Index ranking, based on elements of the enabling environment for innovation, ranked Georgia at the 72nd in its 2014 report. The country’s ranking has not changed much since 2011. Both the research and development (R&D) and capability to adopt or emulate technology were low in Georgia. The Global Competitiveness Index ranked the quality of scientific research institutions and university-industry collaboration in R&D at 119th and 128th respectively. Company spending on R&D is ranked the 126th. According to the 2012 World Bank Entrepreneur Survey, only 7 percent of surveyed entrepreneurs indicated that their firm had introduced a new or substantially improved product or services in the previous three years. More than 90 percent of surveyed firms had no R&D expenditures in the previous five years and did not envision spending on R&D in the next two years. No entrepreneur reported any Georgian products or services new to the world. Universities, technical institutes, R&D firms, and external commercial labs were among the least important sources of knowledge for companies, indicating a lack of innovative activities and industry-relevant research. The number of Intellectual property filings and the number of patent applications during 2010–12 were lower than the pre-crisis years. Filing of intellectual properties in 2012 was 153 cases, down from 267 in 2009. The number of local patent applications in 2012 was half of the 2009 numbers of applications. In addition, survey results from the enterprise and BEEPS surveys showed that the percentage of firms which developed new products in the past three years has dropped from 35 percent in 2008 to 9 percent in 2013. Figure 3.14. Fewer Firms Developed New Figure 3.15. Fewer Firms Spent on R&D Products In percent In percent 60 60 50 50 51 45 40 40 35 30 30 25 20 20 22 16 10 10 14 9 9 7 4 3 0 0 GEO FSU-S ECA GEO FSU-S ECA JJ2008 JJ2013 JJ2008 JJ2013 Source: BEEPs and staff calculations. Source: BEEPs and staff calculations. 54 | Chapter 3: Trade, Labor Outcomes, and Skills Draft 5.indd 72 22/12/14 15:19:12 Georgia: Seizing the Opportunity to Prosper Over same period, percentage of firms that spend any funds on R&D in the previous three years dropped from 16 percent to 4 percent. Most innovation in developing countries is through technology adoption and adaptation, rather than R&D. Technology adoption in Georgia is ranked 106th in the 2013 Global Competitiveness Index. The Enterprise Survey reported that around 8 percent of Georgian firms adopted at least one internationally-recognized quality certification in 2013, low compared to the regional average of 21 percent and the world average of 17 percent. This ratio is also lower than those reported in previous surveys (16 percent in 2008). In addition, although there is a significant presence of FDI in the country, technology transfer is ranked low (101st) by the Global Competitiveness Index. 16 percent of surveyed firms adopted technology licensed from foreign companies, also below the regional average of 19 percent. According to the 2013 Global Competitive Indicators, Georgia’s intensity of local competition, extent of market dominance and effectiveness of anti-monopoly policy were ranked at the 123rd, 119th and 138th out of 148 countries. Higher competition is likely to increase incentives to innovate among both new entrants and incumbents. Moreover the International Property Rights index ranked Georgia at a low 112th out of 130 countries. In particular, the protection of intellectual property rights, patent protection and copyright protection were ranked relatively low at the 107th, the 109th. these offices help raise funds for research, but they also bring the productive sector closer and more aligned to academia, which in turn, help make higher education more relevant to the needs of the industries. F. Conclusions and Policy Messages While a relatively small proportion of firms in Georgia export, these are significantly larger by comparison to firms that do not, better at creating jobs and are also more productive. The number of firms that are export competitive in Georgia is small and declining. However the positive causal relationship between exporting and employment implies that general measures that enhance firms’ productivity and capacity to export—whether through better trade infrastructure and logistics, greater availability of trade-related financing and insurance, or policies to increase matching between Georgian producers and potential importers—will create more jobs. And although the evidence suggests that increasing exports actually reduces the relative demand for female labor, exporting firms are larger and employ more women than non-exporters, implying a net gain in job creation for women as well as for men. There has been limited improvement of the skills content of Georgia’s exports, especially compared to regional performers. Improvements in skills and value added sophistication is largely concentrated in goods destined for ECA. The skills content of goods being exported to Europe remains relatively low, reflecting the dominance of base metals and minerals in the structure of exports. Further development of national quality infrastructure according to DCFTA requirements will support higher value added exports to the EU market. | Chapter 3: Trade, Labor Outcomes, and Skills   55 Draft 5.indd 73 22/12/14 15:19:12 Georgia Country Economic Memorandum | December 2014 Management capacity, technological upgrades, and staff training could help address this gap. In particular, the skills profile of Georgia’s exports has shown less dynamism than that of regional performers such as Poland or Estonia, which have progressed to a higher value added manufactured exports. The structure of exports mirrors that of the labor market and remains largely primary and traditional. Table 3.4. Main Messages Findings Policy Implications Increasing exports has been beneficial for employment Strong emphasis on general education is critical in a world in Georgia; on the demand side of the labor market, firm in which technologies come and go at a fast pace since these productivity growth is critical. However improving labor allow workers to better adapt. mobility is critical if workers are to reallocate to higher VET quality strengthened, linked to general education, and productivity sectors. ensures that tracking into vocational does not happen too Labor sophistication of exports must improve for Georgia early. to compete in a difficult external environment. Supporting Firm incentives to invest in training; stronger engagement the quality infrastructure and encouraging innovation with private sector on VET and life-long learning. will be critical to encourage the development of more Disseminate detailed information about prospective sophisticated products or products with more value earnings and employment circumstances of different added. careers. Skills mismatches are a significant constraint on firms Introduce instruments of financial assistance to encourage and seem to be hindering innovation even though upgrading existing technology and capacity. Georgian innovators outperform non-innovators to a greater extent than in the rest of ECA and along multiple Further develop national quality infrastructure dimensions. Development of new sectors with potential (accreditation, metrology, standardization and conformity for export competitiveness will be as critical as supporting assessment) according to DCFTA requirements to enable competitiveness of existing sectors and would lead to firms to adhere to international standards and strengthen increased job opportunities for Georgia’s citizens. institutional and technical capacity. Provide services to promote exchange of information (fairs, conferences, exhibition, trainings, etc.). Promote enterprise innovation and technological adoption and adaptation and gear skills development towards younger more innovative firms. Establish/facilitate an environment strong industry- research collaboration to promote innovation. Reallocating resources to export sectors will be important if Georgia is to strengthen the positive labor outcomes associated with trade and develop the trade opportunities offered by enhanced market access under the Association Agreement with the European Union. Attracting FDI and capital investment in the tradables sectors could help Georgia expand its product offerings up the value chain, which could spur dynamism in the export sector, with positive knock-on effects for employment and wages. So far capital flows have been largely in the non-tradables sectors and FDI has been import rather than export oriented as noted elsewhere in this report. Labor mobility is the key component of the resource churning needed to support the growth of productive sectors and facilitate job creation and output expansion. Policies that reduce labor mobility costs would reduce distortions affecting labor supply decisions and thus increase labor market flexibility, enabling workers to adjust more quickly to changing market signals. Labor mobility costs are especially high for entry into the manufacturing sector in Georgia, which suggests that there could be a transition period over which policies targeting manufacturing job growth would have a smaller-than-anticipated effect in the absence of concerted policy attention. For example, unmet labor demand 56 | Chapter 3: Trade, Labor Outcomes, and Skills Draft 5.indd 74 22/12/14 15:19:12 Georgia: Seizing the Opportunity to Prosper in manufacturing could keep wages relatively high, reducing firms’ competitiveness. Policies aimed at reducing this transition period and improving labor mobility over the longer run could be designed as direct targeted compensation (e.g., to finance through cost-sharing relocation or re-training costs, or training subsidies), or could address challenges such as education access and quality, VET, and life-long learning needs. In particular, the Government’s ongoing efforts to develop a consolidated education sector strategy (2014–2024) spanning from early childhood education to higher education promoting the notion of life-long learning and quality education for all, provides an excellent venue to address the above challenges . Over time, this would reduce the skills gaps and mismatches that have been consistently identified as constraints on labor mobility and job growth. | Chapter 3: Trade, Labor Outcomes, and Skills   57 Draft 5.indd 75 22/12/14 15:19:12 Georgia Country Economic Memorandum | December 2014 CHAPTER 4: LOGISTICS AND TRADE FACILITATION Georgia has the potential to be a regional transit hub and has committed to increase its footprint in transit cargo. Georgia’s oil and gas pipelines, Black Sea ports, well-developed railway, East-West Highway, and airports are playing an increasingly important role in linking East and West34. At the same time, Georgia functions as the vertical North-South transportation link between Russia and Turkey and, via Armenia, to Iran. Georgia has made it a strategic priority to position itself as a transit country. The government has enacted a 10-point action plan to become a regional hub, and moved aggressively to improve the border management environment. Moreover, it allowed private sector participation into the ports sector, while enhancing road infrastructure along the East-West Highway. The railway sector was reformed, and its freight services operate on a commercial basis. Furthermore, the commercial civil aviation sector operates under a completely liberal regime. Potential transit flows are large when compared to current volumes handled on the Caucasus Transit Corridor (CTC), especially in non-liquid bulk products. While the overall value proposition of the CTC, under current circumstances, might not be able to match other corridor options—particularly the Iranian ports for non-US and EU originated cargo and the Baltic ports for the rest it has high potential in the event of improved cooperation and better integrated systems along the border with Azerbaijan. For cargo destined or originated in Central Asia, only 22 percent of the potential dry bulk transit cargo and 10 percent of the potential containerized transit is currently transported through Georgia. However corridor level planning and facilitation would be critical to realize this potential. Supply-chain management is in its infancy in Georgia. While Third Party Logistics Services providers exist in Georgia, they do not provide a full range of services as in more mature economies, focusing mainly on transit. A small number of sophisticated companies are vertically integrated to address gaps in service provision in the market. In particular, it would be important to support training and skills upgrading in the logistic and supply chain management sector, directly or indirectly. Another priority is to explore alternative regulatory models for the forwarding sector, conducive to improving quality of service and reducing costs, especially for smaller exporters. Two important objectives for Georgia are to make the country a regional and logistics hub and to upgrade multimodal infrastructure. An analysis of the transit potential and corridor performance suggests an urgent need for international cooperation and for addressing pending regulatory, institutional and infrastructure gaps. It will be necessary to take a “big picture” view of the sector and ensure that public investment is strategic and fully accounts for synergies between different modes. Appropriate coordinating mechanisms are therefore necessary, especially to facilitate multimodal transport planning. The development of a national transport strategy is ongoing and will help strengthen planning and coordination. 34 See the supplemental note for a stock taking of key transit related infrastructure, including institutional set up and planned investments. 58 | Chapter 4: Logistics and Trade Facilitation Draft 5.indd 76 22/12/14 15:19:12 Georgia: Seizing the Opportunity to Prosper The rest of the chapter is organized as follows. Section A reviews transit potential and corridor performance. Section B looks at gaps in infrastructure and services relating to various transit modes. Section C reviews pending challenges to trade facilitation and Section D concludes and presents key policy messages. A. Transit Potential and Corridor Performance Logistics is an important determinant of trading costs worldwide. Reducing the costs of transit and establishing functional and efficient trade corridors are viewed as an important mechanism to facilitate external market access and support firm competitiveness. This typically also entails improved trade and supply chain facilitation as well as investment in infrastructure. Despite large investments in infrastructure in recent years, Georgia’s infrastructure appears to be an important bottleneck to trade. Georgia is ranked 36 out of 138 countries in the 2014 Enabling Trade Index (ETI)35, and 13th for market access, 35th in border administration, 71st for infrastructure, and 48th for operating environment. High cost or delays caused by domestic transportation and international transportation rank Figure 4.1. The Enabling Trade Index 2014— as the seventh and eighth most problematic factors for Georgia’s Infrastructure Ranking trade. When breaking down the infrastructure index, Ease and affordability of shipment 130 the availability and quality of transport infrastructure Logistics competence 115 is ranked much higher (56th) than transport services Tracking and tracing 99 Availability and quality (99 ), with a particularly low ranking for ease and th of transport services 99 Timeliness of shipments 89 affordability of shipments and logistics (Figure 4.1). Air transport infrastructure 79 Port infrastructure 62 Georgia has therefore followed a very dynamic Efficiency of transport mode change 59 policy to develop its transport sector over the last Availability and quality 56 of transport infrastructure eight years. This policy is centered on maximizing Road infrastructure 55 private sector involvement in management and Railroad infrastructure 34 investment in ports and airports, and more strategic 0 20 40 60 80 100 120 140 prioritization of government sector investments, Rank/138 Source: World Economic Forum (2014). including the East-West Highway transit corridor and Note: The Availability and Quality of Transport Services pillar consists of six indicators: ease of and affordability of shipment, logistics competence, tracking and tracing ability, timeliness rural roads. Transport services have been liberalized in of shipments in reaching destination, postal services efficiency, and efficiency of transport mode change. The Availability and Quality of Transport Infrastructure pillar consists of seven the railway sector and Georgian Railways operates as a indicators: available international airline seat km/week, quality of air transport, railroad, ports infrastructure, the liner shipping connectivity index, percentage of paved roads, and quality of roads. semi-autonomous entity (owned by the State directly and indirectly through a Joint Stock Company). 35 In 2008 the World Economic Forum launched the Global Enabling Trade Report which is dedicated to measuring and analyzing the factors enabling trade in national economies around the world. It ranks nations according to factors facilitating the free flow of goods across national borders and to destination. The dataset includes both hard data and survey data from the World Economic Forum’s Executive Opinion Survey. The index is structured to cover four major issue areas: (i) market access; (ii) border administration; (iii) transport and communications infrastructure; and (iv) the business environment. Chapter 4: Logistics and Trade Facilitation   59 | Draft 5.indd 77 22/12/14 15:19:12 Georgia Country Economic Memorandum | December 2014 Transit Potential Located at the crossroads of Europe and Central Asia, Georgia is a transit country connecting several important economic regions with a total population in excess of 800 million people. These include the EU (507 million), the Commonwealth of Independent States (CIS) (277 million), Turkey (76 million) and the Caucasus Region (17 million). The Caucasus Transit Corridor (CTC) is a key transit route between Western Europe and Central Asia for the transportation of oil and gas as well as dry cargo. CTC is part of the international and regional corridor TRACECA; an alternative to the north corridor running through the Russian Federation and Belarus, and the southern corridor running through Turkey and Iran, since the latter cannot handle cargo originated in Europe and United States. By mode of transport, transit represents approximately 60 percent of total tonnages hauled by road, 75 percent of Georgian Railways freight throughput, and almost 80 percent of the volume handled by Georgian ports. Developing trade and transit potential is critical to Georgia’s development. With less than 4 million inhabitants, Georgia’s domestic market is much smaller than that of its geographical neighbors. Turkey, Azerbaijan, Kazakhstan and Turkmenistan altogether represent a market twenty times larger in population, with an average purchasing power two to three times higher than Georgia’s. It is not difficult to understand then, why trade and transit are such key ingredients for Georgia’s development plans. Hinging on its central geographical position, the country can capture some of the trade flows moving West to East, and vice versa, expanding its participation in international value chains with the provision of logistics services. This might include transportation, warehousing, forwarding and even value-added services related to supply chain management. Table 4.1. Transit Potential and Composition of Transit Traffic by Product and Transport Mode Estimated Transit Potential (GNIA, 2013) Composition of Transit Traffic via Georgia (USAID, 2011) Overall Estimated total addressable cargo flows add Based on an analysis of the total flows of transit up to 126 million tons per year. Approximately (2009–10), the top 30 products in transit 55 percent comes from exports from Central accounted for 16 million tons in average. About Asia to Europe (including Eastern Europe, 55 percent corresponds to oil, oil products Balkans and Turkey) and the Rest of the World and coal. Other relevant products that transit (excludes Asia and Middle East); and 28 percent through Georgia are meats and meat products correspond to exports from the Caucasus to ; ethyl alcohol; spirits; construction materials, Europe and the Rest of the World. tractors and construction equipment; sugar; malt extract/components for beer; tropical fruits; apricots, cherries, peaches and sugar. The main origin and destination countries of transit traffic are Kazakhstan, Azerbaijan, Turkey, and Turkmenistan. Liquid cargo Estimated liquid cargo addressable transit Excluding pipelines, approx. 60 percent of non- flows add up to 100 million tons per year. crude petroleum and non-crude petroleum oils Approximately 50 percent of this volume comes are transported by rail and the rest by road. from exports from Central Asia to Europe; and Meanwhile, 90 percent of the crude petroleum 20 percent correspond to exports from the and crude oils are carried by rail. Conversely, Caucasus to Europe mineral fuels seemingly use both rail and road in equal proportions. 60 | Chapter 4: Logistics and Trade Facilitation Draft 5.indd 78 22/12/14 15:19:12 Georgia: Seizing the Opportunity to Prosper Table 4.1. Transit Potential and Composition of Transit Traffic by Product and Transport Mode Estimated Transit Potential (GNIA, 2013) Composition of Transit Traffic via Georgia (USAID, 2011) Dry cargo (bulk) Estimated dry bulk cargo addressable transit The majority of the sugar, chemicals, ceramic flows represent up to 17 million yearly tons. products and fats in transit are transported by Approximately 28 percent comes from exports truck (it is not possible to distinguish if these from Central Asia to the Rest of the World (ROW) are carried in containers or bulk trucks). Same and 25 percent comes from exports from Central situation applies for break bulk cargo, such as Asia to Europe machinery. Containerized Estimated containerized cargo addressable The most important transactions for one-to-one transit flows add up to 8.5 million tons. transport to CIS countries were: Salt exports Approximately 35 percent comes from exports from Turkey to Azerbaijan by rail (114 thou. from the Rest of the World (ROW) to the tons) by rail or truck to Armenia (24 thou. tons); Caucasus and Central Asia and 17 percent from Soap exports from Turkey to Azerbaijan by truck European exports to Central Asia (100 thou. tons); Ceramic products from China to Armenia (27 thou. tons) and meat from the US to Armenia, Azerbaijan and Turkmenistan by truck (38 thou. tons) Source: World Bank. Note: GNIA = Georgia National Investment Agency. Georgia’s participation in potential transit flows is still relatively low, especially for non-liquid transit cargo originated or bound for Central Asia. Some market studies have estimated the potential transit traffic that Georgia could attract, based on actual trade volumes flowing across regions located either side of the Caucasus36. Total addressable flows, which indicate transit potential, are estimated in 125 million yearly tons, of which 80 percent belongs to liquid cargo (overwhelmingly West-bound movements), 13 percent to dry bulk cargo, and 7 percent to containerized cargo. For cargo destined or originated in Central Asia, only 22 percent of the potential dry bulk transit cargo and 10 percent of the potential containerized transit is currently transported through Georgia. Meanwhile, between 60 to 82 percent of the potential dry bulk and containerized transit flows from and to the Caucasus actually choose Georgia as a transit country. Table 4.2. Non-Liquid Bulk Cargo: Estimated Transit Potential and Actual Transit via Georgia Million tons Origin or Destination of Cargo: Caucasus Origin or Destination of Cargo: Central Asia Potential Actual Share Potential Actual Share Transit Transit (Percent) Transit Transit (Percent) Dry-bulk 6 3.6 60 12 2.6 22 Container 2.8 2.3 82 3.9 0.4 10 Total 8.8 5.9 67 15.9 3 19 Source: World Bank based on Georgia’s National Investment Agency (GNIA); based on years 2010/2011. Total also includes as origin or destination of cargo regions such as Europe, China and the Rest of the World. The choice of transport mode for some representative products that transit through Georgia also varies substantially. A more detailed view based on transaction-level data of non-liquid transit cargo moving across the country37 suggests that: 36 GNIA (2013). It should be noted that the methodology used is not readily available in the publication. 37 Based on raw data of top 30 most relevant transit transactions, USAID (2012). Chapter 4: Logistics and Trade Facilitation   61 | Draft 5.indd 79 22/12/14 15:19:13 Georgia Country Economic Memorandum | December 2014 i. The largest transactions involving import of meat products from the US to the Caucasus (and beyond) rely intensively on road transport. Refrigerated transport by rail through Georgia was identified as not viable by logistics operators due to operational constraints. ii. Complex machinery (including heavy lift or out-of-gauge cargo) is typically hauled by road. Amongst other considerations, weight and dimension restrictions seemingly affect the use of the railway for this segment. iii. The use of the railway for transporting products such as sugar and ceramic products is relatively high. High-density low-time sensitive products benefit the most from rail freight economics, especially for longer distances. A deeper analysis of transit products—along with a qualitative understanding of each supply chain—would be beneficial to gauge the potential for transport mode diversion (road to rail) in some specific niches. Corridor Benchmarking Corridor choice involves a set of sequential and parallel decisions made to move consignments from origin to destination, based on an understanding between the cargo owners and freight forwarders that carry an obligation of results to its clients. The final choice is made under uncertainty, based on perceived tradeoffs—mostly between costs and the risk of delays in delivery times. The two main competing corridors of the Caucasus Transit Corridor (CTC) to serve Central Asia are: (i) the Southern Corridor (via Iran); and (ii) the Northern Corridor via the Baltic ports, Russia and Ukraine. A third one, alternative, could be drawn from the Russian ports in the Black Sea, all the way to Central Asia. As a way of enacting some of the options that the supply chain principal has at its disposal when choosing the corridor, a schematic comparison of the main features of the corridors was made in terms of costs and transit time. Containerized Cargo Transit to/from Caucasus Countries Based on a comparison of competing corridors for a hypothetical containerized export from China to Baku, Azerbaijan, three different options were formulated: (i) using the CTC via Poti port in Georgia; (ii) using the Southern Corridor through Bandar Abbas port in Iran, and; (iii) via Istanbul port and inland road transit through Georgia. Judging by the presented evidence, the costs and transit times are the lowest via Bandar Abbas to Baku. Poti represents a second best in terms of cost—but not necessarily in transit times—and Istanbul comes third as far as costs are concerned. However, shippers and consignees not only consider monetary costs and times when choosing among competing corridors. Routing via Bandar Abbas might be the preferred route to minimize costs; although presents diminished visibility, difficulties in securing road transportation during the summer, and less predictable border crossing times. Meanwhile, Istanbul seems to have built a stronger proposition in terms of cargo visibility and reliability, mainly from better road freight services, more predictable border crossing times through Georgia, lower likelihood of unexpected delays at Turkish ports, and enhanced flexibility in denser shipping services from 62 | Chapter 4: Logistics and Trade Facilitation Draft 5.indd 80 22/12/14 15:19:13 Georgia: Seizing the Opportunity to Prosper Asia. On the other hand, Poti might offer an intermediate alternative in terms of cost and reliability, considering its developed inland transport services and predictable border crossing times—except for weather related port delays that might occur during the last and first quarter of the year and informal payments on the Azeri side. Transit to/from Central Asia The scenario presented above compares three corridors (CTC, Novorssysk/Russia and through Baltic countries) and three different origin points to Kazakhstan, highlighting the importance of more integrated border and transit management along the CTC. The three origin points are one US port (East Coast, Baltimore), one major Far East hub port (Shanghai), and the largest maritime gateway in the Mediterranean (Valencia, Spain). There is not much difference in cost between the Baltic ports option and the CTC. Nonetheless, the volatility in shipping costs though Baku, and even more importantly, the lack of predictability in transit times (e.g. delays in Caspian ferry services) represent the biggest hiccup to position CTC as the preferred corridor, especially for more time-sensitive shipments. In the case of the Baltic ports, block train operations offer a single tariff between Klaipeda and Almaty/Dostyk—including all handling charges and costs associated to border clearance. The latter are seemingly expedited en route to ensure that transit times do not exceed eight days. Conversely, block train services between Poti and Baku have not passed the trial stage. Break Bulk Cargo Georgia sits next to one of the largest project logistics markets in the World, mostly heavy lift and out-of- gauge cargo bound for the Caspian oil industry. Yet, Georgia’s proposition as a transit country in this segment faces stiff competition from other corridors. Characteristic of the heavy lift and out-of-gauge segment is the need for customized solutions (the opposite of standardization, as in the case of container shipping). This usually involves complex procedures and special handling and means of transport. In the case of heavy lift cargo from China and East Asia to Eastern Azerbaijan and Central Asia, the waterway system of the Volga-Don seems to be the preferred choice of heavy lift cargo for Azerbaijan and the Caspian, Kazakhstan and Turkmenistan. This is due to shortened waiting times to cross the Caspian, and lower shipping rates to Derince port (Turkey) as compared to Black Sea ports and Bandar Abbas. The only hiccup of the Volga river system route is the limits on vessel size and icing conditions during the winter. The competitiveness of CTC corridor in this segment is also very much dependent on costs set on the Azeri side—and out of Georgia’s control. Overall pricing increases significantly due to transshipment costs from railway to truck and distribution for final delivery within Azerbaijan. As mentioned, a single negotiated rail tariff is not in place between Georgian Railways and Azerbaijan Railways, and hence block trains are not functional from a commercial standpoint between Poti and Baku. Significant delays and reliability issues arise due to the Caspian shipping service providers (CASPAR) ferries availability and alleged ‘sudden’ variations in costs. Hence, Bandar Abbas might be the preferred route for Armenia (year-round), or for the rest of the Caucasus and Central Asia, during the winter season—except for cargo originated in the US and Europe. In that respect, Georgia might be a more competitive proposition for certain trade lanes and months of the year. | Chapter 4: Logistics and Trade Facilitation   63 Draft 5.indd 81 22/12/14 15:19:13 Georgia Country Economic Memorandum | December 2014 Corridor Development Georgia’s government has launched a 10-point action plan to position the country as a logistics hub in the region. In addition to the ‘soft’ reforms that improved the business climate, reduced red tape for trade and decreased clearance times at the border, the country has invested significant resources to modernize its transport infrastructure, as is the case of the East-West Highway. As part of the corridor linking Central Asia with Europe, a significant amount of the volumes handled by its ports, road and rail is transit traffic moving East or West. Up to 60 percent of the road freight traffic transported through Georgia’s borders is transit traffic as is 35 and 65 percent of the containerized and bulk volumes respectively hauled by Georgian Railways. The corridor mainly connects the South Caucasus through Georgia, and the Central Asian countries through Azerbaijan and the Caspian Sea. The perennial issue of informal payments at the Azeri side of land border crossings has not reportedly been completely eradicated, and constitutes a setback for the competitiveness of the corridor. Facilitation payments and rent-seeking behavior of Azeri Customs at Red Bridge BCP are reported to add unpredictability to costs at the border. Further cooperation schemes should be pursued to improve the current situation. Whereas containerized cargo demand has increased in the direction of Central Asia, the corridor still suffers from inefficiencies that can undermine Georgia’s proposition as a viable transit route. As mentioned, the inability to offer rail ‘through’ tariffs between Poti and Baku (and beyond) creates uncertainty amongst shippers and forwarders. Volatile handling costs in Baku’s rail yard and port are a frequent cause of concern for logistics operators. Nonetheless, the unreliability of CASPAR ferry services—that prioritize liquid cargo—is probably one of the most detrimental aspects as far as corridor choice is concerned. In sum, the lack of foothold on the Azeri section of the corridor will need to be addressed somehow if Georgia wants to position itself as a preferred choice for transit. The imbalance in containerized cargo to and from Central Asia (overwhelmingly, laden containers move east) and the mismatch in incentives between shipping lines and traders regarding containers in transit creates further problems. Since shipping lines are wary of releasing their containers beyond Baku, containers have to be either bought, or hefty deposits and detention fees paid for transits bound for Central Asia. The lack of backhaul cargo (directional imbalance of flows) magnifies transportation costs. On the Georgian side, facilitation efforts have largely paid off. Still, mandatory security police escorts are required for out-of-gauge shipments during transit through Georgia. Project cargo constitutes one of the market segments where Georgian logistic operators can capture the most value as compared to regular container transit operations. However, there is no choice of provider and a hefty escort fee is charged. Improving Georgia’s role as a transit country and logistics hub for the Caucasus can also facilitate trade. Although currently not operational, a number of projects have aimed at integrating logistics centers with dry port functionality in the Tbilisi hinterland. If container stays are coordinated with (or managed on behalf of) the shipping lines, one trip to reposition empty boxes from the port can be saved. This can reduce outbound transportation costs, and increase the use of groupage cargo for exports coming from Eastern Georgia, Armenia and Azerbaijan through Georgia. The potential to attract these flows can be boosted if any of these centers can be integrated with airport airside access. Such projects should be granted unimpeded access to basic infrastructure, provided they conform to Georgian regulations. 64 | Chapter 4: Logistics and Trade Facilitation Draft 5.indd 82 22/12/14 15:19:13 Georgia: Seizing the Opportunity to Prosper The potential of the CTC, in terms of additional transit traffic, needs to be understood when compared to alternative transit routes. At present CTC plays a relatively minor role in the movement of non-oil cargo between Central Asia and Europe. Instead, the Baltic Ports are the preferred route for imports and exports between these two regions. One particular advantage of the Baltic route is the short-sea leg between the Baltic ports and European ports.38 A recent study by USAID compared the performance of the CTC with alternative routes and found that if the CTC were to become (a) 10 percent less expensive; (b) transit times reduced by a week; and (c) reliability improved in the crossing the Caspian sea, then the corridor could become a much more attractive and competitive alternative, leading to more transit traffic, contributing to reducing the costs of exporting and importing in Georgia and to GDP growth. The division of freight between roads and rail is also not optimal as studies indicate that rail could carry more freight than it does—at present road transport within CTC moves about twice the number of transit containers as railways. Road transport service prices are competitive due to the significant presence of Turkish truck companies that need to secure backhaul cargo. Establishing the right pricing for the use of roads and railways will be important to ensure that the CTC remains competitive vis-à- vis alternative corridors, and attracts more transit traffic. There is a well-identified need to create a bi-national corridor working group that would be essential to address infrastructure and service weaknesses along the corridor, and negotiate single pricing. This is particularly true in the light of the need for improving the performance of Baku Port and Caspian Sea ferries—but would also be critical for the development of a single pricing mechanism or one stop window for railway pricing along the corridor.39 This would help make the CTC more efficient and competitive, in terms of the services offered. An example is the modern railway container services offered from the Baltic ports to the south towards Odessa and to Kazakhstan—a single price has been negotiated by the companies operating the port and rail services, taking care of all border crossing formalities. Proving such enhanced product offerings goes well beyond investments in infrastructure, but requires a corridor level customer oriented approach and the kind of integrated service provision which has been lacking to date. This requires developing a program of investments for multi-modal freight corridors, focusing on transshipment infrastructure and technologies, in coordination with partners along the corridor. The Scenario Planning methodology might be useful to devise a multi-stakeholder-based strategic action plan for Georgia’s role in the CTC corridor. B. Gaps in Services and Infrastructure Road Transport The road network in Georgia is a main facilitator of transit trade and is often a lifeline for economic activity. The East-West Highway carries over 60 percent of total foreign trade and is seen as a central piece in the government’s strategy of transforming Georgia into a transport and logistics hub. Major seaports will require continued investment in road and rail capacity. For this reason, improving the East-West Highway remains the top 38 USAID (2012), Competitiveness Analysis of the Caucasus Transit Corridor: Improving Transit Potential for Central Asia-Europe Traffic. USAID Economic Prosperity Initiative. September 4,2012. Available at: http://pdf.usaid.gov/pdf_docs/pnadz433.pdf 39 Ibid. | Chapter 4: Logistics and Trade Facilitation   65 Draft 5.indd 83 22/12/14 15:19:13 Georgia Country Economic Memorandum | December 2014 priority, as it provides the fastest and shortest surface transport link between the east and west of Georgia, and because it provides a parallel, alternative route to the rail East-West corridor. The European route E60 is the second longest European road corridor and is critical to Georgia’s international connectivity. It runs from Brest, France (on the Atlantic coast), to Irkeshtam, Kyrgyzstan (on the border with the People’s Republic of China). In Georgia, it runs from the Red Bridge at the Azerbaijan Border to the Poti Port at the Black Sea coast, a distance of about 392 kilometers: this section carries an average traffic of approximately 7,800 vehicles per day and accounts for roughly 23 percent of vehicle utilization on Georgian roads. The upgrading of the East-West Highway to international motorway standards has started with 120 kilometers of the E60 section of the highway have already been upgraded with works on-going or planned on the remaining sections. Georgia still needs to complete the East-West Highway Corridor. Most of the sections have been funded and the remainder is pending a feasibility study. In the longer term, reinforcing the country’s North-South links may bring benefits and strengthen the country’s role as a transit corridor. Over the medium to long-term, it will be important to develop a sustainable financing mechanism to maintain the East-West Highway. There is a need to explore the option of bringing in private sector participation for operation and maintenance once the corridor is completed in 2020, which could be financed by tolling. This could free funds for maintenance of secondary and local roads, which are in poor condition, and are critical if Georgia is to reap the full benefits of the East-West Highway. Government’s expenditures on new construction for international and secondary roads significantly increased from 0.61 percent in 2010 to 1 percent in 2013 and this level of expenditures is expected to be maintained in 2014. Beyond large infrastructure construction on motorways, the sustainability of the national road network remains a challenge. The condition of the road network has improved for international roads, but the rest of the network faces significant challenges. While the condition of 76 per cent of international roads is deemed good or fair, the capacity of international and main roads is inadequate to accommodate high traffic growth. This is primarily due to the insufficient funding of routine and periodic maintenance and to the existence of a significant backlog of repairs, which is reflected in the decline of total expenditures on the international and secondary road network from 3.20 percent in 2010 to 2.29 percent in 2013. Improved regulatory oversight and professional norms are a priority to strengthen services overall. Regulations governing the cross-border road freight services industry in Georgia have gradually converged to the spirit of the EU acquis in regards to professional competence, good repute and financial standing. Nonetheless, only a segment of the market is subject to such regulations, which are applied for specific permits needed to operate internationally (‘multiple permits’). Georgian operators are required to fulfill these criteria in order to access the European Conference of Ministers of Transport (ECMT)40 quota allocation for Georgia (and for instance, the multiple entry permits negotiated with Turkey in 2010/2011). However, no Georgian regulation mandates local operators to comply with access to profession norms when applying for a single international permit. 40 ECMT licences are multilateral licences for the international carriage of goods by road for transport undertakings established in an ECMT Member country, on the basis of a quota system. The licences are not valid for transport operations between a Member country and a third country, though can be used for transit. 66 | Chapter 4: Logistics and Trade Facilitation Draft 5.indd 84 22/12/14 15:19:13 Georgia: Seizing the Opportunity to Prosper Changing regulations will affect the cost of transit. The elimination of all EURO-3 trucks from the ECMT quota starting January 1, 2016, is likely to reduce the market share of Georgian operators on EU-Georgia long-haul market, as well as for transit from or to the EU, thereby pushing up costs of such transit. Approximately 95 percent of the trucks registered in Georgia are EURO-3 and below, of which 23 percent are EURO-3 (although not all are ECMT quota-authorized). Nonetheless, EURO-3 trucks will be able to operate under the bilateral system of permits— mostly single permits—provided that they comply with the provisions of the relevant bilateral agreements. This will likely create a downward pressure on regional transport rates, especially for drayage within Georgia, and for medium-haul operations (to and from Azerbaijan and Armenia). The road freight services industry will continue to play a key role in Georgia’s position as a transit country between East and West. A deeper analysis of the possible developments in relation to market access, enforcement of regulations, and the impact on road freight economics of the changing context is probably needed to devise a strategy for the sector. An industry modernization program that can position the industry to reap the benefits of transit would also entail targeted interventions addressing sustainability issues, capacity building (for operators and the regulator) and regulatory reform. While Georgia has signed bilateral road service agreements with over 20 countries, permit-free regimes were found to be in effect only with Armenia and Kazakhstan. In all other cases all three types of permits (bilateral, transit and 3rd country) are subject to a bilateral quota negotiated regularly, and based upon demand of each of the parties. More can be done in this context to facilitate road freight transit. Railway Transport Georgian Railways (GR) underwent a substantial restructuring of its business model. Wholly owned by a state-controlled Joint Stock Company, Georgian Railways provides freight and passenger services, and owns, manages and operates its infrastructure through three strategic business units. The freight business unit operates at a profit and is entirely commercial-based, setting tariffs and engaging in contractual engagements with cargo owners and forwarders at its own will. Under the current institutional setting, however, operations are not separated from the ownership of the infrastructure. This will remain unchanged until 2022, based on the EU DCFTA market and infrastructure access provisions for rail transport. Moreover, the railway has issued Eurobonds and is in charge of much of the investment to maintain or replace the infrastructure. Georgian Railways has recently established a number of subsidiary firms in upstream and downstream markets, in order to develop ancillary infrastructure and operate a container terminal in Tbilisi. It has also acquired the forwarding firm responsible for over 80 percent of the transported liquid cargo sales. The most frequent impediment voiced by logistics agents seems to be the availability and inadequate quality of the rolling stock. Whereas some of the closed wagons are seemingly in good condition, much of the platform cars are seemingly in obsolete state. Special rolling stock for heavy lift or reefer services not readily available (e.g. platform cars have to be even brought from Uzbekistan for heavy lift operations, and gen-set units are not easily obtained). Moreover, many of the locomotives in GR’s fleet have past their service life. This contributes to lower the speeds in certain sections of the corridor, especially where terrain features require more tractive effort and put a strain on other systems (e.g. braking) that lead to suboptimal operational practices. To this effect, the company has a plan to rehabilitate existing rolling stock and purchase new equipment. Moreover, much of the rail infrastructure upgrading in need seems to be underway. | Chapter 4: Logistics and Trade Facilitation   67 Draft 5.indd 85 22/12/14 15:19:13 Georgia Country Economic Memorandum | December 2014 The harmonization of rail tariffs across the corridor is probably the most pressing need in regards to the commercial side of operations along with pending capacity issues. Automation has improved and also GR’s booking system use. Nonetheless, the lack of a single negotiated tariff is a common source of discouragement to use the corridor in the eastbound direction to Baku and beyond. Whereas this might not be entirely up to Georgian Railways to decide—as it involves other parties, for instance Azeri Railways—the plurilateral initiatives under way (e.g. Silk Wind) might be an opportunity to lock-in some of the benefits that could not be otherwise achieved bilaterally. So far, attempts to operationalize block train operations between Poti and Baku have not led to concrete results. Demand, on the other hand, has been feeble—only 15,000 TEUs/year move in transit through Georgian ports to Azerbaijan and Central Asia. Finally, another concern is that, if left unattended, track capacity bottlenecks west of Tbilisi will affect the performance of the entire network. This might involves not only track infrastructure, but also signaling and ancillary investments in marshalling yards. Rail border crossings need more assessment. The operations at Gardabani-Boyuk Kessik BCP seem to offer opportunities for improvement and cross-border cooperation. Whereas some framework agreement exists with Azerbaijan, and discussions are ongoing to adopt an agreement for the operation of the Silk Wind train, the comparison of the current operations with the agreed principles, and the needs imposed by the future legal basis established by these MOUs is still not known41. More in-depth analysis should be made in this regard. A final consideration should be made in regards to the multiple—and potentially competing—initiatives that the country has adhered to—or plans to—in regards to corridor initiatives involving railway transport  (Kars-Tbilisi-Baku, Silk Wind, Viking Train, etc.). An understanding of what the implications are for the other components of the corridor infrastructure of investments in rail services needs to be explicitly incorporated into corridor planning—as for instance, the potential diversion of traffic away from Poti port when the Kars- Tbilisi-Baku rail service is running. While improving conditions and flexibility in corridor operations is a desired outcome, it is not clear that all initiatives might contribute equally to the development of the Georgian logistics industry and infrastructure assets. Maritime Transport Poti port is the main maritime gateway in Georgia, and a spearhead of the corridors that transit the Caucasus region leading to the Black Sea, with direct rail access and connections to the East-West Highway. The port is managed by a global operator that has committed substantial resources for its expansion (including a new Inland Container Depot (ICD)) after acquiring a majority equity position from foreign institutional investors. Considering the port’s limitations in draft and storage space, operations seem streamlined and make good use of the existing resources. There were no major complaints from the community of logisticians and forwarding firms in regards to its operations, except for alleged delays in berthing time during peak season or bad weather. The main question about the port seems to be related to existing capacity constraints, the pace of throughput growth, and the effect of competing alternatives  (e.g. Kars-Tbilisi-Baku rail link). As far as container throughput growth, the diversion of traffic from Bandar Abbas to Poti brings questions about the 41 A copy of the current legal framework governing the Gardabani BCP rail border crossing was requested to the Transport Policy Department but could not be obtained. 68 | Chapter 4: Logistics and Trade Facilitation Draft 5.indd 86 22/12/14 15:19:13 Georgia: Seizing the Opportunity to Prosper port’s ability to consistently expand at rates of 18 percent (CAGR) as in the 2004–12 period. And, if so, how to accommodate those with current capacity levels. All else being equal, taking the port’s design capacity for containerized cargo handling (500,000 TEU/year) and a growth rate of 6 percent (12 percent) CAGR42, technical capacity would exceed 100 percent by 2020 (2017). In truth, it is believed that Poti’s real technical capacity could be as much as 20 percent lower, increasing concerns about congestion down the road. An intermediate solution underway before considering major infrastructure work is the reconversion of one of the multipurpose terminals to handle containers. Currently, container vessels calling Poti have a maximum capacity of 1,500 TEU, considering the port’s maximum allowed draft of 8.5 meters in berths No.7 and 14. However, the port operator is currently upgrading berth No. 12 (planned 285m in length x 13m depth) to accept Panamax-type container vessels, adding another 200–250k TEU/year of capacity . Under the mentioned growth scenarios, reconverting berth no. 12 would allow the port to extend until 2028 (2020 at 12 CAGR) for its design capacity to be reached. A complete overhaul of the port would require significant investment commitments— not necessarily supported by the region’s current commercial reality. As far as Batumi port, should announced investments materialize, this could substantially increase the capacity of the container terminal. Air Cargo Georgia has implemented a fully liberal policy in regards to market access, removing for the most part all frequency, capacity and airline entry restrictions defined in its air service agreements. As a consequence, passenger traffic almost tripled in the last 6 years, and seat capacity has expanded more than twofold. Nonetheless, air cargo volumes are still modest (approximately 16,000 tons) and exclusively concentrated in Tbilisi International Airport. Despite Georgia’s commitment to liberalization, its overall aviation policy has not been formalized. The potential for air freight development in the country is still largely unrecognized. Moreover, proposals to divest away cargo operations from Tbilisi International Airport to other existing or to a new all-cargo airport have been voiced by media outlets and Georgian stakeholders. Whatever claims to the contrary, such proposals do not seem to have any commercial viability under current circumstances. Tbilisi and Batumi airports were concessioned in the form of a Build Operate and Transfer (BOT) agreement with a Turkish consortium until 2027. With skyrocketing passenger growth rates following market liberalization, the levels of service during peak hours have substantially decreased at the new passenger terminal in Tbilisi Int’l Airport. The new apron space—completely rehabilitated together with the passenger terminal— also poses capacity limitations to accommodate wide-body aircraft. Meanwhile, the old apron adjacent to the cargo terminals cannot handle widebodies, due to pavement specifications. In a growing market like Tbilisi, it is unclear when ramp capacity will start to impose constraints for operations, all else being equal. Accordingly, it is important to understand potential demand for ramp space for different types of carriers, and the windows open for their operations—for passenger and all-cargo carriers alike. Based on this, the rehabilitation of the old apron or the extension of the new one will have to be factored in at due time. The current road usage fee introduces a significant bias against air cargo development, dilapidating Georgia’s advantage as preferred transit airport within a 700–900km catchment area. The road usage fee 42 Container throughput grew at 18 percent CAGR during 2004–12. | Chapter 4: Logistics and Trade Facilitation   69 Draft 5.indd 87 22/12/14 15:19:13 Georgia Country Economic Memorandum | December 2014 (GEL 200/vehicle) is charged irrespective of the type of vehicle used, payload and the distance traveled. Since the distance to the Azeri or Armenian border from Tbilisi (60–70 km) is one third of the distance from Sarpi or Poti (325–385 km), the flat structure of the road fee constitutes a tax on all transit shipments leaving or arriving through Tbilisi airport. Furthermore, other cargo-related cost components to access essential services and infrastructure (e.g. cargo security) should reflect the recovery cost of their provision. In sum, a strategy should be defined to position Tbilisi as the preferred choice within its natural catchment area. C. Trade Facilitation and Supply Chain Management The supply chain management industry is still in its infancy in Georgia. Whereas Third Party Logistics Providers (3PLs)43 are present in the country, they do not offer a full range of services as in other more logistics- mature countries. A reliance on typical transit operations is evident for the most part as far as freight-forwarding is concerned. Even so, the footprint of logistics operators along the corridor is rather limited: in Georgia, sales control in the container market is merely 7 percent (only 7 out of 100 containers entering and leaving Georgia are booked in the country). The most sophisticated segment of the market, comprised by a few firms, shows a higher degree of vertical integration, and offers brokerage, transport and warehousing services. They have also developed services specialized in the project cargo, heavy lift and out of gauge segments, which represent higher margins and value added for the industry. Further training, skills and convergence to best practices is needed. Companies try to manage their logistics assets in house for the most part, and there is a lack of awareness of the possibilities and potential to outsource some of these, or even to enter in joint-ventures. There appear to be incentives for vertical integration in transportation activities, internalizing some of the mismatched incentives along the logistics chain. At a larger scale, joint-ventures between with other logistics service providers in different countries should be encouraged; these might represent a bigger challenge for smaller traders and manufacturers, considering the needed investments. Supporting the development of third party supply chain management providers and logistics platforms may be desirable in Georgia. Different initiatives undertaken by international donors and local stakeholders to further develop the market and human resources (curricula in logistics) should be sustained in time. Access to profession regulations for freight forwarding was eliminated in Georgia and the Revenue Services provides brokerage services. A phased transition to a system where the self-assessment principle in filing customs declarations is restored should be accompanied by further capacity building to develop the forwarding and brokerage business. A deep discussion about the model that suits Georgia’s business needs in necessary between all stakeholders. 43 A third-party logistics provider is a firm that provides service to its customers of outsourced (or “third party”) logistics services for part, or all of their supply chain management functions. Third party logistics providers typically specialize in integrated operation, warehousing and transportation services. If these services go beyond logistics and include services that integrate parts of the supply chain, the provider is called a Third-Party Supply chain Management provider (3PSCM). 70 | Chapter 4: Logistics and Trade Facilitation Draft 5.indd 88 22/12/14 15:19:13 Georgia: Seizing the Opportunity to Prosper Box 4.1. Wine Exports Supply Chain Between 2007 and 2012, Georgia exported US$38 million per year in wine. After hazelnuts and liquid spirits, it is the third most important non-extractive product sold overseas by Georgian producers (approximately 2.6 percent of total exports). The most important destinations are Ukraine, Kazakhstan, Belarus and Poland. Georgian wine is believed to be in the mid and high level quality tier, with over 525 grape varieties grown in the country. Two types of wines are being exported: table wines, of lesser quality; and hi-quality wine (i.e. aged in oak barrels, or other unique production system). Georgia itself requires homologation by National Wine Agency previous to export. A reputed wine exporter mentioned that typical handled quantities range between 180,000–200,000 bottles (0.75L bottles) per month with a total turnover of EUR 8 million per year. With 60 to 90 days credit payment, and quantity/price duration contract for at least a year, the typical cash-to-cash cycle is two months. Approximately 1.5 months of production are kept in stock; both production to order or sales from inventory are a common practice. Order times for inputs from placement of order to receipt of goods at factory vary quite widely: EU—6 weeks, Russia—1 week, and 1 week for Georgia. On the other hand, the order times for products from receipt of order to delivery to buyer is 3 weeks for Russia, 6 weeks for China, and 4 weeks for EU. The biggest market for the interviewed firm is Russia (70 percent), followed by other CIS countries like Ukraine, Kazakhstan, Baltics, Armenia account (20 percent), and Asia, EU and US (10 percent). The Russian market has recently opened to Georgian wine, where they must undergo a homologation process. Peaks months are Sep–Dec, and March–April. Nonetheless, previous to the winter, some stocks are built in Russia due to the closure of the land border. During a typical week, up to 11 trucks are dispatched for international markets, at an approximate value of EUR 25,000 and 12–15,000 bottles per order. For sales to Ukraine and Russia, sales terms are mostly EXW, so the company does not handle the logistics, which become the buyer’s responsibility. For the rest of the CIS, it is believed that sales terms are mostly DAP (carriage to place of destination) so that transportation becomes the company’s responsibility. The company hires a third party provider for transportation. Transportation costs are estimated at US$4,500 per truckload to Russia (up to 16,500 bottles), EUR 3,500 per truckload to Poland, EUR 4,000 per truckload to Germany and US$4,500/FEU to China. According to the company, 90 percent of the shipments arrive to the buyer at the agreed schedule. Nonetheless, the company’s main complaints are Russian formalities and informal payments in Azerbaijan. Inputs are mainly imported, coming from Portugal (cork), Russia, Italy (bottle), Turkey and France (cork, oenology materials); bulk wine, labels and boxes (as well) are locally sourced. Trucking from Russia might cost US$2,000–3,000 and EUR 5,000–6,000 from Europe. Trade facilitation has progressed significantly in Georgia. Whereas many years ago transit through Georgia was a lengthy and expensive proposition (e.g. convoys and mandatory escorts, bureaucratic procedures to issue internal transit document “VVT”) border management reforms since 2004 have contributed to radically streamline processes and bring much needed efficiencies. Thanks to the new model, clearance times in Georgia have been | Chapter 4: Logistics and Trade Facilitation   71 Draft 5.indd 89 22/12/14 15:19:13 Georgia Country Economic Memorandum | December 2014 lowered to a bare minimum, and the number of intervening agencies (for clearance) at the border reduced to one—Georgian Revenue Service. Nonetheless, there are several challenges. These particularly relate to border management and are briefly summarized here:44 i. Current inland transit control practices open up space for malfeasance. The time allotted to transit for transport means is not aligned to international practices, and there is little if any mobile enforcement capacity to prevent and mitigate transit fraud inland. This is, as in many countries, a task entrusted to Customs, and not left to regular police forces. It is also unclear if the penalties associated to transit fraud provide the right framework to deter fraudsters. ii. One of the core responsibilities of border control agencies is to safeguard the common good preventing different types of criminal activity or negligent actions that might endanger the public. Focusing in transit control exclusively from a revenue perspective might steer away from fundamental aspects related to supply chain security. iii. The guarantee system45 in place—or the lack of it—raises concerns with regards to the level of enforcement in transit control and the use of the current automation tools already installed. On one hand, ongoing work is being carried out to improve the electronic applications of TIR46 (EPD and SafeTIR) interfacing with ASYCUDA WORLD47, since TIR is in place for international transits. On the other hand, the Guarantee Management module of ASYCUDA WORLD is seemingly not operational.48 Moreover, transit generated within Georgia (i.e., from the border to inland, or from inland to the border, and possibly from one border to another) is apparently more often covered by the Georgian system than by TIR due to the absence of a mandatory guarantee, and is enforced through seal and time control; x-ray scanning is performed based on selectivity, and truck weight reconciliation when applicable. When Georgia accedes to the EU’s Common Transit system (known as the New Customs Transit System or NCTS for Europe), a guarantee system will have to be implemented. iv. The self-assessment principle of a trader filing their own customs declaration has been lost. Instead, Revenue Service officials enter manually most of Customs declarations in Georgia, based on supporting 44 See Saslavsky (2014) for details on institutional set up and current procedures. 45 While national Customs transit can be arranged by the provisions of respective national Customs law, international transit operations require the negotiation of a bilateral or multilateral agreement. Such an agreement generally sets out the form of the goods declaration for transit, the forms of security required acceptable for each administration, as well as sealing requirements and procedures to secure the integrity of the consignment during transit, including technical specifications for transport equipment to qualify for transport under Customs transit. Key to the success of a transit agreement is the guarantee system, i.e. whether it can satisfy a) the requirements of Customs to cover potential duty liabilities, and b) the requirements of the transport and business sector for acceptable terms and conditions. In cases of increased risk of smuggling, in particular for goods which are subject to high duties or excise taxes such as tobacco products, Customs should refrain from imposing Customs escort arrangements, which are very costly, slow down the transport process and provide an opportunity for corrupt practices. Instead, Customs should either look into the use of modern track and trace systems attached to the transport units which enable proper control of the transport, or specify specific guarantee requirements for the transport of high duty goods. 46 Customs Convention on the International Transport of Goods or TIR is one of the most successful multilateral agreement on international transit. 47 Computerized system for managing customs administration as developed by the United Nations Conference on Trade and Development (UNCTAD). 48 Confirmed by UNCTAD. 72 | Chapter 4: Logistics and Trade Facilitation Draft 5.indd 90 22/12/14 15:19:13 Georgia: Seizing the Opportunity to Prosper documents provided by the trader, which constitutes a potential conflict of interest. This process also breaches the practice of physical and sight separation between traders and Revenue Service officials. v. The fee for customs “clearance of goods”49 should be re-assessed to only reflect the recovery costs of service delivery, as stipulated in the context of the Georgia-EU Association Agreement and DCFTA, and WTO’s Trade Facilitation Agreement. vi. If upstream control measures are undermined, any anti-smuggling enforcement activities downstream lose effectiveness. The activation of the Manifest module in ASYCUDA WORLD—allegedly in sight of RS authorities—would allow an accurate automated acquittal of all goods unloaded at entry points against all Customs and transit declarations; creating a more difficult scenario for goods to ‘leak’ into the domestic economy. vii. Substantial improvements have been made in risk management, including the buildup of risk profiles, the integration of other databases (e.g. personal records, plate readers, etc.) and the extension of a transit risk management module. Nonetheless, the selectivity system needs further improvement. Currently, detection rates hover around 1–3 percent, well below a 10–25 percent desirable target range. Furthermore, an over-reliance on the “in situ” documentary check being carried out in parallel with the filing of Customs declarations undermines the purpose and effectiveness of the Yellow Channel. Post-clearance checks (‘Blue channel’) are allegedly based on documentary re-checks, and not on other internationally accepted forms of in-company audit. viii. Gold list membership requirements do not entail any security auditing. While providing expedited treatment to trusted operators is a recognized practice, the enforcement role of border management agencies cannot be entirely divested to post-clearance documentary checks.. Trusted operators might unknowingly attract unlawful activity eager to take advantage of the expedited treatment granted to them. ix. The process for groupage cargo clearance is not aligned with the development of logistics services in Georgia. Due to the limited bonded warehouse space at the customs clearance zones (GEZIs) and the prohibition for groupage cargo to be directed under customs control to the importers’ premises (except for Gold List members), trucks cannot be released until all importers have filed a declaration and goods cleared, causing frequent delays that add to the overall service costs (especially onerous for smaller exporters/importers). A new customs clearance zone (CCZ “Tbilisi 2”) has been recently inaugurated to specifically handle consolidated shipments. Nonetheless, also procedures should be compatible with modern requirements for consolidation services. In other environments like the EU, Customs only interacts with the consolidator (except under special conditions) that can perform all clearance operations on behalf of consignees or shippers, reducing drastically the number of intervening parties. x. In the opposite side of the map, joint border operations with Turkey—which entail infrastructure and procedural improvements—are expected to relieve some of the congestion common in the access road leading to the border area of Sarpi BCP, where processing capacity and time for clearance substantially differs between both sides. The current linear border design (constrained by the lack of physical space 49 Resolution by the Government of Georgia N96 March 30, 2010, and subsequent modifications: http://www.rs.ge/en/5040 | Chapter 4: Logistics and Trade Facilitation   73 Draft 5.indd 91 22/12/14 15:19:13 Georgia Country Economic Memorandum | December 2014 to introduce more efficient BCP arrangements) does not contribute to relief some of that congestion. The problem with the current first come-first serve linear design is that it matches overall waiting time to the longest process. Hence, some form of queue management will need to be introduced in the future if congestion continues to build up. Low risk traffic could be targeted early on, and separated from the rest into a fast-track lane. The establishment of a monitoring and evaluation system that benchmarks current processes and allows assessment of services, enforcement and strategic planning would be desirable. D. Conclusions and Policy Messages Considering its population base, income per capita, and geographical position, Georgia has made it a strategic priority to position itself as a transit country and to enhance its participation in global value chains by providing transport and infrastructure services. The Government remains committed to strengthening Georgia’s capacity to become a regional hub, and to improve the border management environment. Moreover, it allowed private sector participation into the ports sector, while enhancing road infrastructure along the East-West Highway. The railway sector was reformed, and its freight services operate on a commercial basis. Furthermore, the commercial civil aviation sector operates under a completely liberal regime. Potential transit flows are large when compared to current volumes handled on the CTC. The overall value proposition of the CTC however, under current circumstances, might not be able to match other corridor options. In general, the CTC does not seem to be able to compete merely on a cost basis with other options, especially Iranian ports for shipments to the Caspian countries that are not originated in the EU or in the United States. This might change based on the requirements of specific supply chains that need enhanced reliability and visibility. In those cases, the CTC might represent an intermediate solution to Turkish ports. For European and US-originated consignments, Baltic ports would naturally represent a more competitive choice for shippers, given the seamless block train services, especially to Kazakhstan. In the case of heavy lift cargo, the waterway system of the Volga-Don seems to be the preferred choice during the ice-free navigation season. The competitiveness of CTC corridor in this segment is also very much dependent on costs which remain out of Georgia’s control. Overall pricing increases significantly due to transshipment costs from railway to truck and distribution for final delivery within Azerbaijan. As mentioned, a single negotiated rail tariff is not in place between Georgian Railways and Azerbaijan Railways, and hence block trains are not functional from a commercial standpoint between Poti and Baku. Significant delays and reliability issues arise due to the CASPAR ferries availability and alleged ‘sudden’ variations in costs. In addition to this, mismatched incentives add to the costs of the corridor to serve Central Asia, due to hefty deposits and detention fees paid for containers, and the lack of backhaul cargo. Moreover, for out-of-gauge cargo hefty security escorts fees are charged. International cooperation remains critical for the development of the CTC. Considering that some of the avoidable costs and delays do occur outside Georgia’s borders, engaging other partners along the corridor— namely Azerbaijan—will be critical for enhancing the efficiency of the CTC. Regional cooperation through TRACECA and on a bilateral basis could be useful channels. Creating a bi-national corridor working group could set a desirable environment for deeper cooperation. However, a corridor-level customer oriented approach and 74 | Chapter 4: Logistics and Trade Facilitation Draft 5.indd 92 22/12/14 15:19:13 Georgia: Seizing the Opportunity to Prosper integrated service provision will be necessary (lacking to date) including the development of an investments program for multi-modal freight corridors. The Scenario Planning methodology might be useful to devise a multi- stakeholder-based strategic action plan for Georgia’s role in the CTC corridor. Table 4.3. Main Policy Messages Border Management •• Restore self-assessment principle in lodging customs declarations. •• Align fee structure with recovery costs for customs clearance services (filing customs declarations) provided by the Georgian Revenue Service. •• Revise allowed transit times through Georgia for means of transport. •• Enhance domestic transit system and guarantee management functionalities in ASYCUDA WORLD. •• Revise legislation to provide Revenue Service (RS) with competencies to perform inland transit controls and enhance RS mobile control units. •• Enhance upstream control measures such as carrier direct submission of Manifest data into ASYCUDA WORLD (Manifest Module). •• Enhance risk management systems. •• Improve post-clearance checks and procedures beyond documentary checks (Blue channel). •• Enhance audit procedures of Gold List member firms, including supply chain security audits. •• Facilitate procedures for groupage cargo to avoid delays in discharging trucks from the GEZIs. •• Continue cooperation efforts with Turkey for joint border operations and explore traffic management schemes for Sarpi BCP. •• Enhance institutional capacity and strategic planning of Georgian Revenue Service; consider using benchmarking tools. Supply Chain Management •• Support training and skills upgrading in the logistic and supply chain management sector, directly or indirectly, with other international donors. •• Continue support of advocacy groups and efforts to build a supply chain management curricula with local universities and/or partnerships with foreign technical institutions •• Explore alternative regulatory models for the forwarding sector, conducive to improve quality of service (access to profession, professional competence); including a capacity building program Corridor Development •• Strengthen cooperation and institutional framework with Azerbaijan: DD Facilitate border management and ensure transparency of costs and procedures. DD Operationalize block train from Poti to Baku and beyond; including a transparent single negotiated rail tariff (ADDY) and accessorials. DD Deepen cooperation with CASPAR to address charges transparency and schedule reliability for ferry services. DD Operationalize bi-national Corridor Working Groups. •• Allow competition in the escort services fee for out of gauge transit cargo, or set fee on a recovery cost basis. •• Promote pilots of joint-ventures between shippers, forwarders and (alternative) companies providing shipping services in the Caspian. •• Promote establishment of logistics centers with dry port functionality operating on commercial basis. •• Assess incentive scheme for road transit vis-à-vis rail transport, including road usage fee. •• Continue regional cooperation through TRACECA. •• Boost strategic planning at the corridor level; consider using scenario planning methodologies. | Chapter 4: Logistics and Trade Facilitation   75 Draft 5.indd 93 22/12/14 15:19:13 Georgia Country Economic Memorandum | December 2014 Table 4.3. Main Policy Messages Road Infrastructure and Road Freight Services •• Complete remaining sections of the East-West Highway Corridor (already under way). •• Explore different financing mechanisms and future private sector participation for operation and maintenance after completion of the E-W Highway (e.g. tolling). •• Improve secondary and local roads network to help reap full benefits of investments in the East-West Highway. •• Explore role of North-South road links to enhance Georgia’s transit role. •• Devise a strategy for the road freight services sector that would address, inter alia: DD Market access issues DD Access to profession regulations DD Capacity building and sustainability aspects Railway Transport •• Support Georgian Railways’ efforts to separate operations from the ownership of the infrastructure by 2022. •• Upgrade and expand rolling stock. •• Solve track capacity bottlenecks West of Tbilisi (e.g. Poti–Samtredia and Zestaponi-Khashuri) within ongoing rail network modernization efforts. •• Assess existing rail border crossing agreements vis-à-vis current practices, and implications of new service initiatives (e.g. Silk Wind). •• Incorporate future railway initiatives in the context of corridor planning and its impact on other logistics and transport infrastructure assets in the country. Maritime Transport •• Concentrate efforts to upgrade existing port infrastructure. Air Cargo •• Concentrate efforts to promote Tbilisi International Airport as sole cargo gateway. DD Assess apron space needs for cargo operations, and feasibility of enlargement/upgrading of the new and old apron. DD Revise road usage fee structure for transit air cargo in/out of Tbilisi International Airport to support development of Tbilisi as a preferred air transit route. DD Assess charges for essential services and infrastructure use (cargo security) in relation to its recovery cost. 76 | Chapter 4: Logistics and Trade Facilitation Draft 5.indd 94 22/12/14 15:19:13 Georgia: Seizing the Opportunity to Prosper  ONSTRAINTS ON FIRM COMPETITIVENESS: CHAPTER 5: C BEYOND TBILISI Georgia’s regions have benefited unequally from last decade’s rapid economic growth and an intensification of international trade, yet Georgian regions have failed to benefit. Solid growth rates and evolving trade patterns—and the development of large trade deficits—have not been accompanied by regional convergence in terms of either income or productivity. A case in point is Tbilisi, the capital, which contains approximately one third of the national population and accounts for half of the country’s GDP. The city-region’s per capita output levels are almost twice the national average and more than three times that of the most lagging regions. The large internal growth differences that have taken place—largely benefitting capital city-regions50—are, not unique to Georgia and are in common with other transition economies in Eastern Europe. The globalization process offers unprecedented opportunities for firms to grow and regions to prosper; but heightens the risk that some firms and regions will be left lagging behind. Greater integration into global markets, in conjunction with dramatic advances in information technologies, has created significant competitive pressures for firms to adapt to. While the economic literature stresses the importance of trade for growth, the impact of global integration on regions in the presence of spatial inequalities is especially relevant for Georgia. Whether greater trade openness increases or lessens regional disparities depends largely on differences in terms of local endowments, accessibility to trade, and local government capacity51. Figure 5.1. Poverty Differentials by Region Proportion below 60% of median consumption Racha-Lechkhumi-Kvemo Svaneti Samegrelo-Zemo Svaneti Mtskheta- Mtianeti Imereti Shida Kartli Guria Adjara Kakheti Samtskhe- Tbilisi Javakheti Kvemo Kartli JJ0.1319–0.1319 JJ0.1319–0.1916 JJ0.1916–0.2531 JJ0.2531–0.3381 JJ0.3381–0.3669 Source: Hardy and Rodriguez-Pose 2014. 50 Farole et al. 2011. 51 Williamson 2005; Brülhart 2011; Rodríguez-Pose 2012. Chapter 5: Constraints on Firm Competitiveness: Beyond Tbilisi   77| Draft 5.indd 95 22/12/14 15:19:13 Georgia Country Economic Memorandum | December 2014 Georgia is set to embark upon an ambitious strategy for decentralization and the strengthening of government at the regional level, a complex agenda that brings potential benefits but also entails economic risks. The aims of the decentralization reforms include increasing democracy, transparency, and efficiency in the delivery of public goods and services. While improvements in these areas will certainly contribute to strengthening local institutional environments, with a potentially beneficial impact on firm competitiveness, the transfer of powers and resources to sub-national tiers of government also entails risks related to varying capacities to implement policies that may exacerbate existing regional gaps. The questions that arise are about the impact of regional characteristics vis-à-vis firm competitiveness, and how the decentralization process will impact upon existing disparities. Firstly, how do local conditions affect the competitiveness of Georgian firms in different ways in different parts of the country? Secondly, how is the on-going process of decentralization likely to increase or reduce any potential differences in the competitiveness of firms and for the development prospects of the region as a whole? To address these questions, we directly analyze regional differences in firm competitiveness throughout Georgia by analyzing a range of firm characteristics and region-specific factors. We also examine, in light of any existing disparities, how the process of decentralization may contribute to reduce or, by contrast, exacerbate existing disparities. Firm specific characteristics matter more than location or place-specific effects. Local public expenditures, transport infrastructure, and human capital endowments are particularly important for the competitiveness of Georgian firms. As global trade continues to intensify—especially in view of the DCFTA with the European Union (EU)—building capacities in less favored areas is imperative to avoid causing further harm to lagging regions, and to enable more firms to raise their competitiveness, grow, and benefit from scale economies. While the nature of Georgia’s firms, with a large proportion of exports and value added being derived from low-productivity sectors, affect the strength of the findings, the impact of improved public services in the regions is only likely to strengthen as firms evolve to levels that enable better exploitation of the positive externalities associated with knowledge spillovers and trade opportunities. International evidence on the impact of decentralization is mixed, especially given the complexity of the process. Decentralization in emerging countries has frequently led to efficiency losses due to agency problems— linked to the separation across tiers of government taking expenditure decisions and those responsible for collecting taxes—and increased territorial competition52. Decentralization has also been linked to an increase in territorial disparities, which depend on initial conditions. At the same time, however, decentralization has been associated with lower poverty, improved voice, and strengthened public services. The rest of this chapter is organized as follows. Section A reviews regional inequalities. Section B reports the results of the assessment of the extent to which the capacity of Georgian firms to compete is affected by their location (i.e. place-based effects) or by the characteristics of each firm (i.e. sorting and compositional effects), or of a combination of the two. Section C examines the evidence on decentralization and spatial disparities and Section D concludes. 52 Prud’homme 1995; Rodríguez-Pose and Gill 2005. 78 | Chapter 5: Constraints on Firm Competitiveness: Beyond Tbilisi Draft 5.indd 96 22/12/14 15:19:13 Georgia: Seizing the Opportunity to Prosper A. Growth in Georgia: Sub-National Trends Georgia has fostered a strong basis for economic growth over the past ten years. During the last decade the country has achieved impressive GDP growth rates, peaking at an annual rate of over 12 percent in 2007, before a triumvirate of crises’, including mass-demonstrations in 2007, the global financial crisis in mid-2008, and the 2008 August conflict with Russia, brought this to an end. However, Georgia has recovered quickly following this recent economic and political turbulence, resuming solid growth apart from a temporary slow-down in 2013. Increases in Georgian output have moved in parallel with increases in external trade and widening territorial disparities. Over the last decade, the value of imports and exports has escalated, averaging above US$600 million and US$200 million per month in 2013, respectively. However over this period of growth and growing trade, spatial inequalities have become persistent across Georgian regions, and do not show notable improvement.53 Table 5.1. Convergence of Regional GNI, 2006–12 Average across all regions = 100i 2006 2007 2008 2009 2010 2011 2012 TB 187.14 191.39 184.59 180.46 181.04 181.73 190.77 SJ 71.53 76.74 85.37 87.40 87.77 88.47 85.05 GU 88.38 82.84 73.38 76.45 75.75 75.44 74.95 IM_RK 62.81 67.32 72.66 76.79 73.26 72.32 70.47 KK 64.43 62.99 62.24 61.14 66.05 67.34 62.98 SZ 81.77 76.59 62.29 62.69 67.01 66.46 61.88 AD 71.19 67.05 67.22 73.28 70.59 67.42 61.63 KA 74.47 65.64 67.23 64.69 65.57 66.58 60.29 SK_MM 63.89 58.82 64.76 58.57 59.26 60.31 57.01 Standard deviation 0.339 0.355 0.342 0.344 0.334 0.335 0.374 Source: Hardy and Rodriguez-Pose, 2014. Note: Ranked by 2012 GDP index values. i = Based on Enterprise Survey data provided by GeoStat. The magnitude and persistence of regional gross national income inequalities are underscored by the supremacy of the capital, as in many other ECA countries. Tbilisi contains approximately one third of the national population and accounts for half of Georgian GDP. The city-region’s per capita output levels are almost twice the national average and more than three times that of the most lagging regions. The data also suggests that regional inequalities are highly persistent and have tended to expand in periods of economic growth. In other transition economies in Eastern Europe as well, especially in small economies such as Slovakia, Romania, and Estonia, big internal growth differences have taken place, largely benefitting capital city-regions54. The characteristics of Georgia’s regions vary considerably, in line with income disparities. Tbilisi is dominated by the services sector. Kakheti, by contrast, is much more dependent on primary sectors—agriculture—and agri- 53 Based on Enterprise Survey data provided by GeoStat. 54 Farole et al. 2011, Tondl and Vuksic 2003; Farole, Rodríguez-Pose, and Storper 2011, Banerjee and Jarmuzek 2009, Altomonte and Colantone 2008, and Tatar 2010, reveal similar patterns. | Chapter 5: Constraints on Firm Competitiveness: Beyond Tbilisi   79 Draft 5.indd 97 22/12/14 15:19:13 Georgia Country Economic Memorandum | December 2014 business (mainly wine). Kvemo Kartli, and to a lesser extent Shida Kartli and Mtskheta-Mtianeti, are much more orientated towards industrial activities, likely owing to their proximity to the largest market, the capital region. In terms of productivity, the most productive regions, and those that have ascended the national income hierarchy, are those with large services (Tbilisi and Adjara) and industrial (Kvemo Kartli) sectors. The poorest (and most unproductive) are characteristically agrarian. The data also reveal a distinctive East-West divide, which, with the marked exception of Tbilisi, favors the regions in closer proximity to Georgia’s coastline. The presence of significant minority population groups and high incidence of poverty may have significant effects on the productivity of individuals and firms. Minority groups may be socially, economically or politically marginalized, and face greater barriers with respect to their access to the labor market and in the acquisition of skills. In Georgia, there are significant minority groups in the South and Western regions, those bordering Azerbaijan and Armenia and particularly the Samtskhe-Javakheti (53 percent of Azerbaijani nationality) and Kvemo Kartli (51 percent of Armenian nationality) regions. The incidence of poverty, measured in terms of the proportion of the population below 60 percent of median consumption, exhibits a relatively strong Eastern bias, with the exception of Tbilisi and Guria, reaching up to 36 percent of the population in the Kakheti region. Poverty picks out some of the least-productive regions of Georgia.55 Table 5.2. Regional GVA Share by Sector, 2012 In percent Agriculture Industry Construction Services Public Administration Tbilisi 0 (0) 12.8 (-0.9) 11.6 (0.4) 59.3 (-2.5) 16.2 (3.1) Adjara 7.1 (-5.7) 11.3 (0.8) 11.9 (-1.0) 44.3 (4.9) 25.4 (1.0) Samegrelo Zemo Svaneti 19.6 (-4.1) 13.6 (4.0) 3.8 (-3.1) 38.4 (0.7) 24.5 (2.5) Imereti_Racha-Lechkhumi Kvemo Kartli 13.2 (-11.5) 17.1 (5.1) 2.0 (-0.5) 35.3 (5.2) 32.4 (1.8) Guria 27.3 (-4.7) 11.0 (-5.7) 1.8 (-0.9) 31.0 (8.7) 28.9 (2.6) Kakheti 24.7 (-5.3) 17.2 (2.0) 2.1 (-0.2) 26.6 (2.7) 29.4 (0.8) Shida-Kartli_Mtksheta Mtianeti 17.9 (-7.1) 27.7 (3.4) 3.0 (-6.9) 21.2 (1.6) 30.2 (8.9) Samtskhe-Javakheti 29.2 (-0.5) 10.4 (-6.4) 4.5 (3.9) 19.7 (-2.8) 36.3 (5.8) Kvemo-Kartli 17.7 (-5.1) 41.1 (-1.9) 2.3 (0.9) 17.5 (-0.4) 21.4 (6.5) Source: Hardy and Rodriguez-Pose, 2014. The competitiveness of places, and their firms, is central to regional economic development and a key concern for policy. The firm, and its ability to compete, plays a central role in explaining the differential economic performance of regions and countries. Global trade, by increasing competitive pressures, can demand greater efficiency from firms, and make them increase investment in efficiency-boosting areas such as R&D and new technologies in order to rise to global challenges. Naturally, firm- and industry- specific characteristics explain much of a firm’s ability to survive, meet global competitiveness challenges, and increase export performance. It is becoming increasingly evident, however, that the characteristics of regions also play a significant role in facilitating firm competitiveness and economic development, and offer considerable scope for policy. By mapping firm competitiveness we can clearly see that there is a distinct geography to firm competitiveness in Georgia, matched closely by capital intensity of production. Interpreting competitiveness as simply ‘a 55 Percent change since 2006 in brackets. 80 | Chapter 5: Constraints on Firm Competitiveness: Beyond Tbilisi Draft 5.indd 98 22/12/14 15:19:13 Georgia: Seizing the Opportunity to Prosper Figure 5.2. Labor Productivity in 2012 GVA per worker in GEL Racha-Lechkhumi-Kvemo Svaneti Samegrelo-Zemo Svaneti Mtskheta- Mtianeti Imereti Shida Kartli Guria Adjara Kakheti Samtskhe- Tbilisi Javakheti Kvemo Kartli JJ5,385.16 JJ5,385.16–9,061.20 JJ9,061.20–11,408.04 JJ11,408.04–17,384.91 JJ17,384.91–27,427.89 Source: Hardy and Rodriguez-Pose 2014. Figure 5.3. Capital Intensity Pattern is Similar that of Productivity Capital per worker in GEL Racha-Lechkhumi-Kvemo Svaneti Samegrelo-Zemo Svaneti Mtskheta- Mtianeti Imereti Shida Kartli Guria Adjara Kakheti Samtskhe- Tbilisi Javakheti Kvemo Kartli JJ734.6875–734.6875 JJ734.6875–1,422.0013 JJ1,422.0013–1,919.2114 JJ1,919.2114–2,993.7359 JJ2,993.7359–4,335.4360 Source: Hardy and Rodriguez-Pose 2014. poetic way of saying productivity’56, a measure of labor productivity, namely gross value-added (GVA) per worker, can obtained with data from the Industrial Census. The difference between the productivity performance of Tbilisi and the small, lagging coastal region of Guria is considerable, with the former achieving productivity levels five times that of the latter. Although large firms are generally the most productive, on average, the differences across firm size are relatively small, with the exception of those located in Shida Kartli and Mtskheta-Mtianeti. For the most productive regions, including Tbilisi, Adjara, and Kvemo Kartli, it is, surprisingly, small- and medium- sized firms that are the most productive.57 The geography of the average capital intensity (Figure 5.3) of Georgian firms follows a similar pattern to that of productivity (Figure 5.2). Ideally, several other firm-level factors, such as 56 Krugman 1997. 57 As a robustness measure, we create additional measures of firm performance. First, we calculate TFP for a sub-set of manufacturing firms. We follow Levinsohn and Petrin (2013), using intermediate inputs such as materials and energy expenditures as instruments to overcome bias evident in traditional productivity estimates. We are unable to compute common alternatives, such as the Olley-Pakes method, due to very limited data on firm investments. Secondly we compare the results with simple measures of firm competitiveness, profit per worker and market share, to check the validity of our results. In the case of these simple measures, the results largely corroborate the main labor productivity results. Chapter 5: Constraints on Firm Competitiveness: Beyond Tbilisi   81 | Draft 5.indd 99 22/12/14 15:19:14 Georgia Country Economic Memorandum | December 2014 age and export activity, among others, would have been welcome. Nevertheless, the assembled set of firm-level characteristics represents the best set available from the available industry census data. Table 5.3. Labor Productivity in 2012 GVA per worker in GEL by firm size All Micro Small Medium Large Tbilisi 27,427.89 29,768.96 26,789.13 17,480.57 19,014.50 Adjara 17,384.91 16,602.88 20,837.37 13,463.63 12,067.78 Kvemo Kartli 13,623.20 12,387.34 13,456.18 27,315.59 19,476.06 Shida Kartli & Mtskheta-Mtianeti 11,408.04 8,955.16 16,807.26 18,183.53 46,022.65 Samagrelo-Zemo Svaneti 11,030.35 10,258.57 11,649.84 15,638.25 25,033.61 Kakheti 9,061.20 6,787.73 12,624.84 31,902.82 12,960.49 Samtskhe-Javakheti 8,027.00 6,568.94 11,091.19 16,158.18 Imereti, Racha-Lechkhumi & 7,696.49 6,935.76 9,612.47 13,772.94 7,859.06 Kvemo Svaneti Guria 5,385.16 5,922.31 3,718.93 2,383.36 Source: Hardy and Rodriguez-Pose (2014). Note: Sizes are defined as follows: Micro (<10 employees), Small (10–49 employees), Medium (51–250) and Large (>250). The magnitudes of differences in unemployment, skills and labor productivity across regions are wider than those in average incomes. Although Tbilisi is clearly a key center of skilled employment and high incomes, the proportion of highly educated inhabitants is the country’s lowest (although nominally highest) and the unemployment rate is more than double that of the other regions in Georgia. This observation is no doubt partly attributable to migrant inflows, but may also be affected by the nature of unemployment statistics in largely agrarian economies. As self-sufficient farming is recorded as self-employment, the heads of households engaged in agriculture are recorded as ‘individual entrepreneurs’, and family members are recorded as ‘unpaid family business workers’, unemployment may be underestimated in such areas. Differences in regional patterns with respect to incomes, unemployment and average labor productivity levels suggest that differences in regional labor market conditions (labor demand and impediments to the flows of labor and capital) shape the variation we see in the ranking of regions. Figure 5.4. Labor Market Conditions and Constraints on Mobility May Explain Variations in Regional Patterns Human capital, proportion with post-secondary and higher education Unemployment, proportion unemployed Racha-Lechkhumi- Racha-Lechkhumi- Kvemo Svaneti Kvemo Svaneti Samegrelo- Mtskheta- Samegrelo- Mtskheta- Zemo Svaneti Shida Mtianeti Zemo Svaneti Shida Mtianeti Imereti Imereti Guria Kartli Guria Kartli Adjara Samtskhe- Kakheti Adjara Samtskhe- Kakheti Javakheti Kvemo Tbilisi Javakheti Kvemo Tbilisi Kartli Kartli JJ0.2340–0.2340 JJ0.2340–0.4027 JJ0.4027–0.4236 JJ0.4236–0.4395 JJ0.4395–0.5421 JJ0.0512–0.0512 JJ0.0512–0.0698 JJ0.0698–0.1259 JJ0.1259–0.1675 JJ0.1675–0.2992 Source: Hardy and Rodriguez-Pose 2014. 82 | Chapter 5: Constraints on Firm Competitiveness: Beyond Tbilisi Draft 5.indd 100 22/12/14 15:19:14 Georgia: Seizing the Opportunity to Prosper B. Firm Competitiveness and Location The 2009 World Development Report emphasizes a range of structural, place-specific and ‘softer’ government and institutional factors that are instrumental in shaping the evolution of disparities between regions. The range of regional attributes that can potentially influence the competitiveness of firms can be broadly decomposed into two categories, pure location factors (so-called first-nature geography) and second-nature geography, including agglomeration economies58, endowments relating to the local labor market, the business and investment climate, and the institutional setting and policy context. For a country like Georgia, with a dominant urban core, regions in closer proximity to this leading node may benefit from important inter- regional externalities and spill-overs, which must also be reflected in an empirical analysis. In the analysis, firm size (turnover), ownership type (public or private), capital intensity, and female employment share are included as controls for the firm-specific characteristics that affect firm competitiveness. We draw on a wide range of variables to address the question of how and which region-specific factors drive and interact with firm and market characteristics to influence competiveness. The data utilized covers a total of nine Georgian regions59 over the period 2006–12 and relies on a number of sources, including GeoStat’s household surveys and industrial census. Firm competitiveness is measured as gross value added, but robustness checks utilized total factor productivity and confirmed that the analysis remains valid. Box 5.1 summarizes the variables associated with firm competitiveness in the empirical models. Variance in labor productivity is driven by spatial, sectoral and firm-specific factors60. Using a multi-level modelling framework, structured with region and sector level components, the analysis indicates that 5 percent of the variability in firm labor productivity is attributable to region-level factors, 13 percent to sector-specific factors and the remainder, 82 percent, is a function of firm-level differences. If Georgia is to continue to grow and become a larger player in global trade, it will have to maximize the performance of firms in all areas. The primary firm-level findings show that bigger firms, higher capital intensities and private-sector ownership are highly positively associated with competitiveness, while the female labor force share is negatively correlated with competitiveness.61 As expected, larger firms with more capital intensive production are more competitive. Female participation rates seem to have a negative correlation with competitiveness. This may reflect relative differences in gender composition between manufacturing and service sector firms, where productivity tends to be higher in the former, and also within sectors, such as a potential sectoral bias in manufacturing sectors, where female participation tends to be highest in more labor intensive activities (e.g. apparel) and at the lower end of the value chain. 58 The qualities of specific places, such as the benefits of agglomeration and the availability of a critical mass of skilled workers and knowledge- intensive activities feature heavily in contemporary research on firm competitiveness (Malmberg et al 2000). 59 The 9 regions include Tbilisi (1), Adjara (2), Guria (3), Imereti, Racha-Lechkhumi and Kvemo Svaneti (4), Kakheti (5), Shida Kartli and Mtskheta- Mtianeti (6), Samagrelo-Zemo Svaneti (7), Samtskhe-Javakheti (8) and Kvemo Kartli (9). Regions 4 and 6 are each aggregated from two administrative divisions due to data constraints and following the methodology of the Georgian Statistics Office. The Autonomous Republic of Abkhazia is excluded from the analysis due to no data availability. 60 To give a sense of the relative magnitude of the determinants of firm performance, we first specify a multilevel model that provides insights into the drivers of labor productivity. The constant term is significant because it is an `empty’ hierarchical model and only contains a constant term. It is the variance components that are used to estimate how different levels are associated with labor productivity, as presented in Table 5.4. 61 Regression results are not reported here but are available on request from the authors. | Chapter 5: Constraints on Firm Competitiveness: Beyond Tbilisi   83 Draft 5.indd 101 22/12/14 15:19:14 Georgia Country Economic Memorandum | December 2014 Box 5.1. The Drivers of Regional Differences First-Nature Geography Two measures of first-nature geography are: (i) the ‘ruggedness’ index based on the coefficient of variation of elevation in each “grid” in Georgia’s map, which is a proxy for accessibility and transportation costs; and (ii) climate proxied by average July rainfall. As the values of these indicators are implemented as time- invariant regional factors, random effects specifications are necessary to measure their associations with firm performance. Second-Nature Geography Agglomeration economies are expected to favorably impact firm capacity to compete by lowering costs though the sharing of infrastructure, facilitating logistics/transit, increasing interactions between suppliers and customers and providing a critical mass of complementary services, skilled human capital and competitive rivals to engender greater levels of performance. Externalities from specialization, also known as localization economies, are measured as the level of own- industry regional specialization for each firm (i.e. the total employment in firms in sector j in region i as a percent of total employment in sector j). We use a relative measure that weights each of these shares by the national average for each sector. To assess urbanization economies, we employ indicators of density and diversity. Population density accounts for interactions between people and firms within urban areas that support efficiency gains and knowledge spill-overs. We compare this measure with alternatives such as road density, to give an indication of the level of regional infrastructure. Policy makers view infrastructure (such as roads, power, communication and other utilities) as a core component of development, where an adequate In a small country like Georgia, it appears that firm competitiveness, and likely the performance of the regional economy, benefit from agglomeration—and principally localization—economies. Several sets of regional characteristics are examined for their association and potential influence on the competitiveness of local firms62. Localization economies are found to be important suggesting clustering around a pool of knowledge, skills and infrastructure generates spill-overs that are beneficial for productivity growth. Each measure of agglomeration is found to be important when considered individually though neither diversity nor density matter when considered together with localization. When accounting for spatial spill-overs, it seems there are constraints on taking advantage of neighboring clusters. This corresponding to the 2009 World Development Report narrative, signifying that the three dimensions of density, distance, and division are fundamental for development and to firm competitiveness in Georgia. Investment in transit infrastructure that can reduce the cost and time of travel could be one solution. 62 Furthermore, in light of the unipolar structure of the Georgian economy, with Tbilisi at its economic center, we trailed a dummy variable to mediate the disproportionate effect this region may have on the results. However, the inclusion of the dummy variable did not alter the main results, and is omitted from the analysis. 84 | Chapter 5: Constraints on Firm Competitiveness: Beyond Tbilisi Draft 5.indd 102 22/12/14 15:19:14 Georgia: Seizing the Opportunity to Prosper quantity and quality are critical for competitiveness and international trade (World Bank 1994). Better endowments of infrastructure lower transactions, communications, and transport costs and ease market access. Both measures are highly correlated, so we retain population density to exploit its time variability. The diversity index measures the sectoral mix of the regional economy and achieves a minimum value of 0 when total regional employment in concentrated in a single sector and increases with higher levels of economic diversity. Relatively high levels of diversity are evident throughout Georgia. Labor Market The proportion of the economically active population with post-secondary and higher education represents the quantity of skilled human capital available to firms. Unemployment is calculated using the ‘strict’ ILO criteria. Business Environment and Investment Climate The proportion of total regional employment in SMEs (fewer than 250 employees) is utilized to proxy for economic dynamism associated with small, entrepreneurial firms (Glaeser and Kerr 2009). The proportion of employment in local firms active in a knowledge intensive or high technology sector (KIS and HTM sectors63), proxies for the regional technology and knowledge endowment. Tbilisi dominates due to the size of the population. The investment climate is proxied by per capita private investments where Tbilisi is again dominant. Spatial Spill-Overs The spatial lag of each variable helps assess the impact of inter-regional interactions and spill-overs on local firms. This complements the analysis of a region-specific factor with an examination of the influence of the same factor in neighboring (contiguous) regions. First-nature geography is not a driver of firm competitiveness. Considering the role of first-nature geography, such as topographic ruggedness and climactic factors, suggests that it is important only for economic structure and initial conditions but firm competitiveness is associated mainly with second-nature drivers. And as Georgia continues to modernize and raise competitiveness this is only likely to continue. The proportion of highly educated workers within each region is robustly productivity-enhancing, though not evidence is found for spatial-spillovers from neighboring regions with higher endowments of skilled labor. The skills endowment of a region is closely aligned its labor productivity and emerges as one of the main drivers of firm competitiveness. Spatial spill-overs, in contrast, are found to be negatively associated with a firm’s labor productivity (although statistically insignificant). The direction of this association may show that while firm density, infrastructure, and regional wealth do generate spill-overs, there is no possibility of free riding on human resources in nearby regions. The absence of a skilled population is therefore a fundamental handicap for the performance of local firms. 63 KIS and HTM sectors include: KIS (NACE codes 64, 72, 73) and HTM (NACE codes: 24.4, 30, 32, 33). | Chapter 5: Constraints on Firm Competitiveness: Beyond Tbilisi   85 Draft 5.indd 103 22/12/14 15:19:14 Georgia Country Economic Memorandum | December 2014 Table 5.4. Labor Productivity Heterogeneity (1) (2) (3) Constant 7.943*** (0.13) 8.379*** (0.08) 7.977*** (0.11) Variance Firm 1.862 (0.02) 1.859 (0.02) 1.638 (0.02) Region 0.157 (0.07) 0.093 (0.05) Sector 0.232 (0.06) 0.259 (0.03) Variance partition coefficient Firm 9.2% 88.9% 82.3% Region 7.8% 4.7% Sector 11.1% 13.0% No. 13,958 13,958 13,958 Log likelihood -24,163.9 -24,187.1 -23,449.5 LR Test 1,887.7*** 1,841.2*** 3,316.5*** Groups 9 43 Minimum number of firms 332 2 Maximum number of firms 6,751 2,377 Average firms 1,550.9 424.6 Source: Author calculations. Despite the high rates of unemployment in Tbilisi, the city is still likely to be a draw for the most highly skilled workers. Unemployment data should be treated with caution due to the way in which the Georgian statistics office records instances of self-sufficient farming as self-employment, the heads of households engaged in agriculture as ‘individual entrepreneurs’, and family members as ‘unpaid family business workers’, resulting in relatively low estimates for unemployment in rural environments. This appears to affect regions like Guria, which are characterized by relatively low unemployment rates, suggesting that workers are absorbed by the agricultural sector and by related low-productivity activities, and distorts the correlation between unemployment and competitiveness. The dominance of Tbilisi, which has the highest unemployment in the country, does the same, so that even though unemployment is negatively correlated with firm competitiveness, there are significant and also negative spatial spill-overs as neighboring regions find it difficult to attract away skilled workers from the capital. High-income regions are where the most competitive firms are to be found in Georgia. We find that in addition to density and diversity, human capital rich and high-income regions provide the most fertile environment in which firms can thrive, according with the prescriptions of the 2009 World Development Report. Locations with more wealthy neighboring regions seem to be detrimental to competitiveness, suggesting constraints to benefiting from spill-overs in terms of access to resources. There is a broadly negative association between firm competitiveness and social disadvantage, such as that experienced by the large minority and displaced populations along the Southern border of Georgia and other regions with relatively high rates of poverty. A wide range of social indicators, such as numbers of pension recipients, levels of infant mortality, and incidences of cardio-vascular disease were also evaluated. In most instances these additional data were only available for single years, and were treated a time invariant over the period. These factors also proved negative and significant suggesting the importance of demographic factors and of an enabling social and institutional environment. Reversing the marginalization of minority groups and 86 | Chapter 5: Constraints on Firm Competitiveness: Beyond Tbilisi Draft 5.indd 104 22/12/14 15:19:14 Georgia: Seizing the Opportunity to Prosper IDPs could be important for engendering virtuous circles of development, and provide the boost needed to reduce regional disparities. These results also reflect issues relating to a lack of agglomeration benefits, such as a critical mass of firms, both in the same sector (specialization economies) and diverse (Jacobian economies) sectors, and cultural (minorities and IDPs) and physical divisions (mountains) with key regional markets. Size matters, especially in manufacturing, suggesting a “glass ceiling” on productivity growth for SMEs. While positive, the impact of knowledge intensive firms has a statistically insignificant impact on firm competitiveness; however the share of employment in SMEs, a typical indicator used to proxy for a vibrant, competitive, and entrepreneurial business environment—i.e. the small and medium firms that make important contributions to economic and social development globally—is negatively associated with firm performance. As manufacturing firms comprise a significant proportion of the dataset, it is possible that economies of scale predominate in this result. As SMEs tend to be most prevalent in less productive regions, their size may reflect barriers to growth, and represent a glass ceiling to firms becoming large. This is consistent with the finding in World Bank (2013) that firm productivity does not evolve over the life-cycle in Georgia as it does in other, higher value-added exporters. C. Decentralization and Firm Competitiveness in Georgia Despite the considerable progress that has been made, Georgia remains one of the poorest countries in the ECA region with a significant pending reform agenda. Persistent territorial disparities, high levels of poverty, a high incidence of rural subsistence agriculture and urban unemployment highlight the underlying economic, social and political challenges that threaten to stall the future development of Georgian regions (Japaridze 2010; World Bank 2013). Significant reforms across all regions are necessary if Georgian firms are to modernize and compete more widely, and for Georgia to begin to reduce the rampant trade deficit that is being accumulated. In particular, gaps in infrastructure, health and education, financial access and property rights need to be addressed. This will call for greater voice and better, more inclusive governance and strengthened public service provision, both in terms of territorial and demographic access and quality. A fundamental issue for Georgia is that the system of governance has largely failed to make inroads towards the aspiration of greater local self-government, or address the identified regional disparities highlighted in our analysis  (Jackson 2004). A critical question for the future of Georgia is how to resolve the significant regional cleavages that threaten the future socio-economic stability of the country, and undermine the competitiveness of Georgia’s firms. Addressing the limited autonomy and capacity, including key obstacles to effective governance, of local and regional government units, is now a major priority for the Georgian state. The drive for better local governance, and a push for bottom-up development, is also part of growing global trend. Decentralization is increasingly seen as a strategy to engender social change and ensure the generation of an “economic dividend.” For Georgia, a principle objective behind the current State Strategy for Regional Development 2010–2017 (Government of Georgia 2010) is to create a favorable environment for the socio-economic development of the regions and improve living standards and conditions of the population. A core component is to strengthen local self-governance and to increase economic efficiency and democracy | Chapter 5: Constraints on Firm Competitiveness: Beyond Tbilisi   87 Draft 5.indd 105 22/12/14 15:19:14 Georgia Country Economic Memorandum | December 2014 through the de-concentration and decentralization of authority and decision-making to lower tiers of governance where appropriate. To do so, Georgia is putting in place significant efforts to redefine the constitutional, legal, financial and economic basis on which local self-government can be affected throughout the country with the implementation of a new Local Self-Government Code.64 Decentralization, when well designed and successfully implemented, is an effective tool for tackling many of the issues that Georgia currently faces. Decentralized government systems can improve the transparency, accountability and the responsiveness of the public sector, by invoking greater levels of participation in local government decision-making and increasing representation of diverse interest groups at the local level to bring decision making processes and responsibilities as close to the people as possible, thus, intensifying pressure on governments to work more effectively. For supporting firm competitiveness and growth, the development of SMEs, and broader issues of job creation, poverty reduction, better public services delivery—such as that of local infrastructure, education, healthcare and social protection—decentralization is often also argued to be more efficient, effective, and targeted, given that it encourages the demand for good governance and for better quality of service provision.65 While this can form a strong basis for sustainable local economic development, the process itself presents a range of political and economic risks.66 There is a range of complexities and challenges associated with decentralization initiatives, including agency problems (especially where the process is top-down as in Georgia), institutional weaknesses and territorial competition, associated with important trade-offs between efficiency, inequality and competitive capacities. This risks compromising local capacity to deliver services because of lack of adequate resources. At the same time there have to be checks and balances for the central government to ensure regions spend sustainably and efficiently. The process therefore requires strong coordination and clear lines of responsibility between different tiers of government, and oversight from strong central governments to ensure accountability, transparency and fiscal discipline from local government units. There is also a risk that fiscal decentralization is likely to disproportionately benefit the most prosperous regions. If resources flow to regions better endowed economically and in institutions, there is a risk of increasing interregional disparities and further marginalizing lagging and less-favored. As described by Rodríguez-Pose and Gill (2005), ‘poorer infrastructure, as well as smaller tax bases, less access to financial markets, less influence over the discretionary aspect of central government finances and fewer, or smaller, input and output markets, may lose the battle for some states before it has even begun’ (p. 413). In light of the large and growing disparities between regions in Georgia, decentralization may lead a reinforcement of leading regions67. Further downsides for lagging regions include a lower provision of public goods, due to an inability to maximize economies of scale, and greater potential for clientelistic practices by bureaucrats and stakeholders. Thus, the success of decentralization hinges on its design. In particular, policies 64 Keating 1998; Rodríguez-Pose and Gill 2005. 65 Von Braun and Grote 2000; Bardhan 2002; Klugman 1994; Agrawal and Ribot 1999; Ebel and Yilmaz 2002; Brenner 2004. 66 Prud’homme 1995; Rodden 2002; Gill 2005; Felzenshòtain and Portnov 2005, Dabla-Norris 2006, Cheshire and Gordon 1996; Martínez-Vázquez and McNab 2003; Ezcurra and Pascual 2008. 67 Rodríguez-Pose and Krøijer 2009, Treisman 2000; Rodríguez-Pose and Gill 2005. 88 | Chapter 5: Constraints on Firm Competitiveness: Beyond Tbilisi Draft 5.indd 106 22/12/14 15:19:14 Georgia: Seizing the Opportunity to Prosper must be sequenced to match administrative, fiscal, and political powers at the right time in the process and must provide an adequate level of support to regions to carry out their growing responsibilities. By extending the modelling framework to examine the influence of public expenditures at the regional level, we can arrive at a better assessment of the potential benefits of decentralization. As we have established, local government capacities and regional endowments seem to have a clear role in the divergent patterns of firm efficiency. Are current spending patterns configured in such a way as to be disparity enhancing, as opposed to disparity reducing? Moreover, what are the implications for future government policy, and more specifically, for the decentralization process in Georgia? Figure 5.5. Public Spending by Region Total expenses, per capita expenditure (GEL, 000’s) Racha-Lechkhumi-Kvemo Svaneti Samegrelo-Zemo Svaneti Mtskheta- Mtianeti Imereti Shida Kartli Guria Adjara Kakheti Samtskhe- Tbilisi Javakheti Kvemo Kartli JJ11.8019–11.8019 JJ11.8019–15.0299 JJ15.0299–17.1351 JJ17.1351–37.2119 JJ37.2119–47.1175 Source: Hardy and Rodriguez-Pose 2014. Currently, the leading regions of Georgia tend to enjoy the highest levels of per capita funding, particularly Tbilisi and Adjara. Per capita expenditures across regions also vary widely, with four- and five-fold inter-regional disparities between the best funded (Adjara and Tbilisi) and worst funded (Kakheti) administrative divisions. These differences are further magnified at the municipality level, as noted in the World Bank’s Public Expenditure Review (2014). In addition, the regional patterns differ according to the core component of regional public expenditure. Where transportation expenditures are particularly high (in per capita terms) in Tbilisi and Georgia’s leading Western regions—Adjara in particular—education and social expenditures appear to be structured in a more balanced way, with some intermediate and lagging regions ranking higher in per capita expenditures. Social expenditures, in particular, are higher in the most lagging regions, in regions where you would expect to find the most need, largely because of transfers through pensions and the TSA. However, the overall balance tends to place upward pressure of existing disparities, with Tbilisi consistently outspending all other regions in per capita terms. Regional public expenditures per capita are strongly associated with firm competitiveness. When we analyze the components of this spending, we find a range of factors that are conditionally correlated with labor productivity and the TFP of manufacturing firms. Core components such as transport expenditures are positively associated with firm competitiveness. Expenditure on subsidies and social protection are both also positively and significantly associated with firm performance, which may suggest that the ways in which Georgian regions support and incentivize local firms may prove to yield some benefits. However, we do not have data on the nature of subsidies spending within regions to test this claim. Chapter 5: Constraints on Firm Competitiveness: Beyond Tbilisi   89 | Draft 5.indd 107 22/12/14 15:19:14 Georgia Country Economic Memorandum | December 2014 While per capita public spending on education is not generally associated with firm performance, this finding is likely to reflect variations in the costs of education as well as local capacity and quality. As illustrated by Chankseliani (2013), rural disadvantage in education is a concern—81 percent of university students are from urban areas, which constitute over half the population—with an even wider urban-rural divide in the most prestigious higher education institutions in the country. In the long term this can only serve to curb the aspirations of the rural youth and to further regional disparities. As the costs of education are likely to vary widely per head, with remote regions likely to receive higher spending per head than schools in large agglomerations, this finding should be interpreted with caution. It may also reflect expenditure efficiency and local capacity. Although some of the regional discrepancies may simply reflect differences in costs related to physical geography and inherited infrastructure and facilities, differences may also point toward a range of real regional concerns. It is also vital to ensure that the decentralization processes supports rather than undermines each of the four key pillars of local and regional development; the competitiveness of local firms, attraction of inward investment, development of a regional base of skilled human capital, and upgrading of local infrastructure. Firm competitiveness is closely related to endowments of local infrastructure and human capital, and is responsive to public spending per capita. If decentralization is to be performance enhancing, regional governments need the power and resources to tackle the areas, by investing in infrastructure, education and public policies that help to reduce transactions costs for firms, allowing them to compete more widely, domestically and internationally. Much of the progress made to date on decentralization in Georgia has not been part of a comprehensive strategy, also a problem for several other transition countries in the region. The current State Strategy for Regional Development provides a more solid and well-articulated program for decentralization, with the potential to create a solid foundation for future reform and development progress. This research and discussion has identified some key areas that need to be addressed in order to ensure that decentralization does not inadvertently harm particular regions, or compromise the competitiveness of local firms. Accordingly, if the decentralization process is to achieve its stated goals, including increased competitiveness and balanced socio-economic development throughout the regions, it will require some consideration of the following points: •• Firstly, and generally, the decentralization process requires effective, committed and empowered local and central government units with the sufficient capacity to coordinate and oversee the process, ensuring that the process is inclusive for all groups within the region. •• Secondly, and more specific to Georgia, having already made some progress with the vertical structure of government, resulting in more appropriate territorial divisions, better suited to addressing the needs of the region, and increasing autonomy with respect to carrying out public policies, measures need to be taken to maximize efficiency and develop the capacity of lagging and peripheral regions to administer public services and investments effectively. Divergences in the transparency, efficiency and accountability of regional governments must be managed to correct market failures, provide access to markets and manage economic growth and development. Responsibilities should only be decentralized in conjunction with specific plans to enhance the capacity of sub-national governments, especially in those regions lagging behind. 90 | Chapter 5: Constraints on Firm Competitiveness: Beyond Tbilisi Draft 5.indd 108 22/12/14 15:19:14 Georgia: Seizing the Opportunity to Prosper •• Thirdly, the discrepancy between responsibility and resources needs to be better addressed with increased fiscal decentralization. •• Finally, the data show that there are the significant divergences in current public expenditures per capita across regions, which serve to sustain or exasperate, rather than ameliorate, existing inter-territorial disparities and the transfer system needs to be overhauled to better reflect these differences. In addition, as regions are granted greater autonomy to generate revenues locally as part of the decentralization program, divergences in capacities to generate revenues need to be monitored to support equalization across territories. D. Conclusions Georgia has made considerable progress in terms of economic growth, development and governance over the last two decades since independence, though trends are variable across the country. Strong performance on growth has been accompanied by strengthened institutions, opening up and success at increasing trade and foreign investment. However, longstanding territorial disparities have to be addressed. There are divergences in poverty, labor productivity, per capita spending and core investments, skills, infrastructure, and employment that suggest that Georgia’s growth process has affected firms located in different regions in different ways. In particular, as Georgia seeks “export-led” growth and enhanced market access through improved economic ties with the EU, these disparities will continue to affect how the benefits of trade are shared geographically and across demographic groups. Significant attention needs to be paid to building a better business environment in less-favored areas, so they can benefit from inward investment and productivity growth. Georgia has done outstandingly well on the Doing business index, which captures a small range of business environment characteristics—however the Doing business survey only covers Tbilisi. The analysis highlights that local expenditures, especially in transport infrastructure and human capital, significantly affect the competitiveness of Georgian firms in different parts of the country. Therefore, given the current endowments across regions, if global trade is to bring prosperity to peripheral or lagging regions, policy attention needs to focus on investment in public services to help increase firm productivity and allow businesses to grow, compete domestically and abroad, and benefit from economies of scale. The ongoing decentralization process in Georgia also has implications for firm competitiveness and for regional economic development in general. The decentralization process does offer an opportunity to strengthen public participation, voice, and promote inclusive governance. Where successfully implemented, this process can support the specific needs of lagging regions and groups in Georgia, and improve the prospects for firms to compete and regions to develop. However, evidence tends to be mixed on whether impacts of decentralization are positive or negative, and suggests that the complexities of design, planning and sequencing are major stumbling blocks that require close consideration. Regional government units need to be able to diagnose and evaluate bottlenecks in the regional environment, including shortcomings in local socio-economic conditions, identifying | Chapter 5: Constraints on Firm Competitiveness: Beyond Tbilisi   91 Draft 5.indd 109 22/12/14 15:19:14 Georgia Country Economic Memorandum | December 2014 on the way appropriate actors and stakeholders to instigate reforms and balance the complex needs of the local territory. Table 5.5. Main Policy Messages Spatial inclusion calls for strengthened firm Stronger focus on access to and quality of public services competitiveness throughout the country. in the region, particularly health and education, and infrastructure, will be important. Job creation and support for small firms should be a Supporting the SME sector will be critical for job creation priority, particularly in areas where unemployment is and to strengthen regional opportunities and narrow high (which includes the capital). disparities. The next chapter explores policy options in this context. The question for policy is how the decentralization Careful attention to sequencing reforms so that there is a process could maximize the benefits while minimizing balanced matching of power, responsibility and resources risks and avoiding the pitfalls. along with institutional capacity building. 92 | Chapter 5: Constraints on Firm Competitiveness: Beyond Tbilisi Draft 5.indd 110 22/12/14 15:19:14 Georgia: Seizing the Opportunity to Prosper DRIVERS AND SUSTAINABILITY OF CHAPTER 6:  EXTERNAL IMBALANCES Over the last twenty years, Georgia has run current account deficits that on occasion have surpassed 20 percent of GDP. Persistent current account deficits (CAD) have led to high external financing needs and to the country’s net external liability position exceeding 100 percent of GDP, and to risk perceptions that have increased borrowing costs for firms and households. High capital inflows have also contributed significantly to the dollarization68 of Georgia’s financial system. Two thirds of loans taken in Georgia are in foreign currency; when assets are predominantly Lari-denominated, the risk of currency movements generating balance sheet effects is high. Georgia’s external accounts reflect challenges that can be addressed through a shift towards new, export- oriented and sustainable sources of growth. About two-thirds of Georgia’s current account deficit is structural and is driven by low savings, high public spending during the crisis and its immediate aftermath which lowered national savings, and rapid consumption growth. Although small, the contribution of trade openness to the current account is nevertheless positive, reflecting the potentially important role of more focused policy support to improving export competitiveness and diversification. Although the high dollarization of the financial system dampens the effect, real exchange rate movements do impact net exports. However large capital inflows have led to a trend of real exchange rate appreciation that has been detrimental to the price competitiveness of Georgia’s exports. While Georgia has been successful at attracting foreign direct investment (FDI), which has largely financed the current account deficit, it has flowed mainly to non-tradables and has generated higher imports rather than exports. The predominance of FDI in the financial account lowers risks, though the structure of FDI needs to shift towards the tradable sectors, especially manufacturing and modern services, to generate sufficient export flows that can offset future profit repatriation. While in other developing countries, FDI inflows have both financed the current account deficit and contributed to reduce it via higher exports by foreign owned companies, in Georgia FDI has exhibited a high propensity to import, which has widened the current account deficit. The objective of the assessment summarized in this chapter is to examine the determinants of the current account, assess the structure of its financing and its macroeconomic impact, and infer a policy path consistent with external sustainability and Georgia’s growth ambitions. The rest of this chapter is organized as follows. Section 1 reviews basic trends in the current and capital accounts. Section 2 examines current and capital account determinants and the decomposition of the current account deficit (CAD) into structural and 68 It’s important to note, that dollarization has been declining in recent years. Since 2008, loans’ dollarization decreased by 13.8 p.p. to 59.4% and deposits’ dollarization—by 14.1 p.p. to 59.8%. | Chapter 6: Drivers and Sustainability of External Imbalances   93 Draft 5.indd 111 22/12/14 15:19:14 Georgia Country Economic Memorandum | December 2014 cyclical components. Section 3 presents an assessment of sustainability under different scenarios. The chapter concludes with key messages and policy implications. A. Basic Trends: Building Up External Liabilities The structure, financing and sustainability of the current account correspond directly to current and future patterns of growth and economic competitiveness. The current account balance (CAB) reflects the excess of domestic investment and consumption over domestic production, and is equivalent to how much the economy borrows from the rest of the world to finance that excess. A growing economy may need to borrow to finance high investment in which case a current account deficit now may be the source of higher growth, exports, and savings in the future. What matters are the underlying sources of the current account deficit. Whether it is driven by high public spending or private investment or by low private savings, how it is financed, and which sectors capital is flowing, all have significant implications for current account dynamics and for future growth and job creation. Table 6.1. External Accounts Million US$, unless otherwise indicated 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Current Account -384 -354 -710 -1,176 -2,009 -2,813 -1,134 -1,193 -1,840 -1,854 -951 In percent of GDP -9.6 -6.9 -11.1 -15.1 -19.8 -22.0 -10.5 -10.2 -12.7 -11.7 -5.9 Financial Account 370 300 625 1,066 1,915 2,759 916 1,009 1,673 1,742 841 In percent of GDP 9.3 5.9 9.7 13.7 18.8 21.6 8.5 8.7 11.6 11.0 5.2 Source: Georgian authorities and Bank staff calculations. Georgia’s current account has been in deficit for nearly 20 years. The CAD peaked at close to 22 percent of GDP in 2008 following four years of high growth and a pro-cyclical fiscal policy supported by high FDI inflows. The 16 percent depreciation following the onset of the global economic crisis and the conflict with Russia in 2008, which had led to lower domestic demand, falling FDI, and a halving of the CAD to 10 percent by 2009, supported recovery in exports. As growth recovered on the back of high public spending and the currency appreciated in 2010 the CAD widened again to close to 13 percent in 2012 before contracting again in response to a sharp fall in growth and domestic demand in 2013 (Figure 6.1). The trade deficit has been driven by large increases in the imports of goods that more than offset growth in exports, which was concentrated in services. In fact the services trade balance reached 8.7 percent in 2013. Underpinning these trends is a steady real exchange rate appreciation since 2004, briefly interrupted by the 2008 crises, and supported by large capital inflows, which has been detrimental to export competitiveness. The CAD also reflects an increasingly negative income balance, more than offset by growing current transfers, which are currently at 9 percent of GDP. Net income is driven mainly by the repatriation of profits by foreign-owned companies operating in Georgia. While this is consistent with patterns observed in Eastern Europe, since the investments underpinning the repatriated profits are mainly in non-tradables in Georgia, these do not reflect an improved capacity to manage the current account deficit in the future. Net income from abroad 94 | Chapter 6: Drivers and Sustainability of External Imbalances Draft 5.indd 112 22/12/14 15:19:14 Georgia: Seizing the Opportunity to Prosper fell by nearly 2 percentage points of GDP between 2004 and 2013. On the other hand current transfers, including both remittances and official transfers, have grown significantly, with private remittances catching up with official transfers for the first time in 2012 (Figure 6.1). Figure 6.1. Georgia’s External Imbalances Current Account Balance, Trade Balance, and Oil Balance Net Exports of Goods and Services In percent of GDP In percent of GDP 20 10 15 5 10 0 5 -5 0 -5 -10 -10 -15 -15 -20 -20 -25 -25 -30 -30 -35 -35 1995 96 97 98 99 2000 01 02 03 04 05 06 07 08 09 10 11 12 2013 1995 96 97 98 99 2000 01 02 03 04 05 06 07 08 09 10 11 12 2013 ▬▬CA ▬▬Trade balance ▬▬Oil balance ▬▬Net exports of goods ▬▬Next exports of services Real Effective Exchange Rate and Terms of Trade Current Transfers 2005=100 In percent of GDP 130 9 125 8 120 7 115 6 110 5 105 4 100 3 95 90 2 85 1 80 0 1995 96 97 98 99 2000 01 02 03 04 05 06 07 08 09 10 11 12 2013 1995 96 97 98 99 2000 01 02 03 04 05 06 07 08 09 10 11 12 2013 ▬▬Real effective exchange rate, CPI based ▬▬Terms of trade ▬▬Official transfers ▬▬Private transfers Source: World Bank staff calculations. The volume, dynamics and structure of capital flows have important implications for risk, growth, future current account movements, and policy responses. A higher share of FDI relative to portfolio investments or loans is associated with lower vulnerability for a given CAD. However the destination sectors of FDI matter. The size of the capital inflows needed to cover the CAD is important: larger flows may require counter-cyclical measures or could have a detrimental impact on competitiveness by causing a real exchange rate appreciation or by pushing up interest rates; smaller flows have a better chance of being addressed through export growth. While FDI has been the main source of financing for the current account, it has largely flowed to the non- tradable sectors, construction and services, and has in fact contributed to further widening of the deficit. Improvements in macro-fundamentals and the acceleration of the privatization process together with growing Chapter 6: Drivers and Sustainability of External Imbalances   95 | Draft 5.indd 113 22/12/14 15:19:15 Georgia Country Economic Memorandum | December 2014 global liquidity during the great moderation resulted in a strong increase in FDI inflows before the 2008 crises. FDI inflows peaked in 2007 at almost 18 percent of GDP, while the average annual inflows over 2003–08 were nearly 12 percent of GDP. The global financial crisis more than halved these inflows between 6 and 8 percent of GDP after 2008. In 2013 FDI inflows reached a 6.3 percent of GDP, only slightly higher than the 10-year low. However FDI has largely been concentrated in non-tradables, particularly real estate, and services, and has been associated with a high propensity to import and low export orientation, which may partly explain higher productivity levels in other countries with similar or lower foreign investment levels. Figure 6.2. Financing the Deficit Net FDI and Line Graph and Sectors (Tradables and Non- Net Portfolio and Net Loans (Bank & Corporate) tradables) Over Time In percent of GDP In percent In percent of GDP In percent of GDP 18 25 7 30 16 20 6 20 10 14 5 0 12 4 -10 15 10 -20 3 8 -30 10 2 -40 6 1 -50 4 5 -60 2 0 -70 0 0 -1 -80 1995 96 97 98 99 2000 01 02 03 04 05 06 07 08 09 10 11 12 2013 2000 01 02 03 04 05 06 07 08 09 10 11 12 2013 ▬▬Net FDI, lhs ▬▬Share manufacturing, rhs ▬▬Net portfolio flows, lhs ▬▬Net loans flows, rhs Sources of FDI by Region FDI and GDP per Capita In percent In US$ billions In US$ 100 4,000 1.8 90 3,500 1.6 80 1.4 3,000 70 1.2 2,500 60 1.0 50 2,000 0.8 40 1,500 0.6 30 1,000 20 0.4 10 500 0.2 0 0 0 1996 97 98 99 2000 01 02 03 04 05 06 07 08 09 10 11 12 2013 1995 96 97 98 99 2000 01 02 03 04 05 06 07 08 09 10 11 12 2013 JJCIS countries JJEU countries JJOther countries ▬▬Net FDI, lhs ▬▬GDP per capita, rhs Source: World Bank staff calculations. Portfolio and lending flows have increased in recent years, contributing to the accumulation of net foreign liabilities. After a sharp drop during the crisis years, portfolio inflows reversed in response to Eurobond repayment, local bond market development, and lower yields in developed markets, standing at -0.3 percent in 2013. Net lending, especially corporate, picked up starting 2006 and remains high, at nearly 3 percent. 96 | Chapter 6: Drivers and Sustainability of External Imbalances Draft 5.indd 114 22/12/14 15:19:15 Georgia: Seizing the Opportunity to Prosper  he Drivers: Brisk Growth in Credit and Imports Supported by High Capital B. T Spending The CAB is the excess of investment and consumption over domestic production. It can be thought of as the balance on earnings from abroad or as the sum of private and public sector savings net of private and public investment and consumption. The current account therefore is driven by private saving and investment decisions, fiscal policy and factors that affect overall export competitiveness. Falling national savings have contributed significantly to the CAD in recent years, a challenge given the investment needs implicit in Georgia’s high target growth rates and the risks embedded in continued reliance on external financing. Savings rates fell sharply following 2004 and were negative in 2008, when the CAD peaked. Three main drivers of lower savings following 2004 have been identified as post-transition optimism about growth prospects, the sharp increase in domestic credit to the private sector over the decade of the 2000s, Figure 6.3. Snapshot of Savings and Investment Trends in Georgia Savings vs. Investments Private Savings vs. Private Investment In percent of GDP In percent of GDP 40 30 35 25 30 20 25 20 15 15 10 10 5 5 0 0 1995 96 97 98 99 2000 01 02 03 04 05 06 07 08 09 10 11 12 2013 2001 02 03 04 05 06 07 08 09 10 11 12 2013 ▬▬Total investments ▬▬Gross national savings ▬▬Private investments ▬▬Private savings Structural and Cyclical Savings Structural and Cyclical Investments In percent of GDP In percent of GDP 25 40 20 30 15 10 20 5 10 0 -5 0 -10 -10 -15 -20 -20 -Q1 03-Q1 04-Q1 05-Q1 06-Q1 07-Q1 08-Q1 09-Q1 10-Q1 11-Q1 1 12-Q4 12-Q2 -Q1 03-Q1 04-Q1 05-Q1 06-Q1 07-Q1 08-Q1 09-Q1 10-Q1 11-Q1 1 12-Q4 12-Q2 2002 0 2002 0 ▬▬Structural savings ▬▬Cyclical savings ▬▬Structural investments ▬▬Cyclical investments Source: World Bank staff calculations. Chapter 6: Drivers and Sustainability of External Imbalances   97 | Draft 5.indd 115 22/12/14 15:19:15 Georgia Country Economic Memorandum | December 2014 and an increase in the fiscal deficit between 2007 and 2010, largely due to the fiscal stimulus needed during the crisis years.69 Moreover Georgia has had historically lower savings rates than other countries with similar demographic and income characteristics. Investment rates have been driven largely by FDI pre-crisis and by FDI and public capital spending after and have displayed less volatility. The reliance on foreign savings to finance investment and growth is not unusual for developing countries; however the low levels of savings is a matter of concern, given that investment rates have to be sustained or even increased if Georgia is to achieve faster growth. In fact a substantial increase in investment and national savings would be needed, even with faster productivity growth than the trend, to sustain long-run growth at an average of 5 percent. The savings-investment gap reflected in the CAD is due to a combination of structural and cyclical factors. The structural component of the current account, that is the product of the country’s income and demographic characteristics and the accumulated decisions of its agents in terms of consumption, investment and saving, is roughly six percent of GDP, while the cyclical component, due to conditions in the global marketplace, and the temporary behavior of domestic variables such as credit, aggregate demand, real exchange rates, and macro uncertainty, is about three percent of GDP. Figure 6.4. Structural vs. Cyclical Components of the CAB and FDI Current Account Balance Foreign Direct Investment In percent of GDP In percent of GDP 10 24 20 5 16 0 12 -5 8 4 -10 0 -15 -4 -20 -8 -Q1 03-Q1 04-Q1 05-Q1 06-Q1 07-Q1 08-Q1 09-Q1 10-Q1 11-Q1 1 12-Q4 12-Q2 -Q1 02-Q1 03-Q1 04-Q1 05-Q1 06-Q1 07-Q1 08-Q1 09-Q1 10-Q1 11-Q1 12-Q1012-Q4 2002 0 2001 2 ▬▬Structural CA ▬▬Cyclical CA ▬▬FDI inflows ▬▬Structural ▬▬Cyclical Source: World Bank staff calculations. Sources: Author’s calculations and National Bank of Georgia. Overall, the stage of development of Georgia explains a sizeable portion of the structural current account deficit. About half of the observed CAD is accounted for by two long term drivers that have remained stable over time and seem to set a lower bound for the deficit: Georgia’s income (in ppp terms) relative to its main trading partners—convergence reduces the deficit—and the lagged fertility rate, which increases the deficit. The findings are consistent with theory. Developing countries with high investment needs are typically expected to be importers of capital. High dependency rates, and a smaller share of the population in the active working age group, are typically associated with lower savings rate and Georgia is aging rapidly relative to its income cohort. Private credit growth, government spending, and FDI emerge as the most significant cyclical determinants of current account dynamics. Rapid growth in private credit to the domestic sector contributed to lower savings 69 See “Georgia Rising,” World Bank 2013. 98 | Chapter 6: Drivers and Sustainability of External Imbalances Draft 5.indd 116 22/12/14 15:19:16 Georgia: Seizing the Opportunity to Prosper and higher investment and to dollarization of the financial system; two-thirds of domestic loans are foreign currency denominated. Government spending—which peaked around the 2008 crises at 37 percent of GDP— reduced public savings and increased the CAD, while the post-crisis fiscal consolidation helped offset the impact of higher domestic demand on the current account. FDI in non-tradables has been associated with high imports, especially of capital goods and intermediates, and low export orientation. Georgia has been successful at attracting FDI, whose structural component matches that of the current account at about 6 percent of GDP. Lower world growth rates and increasing global liquidity are the main “external” or “push” factors driving the structural component of FDI inflows. The results suggest that Georgia faced a competitive market for FDI and managed to secure substantial inflows. In fact Georgia attracted relatively high FDI even during the crisis years from countries whose growth rates were not significantly affected. The post- 2004 reforms, Georgia’s high growth rate, reduced inflation volatility, and trade openness, were the significant domestic “pull” drivers of structural FDI. Shorter-run cyclical movements were also driven by push factors, such Figure 6.5. Some of the Drivers of Reliance on External Financing in Georgia Private Credit Foreign Direct Investment Inflows In percent of GDP In percent of GDP 45 20 40 18 35 16 14 30 12 25 10 20 8 15 6 10 4 5 2 0 0 1995 96 97 98 99 2000 01 02 03 04 05 06 07 08 09 10 11 12 2013 1995 96 97 98 99 2000 01 02 03 04 05 06 07 08 09 10 11 12 2013 Government Spending Population Pyramid In percent of GDP Age 40 85+ 80–84 35 75–79 70–74 30 65–69 60–64 25 55–59 50–54 45–49 20 40–44 35–39 15 30–34 25–29 10 20–24 15–19 5 10–14 5–9 0 0–4 1995 96 97 98 99 2000 01 02 03 04 05 06 07 08 09 10 11 12 2013 200 150 100 50 0 50 100 150 200 In thousands JJFemales JJMales Source: World Bank staff calculations. | Chapter 6: Drivers and Sustainability of External Imbalances   99 Draft 5.indd 117 22/12/14 15:19:16 Georgia Country Economic Memorandum | December 2014 as cyclical worsening of global growth and higher oil prices, and pull factors such as an increase in the output gap and lower public consumption spending. While portfolio flows have also been driven by local profitability and macroeconomic fundamentals, global factors played a lesser role until recently. Though FDI benefited from the period of great moderation, marked by lower interest rates, high commodity prices, high global liquidity, low risk aversion, and strong global growth, portfolio flows to Georgia did not increase, mainly because of underdeveloped capital markets. It was only with the Eurobond issues and the development of local bond markets starting 2008 that Georgia started attracting portfolio flows, albeit at a time when most countries were experiencing outflows. Starting 2010, flows increased in response to greater liquidity and softening risk. Mainly however the interest rate differential—which persisted in spite of lower risk premiums and was larger than could be accounted for by observed exchange rate volatility—has increased the appeal of Georgian debt instruments. Equity markets remained undeveloped leading to mainly longer term portfolio bond flows. Macroeconomic fundamentals have been sound in terms of lower fiscal deficits, improved inflation, steady reserve accumulation, and term structure of external debt, which has also supported higher inflows. C. External Sustainability The accumulated impact of persistent current account deficits has led to a worsening net foreign asset position, currently -104 percent of GDP, leading to concerns about external sustainability. External vulnerability is exacerbated by the high levels of dollar debt held by firms and households. In this context, the sharp exchange rate adjustments that may result from sudden stops in capital inflows could have important balance sheet effects, calling for a painful adjustment to investment and consumption that would be detrimental to growth and welfare. Here we define current account sustainability as the stable state in which the CAB generates no economic forces of its own to change its trajectory. In particular, the CAB is sustainable if continuation of the current government policy stance and private sector behavior does not result in future rapid policy shifts (such as, for example, a sudden policy tightening causing a large recession) and/or substantial changes in other economic variables, such as large exchange rate depreciations or interest rates hikes. The objective is to generate alternative scenarios of the future paths of the NIIP. The question addressed is the following: given the identified determinants of the current account and the stock at end-2012 of foreign assets and liabilities of the country, what will be the future paths of NIIP under different scenarios? The methodology does not impose any steady-state assumption on the evolution of the economy. A fast expansion of credit growth could bring the CAD to almost 11 percent of GDP, and the NIIP to -140 percent of GDP. The scenario analysis for the effect of credit growth suggests that, allowing all other relevant determinants to evolve at the benchmark trends, the CAB contracts mildly in 2014 to reach 7.1 percent of GDP, increasing gradually thereafter, to surpass 8 percent of GDP in 2018. In the most expansive scenario, in which credit is assumed to grow at 16 percent per annum after 2015, the CAD would be close to 11 percent of 100 | Chapter 6: Drivers and Sustainability of External Imbalances Draft 5.indd 118 22/12/14 15:19:16 Georgia: Seizing the Opportunity to Prosper GDP. Analogously, the NIIP would increase to up to 130 percent of GDP, from 104 percent in 2013 under the most conservative scenario, in which credit is assumed to grow at half or 8 percent of GDP. Figure 6.6. Key Balance Sheet Indicators Short-Term Debt/Total External Reserves and Short-Term Debt/Reserves In percent of total external debt In US$ billion In percent of total reserves 25 3.0 80 70 2.5 20 60 2.0 15 50 1.5 40 10 30 1.0 20 5 0.5 10 0 0 0 1995 96 97 98 99 2000 01 02 03 04 05 06 07 08 09 10 11 2012 1995 96 97 98 99 2000 01 02 03 04 05 06 07 08 09 10 11 12 2013 ▬▬Gross reserve assets, lhs ▬▬Short-term debt, rhs Interest Rate Differential and EMBI Net Foreign Assets and Total External Debt In percent In percent of GDP In percent of GDP 20 0 100 18 90 -20 16 80 14 70 -40 12 60 10 -60 50 8 40 -80 6 30 4 20 -100 2 10 0 -120 0 2003 04 05 06 07 08 09 10 11 12 2013 1995 96 97 98 99 2000 01 02 03 04 05 06 07 08 09 10 11 12 2013 ▬▬3-mo interest rate differential ▬▬1-year interest rate differential ▬▬NFA, lhs ▬▬External debt stocks, rhs ▬▬1-year FX deposits interest rate diff. ▬▬Georgia EMBI spread ▬▬EMBI global spread Source: World Bank staff calculations. The real exchange rate channel to address external imbalances is weak. The comparison of the impact of different real exchange rate scenarios on both the CAD and the NIIP reveal that the real exchange rate only has a weak effect on the external imbalances. The projected CAB for 2018 under the assumption of the strongest depreciation is only one-twentieth of a percentage point. Similarly, the difference between implied NIIP is insignificant. This reflects both high dollarization and a need to strengthen export competitiveness, as well as the dominant role of non-tradables in current account dynamics. The projected effects of GDP growth on the CAB and to a lower extent on NIIP, given a relative level of income are sizable. Under the benchmark path for GDP growth, the CAB would reach almost 9 percent in 2018, while the NIIP would reach -133 percent of GDP. On the other hand, under the assumption of stronger growth by Chapter 6: Drivers and Sustainability of External Imbalances   101 | Draft 5.indd 119 22/12/14 15:19:16 Georgia Country Economic Memorandum | December 2014 Box 6.1. Should the Level and the Composition of Georgia’s Net Foreign Liabilities be a Source of Concern Accumulation of large Net Foreign Liabilities, or analogously, a large negative NIIP can pose significant risk in terms of sustainability of large and persistent CAD. Accumulated negative NIIP directly leads to worsening of future CABs through larger income outflows. Moreover, higher negative NIIP also increases country risk and reduces the willingness of investors to finance parts of future CAD. Catao and Milesi- Ferretti (2013) recently documented in a cross-country study of external crises over the period 1970–2011 that crises risk increases with net foreign liabilities, and particularly so when these exceed 50 percent of GDP or 20 percent of its historical mean value. Moreover, they find that the composition of the NIIP also matters for vulnerability. Crises risk increases with the share of net debt liabilities in overall NIIP and suggested a (rough) threshold of 35 percent of GDP for the net debt liabilities as a signal of the crises potential. Both gross and net liabilities increased significantly in nominal value over the last decade. Gross liabilities increased from below 5bn USD in 2004 to close to 22bn USD in 2012. As a share of GDP, NIIP temporarily improved over 2004–07 and expanded strongly from 2009, reaching -98.9 percent of GDP by end-2012. The debt versus equity structure of net liabilities worsened over the crises period. The composition of the NIIP improved between 2004 and 2007, as the share of net debt liabilities in GDP was reduced from 45 percent in 2003 to 18 percent at the end of 2007, which, should have contributed to decrease Georgia’s vulnerability to external crises.70 Improvements in the relative structure followed from record levels of FDI, largely driven by the acceleration of the privatization process. However, this improvement was not long lasting. While both net debt and net equity increased strongly over the crises period, the speed of net debt accumulation was stronger and led to the rise in net debt to GDP ratio which reached 38 percent at the end of 2012. Although the share of short-term debt in the new debt increased more recently, given a dominant share of the short term assets in gross Georgia’s assets, the maturity structure of the NIIP remains favorable (short-term net assets are below 2 percent of GDP. The pace of recent NIIP accumulation, its current level and the relative structure indicate high risks for sustainability of the future current account deficits. An increased share of net debt in GDP implies a larger external exposure to global portfolio investors although the rollover/shifts in global sentiment risks are (currently) not high given the long maturity of the issued instruments. The reduction of non-resident investors’ holdings of local currency government bonds in 2013 suggests that these risks are nevertheless present. High accumulated levels of net debt imply a limited role for debt instruments in financing the CAD in the short-run and signal the need for reducing CAD to the levels that can be primarily financed with net FDI inflows and net transfers. Given the expected worsening of the income balance in relation to debt servicing and profit repatriations, adjustment in the CAD requires stronger improvements in the trade balance). 1 percentage point higher than the WEO forecast, the projected CAD would surpass 9 percent in 2018, while the NIIP would reach -136 percent of GDP. 70 Net debt liabilities are equal to the difference between the debt assets (portfolio debt, other investment debt and foreign exchange reserves) and debt liabilities (portfolio debt and other investment debt). 102 | Chapter 6: Drivers and Sustainability of External Imbalances Draft 5.indd 120 22/12/14 15:19:16 Georgia: Seizing the Opportunity to Prosper D. Conclusions Georgia’s current account is a symptom of the growth path taken since transition; sustainability calls for renewed focus on improving competitiveness. Georgia benefited from the 2004 reforms; it became and remains a competitive destination for FDI and has performed well in terms of growth. Capital accumulation and productivity growth has however largely been concentrated in non-tradables, which absorbs much of the FDI entering Georgia. These sources of growth, namely high credit growth fueled by foreign currency loans, high reliance on fiscal policy to manage even temporary shocks, and investment that is import-oriented rather than export generating, are not sustainable. FDI has been responsive to Georgia’s policy framework and to global market conditions, and has largely financed the current account deficit; however its impact on the evolution of the CAD has not been positive. Going forward, FDI is projected to remain at about 6 percent in the medium term, barring policy reversals or shocks, which is sufficient to cover Georgia’s estimated structural CAD. International experience shows however that the beneficial effects of FDI inflows on the economy and on the CAD depend on their composition. In Georgia, most FDI has been directed to non-tradable sectors, exhibiting a high propensity to import and to generate net income and profit repatriation without offsetting increases in exports down the line. In other countries where foreign investors have targeted tradable sectors instead, such as Morocco or Poland, FDI has supported growth and narrowed the CAD. Investment promotion strategies should therefore focus on sectors with potentially high spillover effects in terms of productivity and competitiveness to the rest of the economy to support exports and job growth. Recently, other and riskier sources of financing have grown in importance. Global deleveraging is likely to result in a reduction in bank and corporate flows. Higher portfolio bond inflows increase the vulnerability of Georgia’s external sector due to the higher sensitivity to the risk-return calculus. It is therefore important to broaden the investors’ base as diversified demand is important to minimize shocks. Higher participation of foreign investors in domestic markets is also beneficial for liquidity and competition which will help support the development of the long-term capital market. Better developed capital markets may support equity investment, suggesting an important area of reforms to focus on given concerns regarding the growing share of debt in external liabilities. Although small, trade openness has helped improve the CAB, suggesting a potentially important impact of trade policy. Large FDI flows suggest that the income balance will remain negative in the medium-term and that improvements in the CAD would have to come from an improved trade balance. There are three main reasons why policies aimed at improving export competitiveness will be critical for reducing the current account deficit. First, though small, trade openness has had a significant and positive impact on the CAB. Second, while real exchange rate movements do have an impact on net exports this is weakened by the high degree of dollarization of the financial system and the reliance of firms on external financing, as well as by the limited development of the tradables sectors. Third, sustained capital flows have led to a trend of real exchange rate appreciation that has hindered competitiveness. Policies to encourage savings and develop attractive local currency financial instruments will support de-dollarization and reduce the vulnerabilities arising from the CAD. Strategies need to be designed to | Chapter 6: Drivers and Sustainability of External Imbalances   103 Draft 5.indd 121 22/12/14 15:19:16 Georgia Country Economic Memorandum | December 2014 incentivize private saving through the development of a wider range of financial instruments, including for longer term saving. Pension reform would also support capital market development in the longer run. High credit growth has played a large role in widening the CAD, largely because of its composition. Two thirds of domestic loans are denominated in foreign currency, which constitutes a substantial risk for households and firms with currency mismatches in their balance sheets, for the banking system, and for the economy as a whole. Creating attractive saving instruments denominated in domestic currency will also support accessible credit and gradual de- dollarization of the financial system. Finally, in addition to supporting both savings and the financing of domestic investment through domestic resources, these policies will strengthen monetary policy channels and take some of the pressure off fiscal policy. 104 | Chapter 6: Drivers and Sustainability of External Imbalances Draft 5.indd 122 22/12/14 15:19:16 Georgia: Seizing the Opportunity to Prosper References Acemoglu, D. and D. H. Autor (2011), “Skills, Tasks and Technologies: Implications for Employment and Earnings”, in O. Ashenfelter and D. E. Card (eds.), Handbook of Labor Economics, Volume 4, Elsevier, Amsterdam. http://economics.mit.edu/files/7006 Aedo, C., J. Hentschel, J. Luque, and M. Moreno (2013) “From Occupations to Embedded Skills: A Cross-Country Comparison”, World Bank Policy Research Working Paper, 6560. Agrawal, A. and J. Ribot. (1999). Accountability in Decentralization: A Framework with South Asian and West African Cases, The Journal of Developing Areas, 33 (4): 473–502. _______, I. Cockburn, A. Galasso, and A. Oettl, (2012). Why are Some Regions More Innovative than Others? The Role of Firm Size Diversity, NBER Working Paper: 17793. Altomonte, C. and I. Colantone. (2008). Firm Heterogeneity and Endogenous Regional Disparities, Journal of Economic Geography, 8 (6): 779–810. Arias, O., S. Carolina Sánchez-Páramo, M. E. Dávalos, I. Santos, E.R. Tiongson, C. Grun, N. de Andrade Falcão, G. Saiovici, and C.A. Cancho (2014) Back to Work: Growing with Jobs in Europe and Central Asia. Washington, DC: World Bank. Artuç, E., S. Chaudhuri, and J. McLaren (2010). “Trade Shocks and Labor Adjustment: A Structural Empirical Approach.” The American Economic Review, 100: 1008-45 Autor, D. and D. Dorn (2013) “The Growth of Low-Skill Service Jobs and the Polarization of the US Labor Market”, American Economic Review, 103(5), p. 1553–1597. _______, H. Levy and F.R. Murnane (2003), “The Skill Content of Recent Technological Change: An Empirical Exploration”, Quarterly Journal of Economics, Vol. 118, No. 4, pp. 1279–1334. http://economics.mit. edu/files/569 Backman, M. (2013). Human Capital in Firms and Regions: Impact on Firm Productivity, Papers in Regional Science, doi: 10.1111/pirs.12005. Banerjee, B. and M. Jarmuzek. (2009). Anatomy of Regional Disparities in the Slovak Republic. IMF Working Paper No. 09/145. Bardhan, P. (2002). Decentralization of Governance and Development, Journal of Economic Perspectives, 16 (4): 185–205. Brambilla, I., D. Lederman, and G. Porto (2012). “Exports, Export Destinations, and Skills.” American Economic Review, 102(7): 3406–38. Beeson, P. E., D.N. DeJong, and W. Troesken. (2001). Population Growth in US Counties, 1840–1990. Regional Science and Urban Economics, 31: 669–699. Bernard, A., and J.B. Jensen (1995). “Exporters, Jobs and Wages in US Manufacturing, 1976–1987.” Brookings Papers on Economic Activity: Microeconomics, 67–112. _______, and J.B. Jensen (1999). “Exceptional Export Performance: Cause, Effect, or Both?” Journal of International Economics, 47(1): 1–25. _______, and J.B. Jensen (2004). Why Some Firms Export, Review of Economics and Statistics, 86 (2): 561–569. _______, J.B. Jensen, and P. Schott, 2005. Importers, Exporters, and Multinationals: A portrait of Firms in the U.S. that Trade Goods. NBER Working Paper, p. 11404. | References   105 Draft 5.indd 123 22/12/14 15:19:16 Georgia Country Economic Memorandum | December 2014 _______, J.B. Jensen, S. Redding and P. Schott (2007). “Firms in International Trade.” Journal of Economic Perspectives, 21(3): 105–30. _______, S.J. Redding, and P.K. Schott (2006): “Multi–Product Firms and Product Switching,” NBER Working Papers 12293, National Bureau of Economic Research. _______, S.J. Redding, and P.K. Schott (2010) ‘Multi-Product Firms and Product Switching,’ American Economic Review 100, 70–97 Besedes, T., Prusa, T., (2006). Ins, Outs and the Duration of Trade. Canadian Journal of Economics 104, 266–295. _______. T. Prusa (2007). The Role of Extensive and Intensive Margins and Export Growth, NBER Working Paper 13628 Bolashvili, P. (2002). Fiscal Autonomy Problems of Local Government in Georgia, In Davey, K. (Eds.), Fiscal Autonomy and Efficiency: Reforms in the Former Soviet Union, OSI/LGI: Budapest, 59–87. Branstetter, L.G. (2001). Are Knowledge Spillovers International or Intranational in Scope? Journal of International Economics, 53 (1): 53–79. Brenner, N. (2004). New State Spaces: Urban Governance and the Rescaling of Statehood, Oxford University Press: Oxford. Brenton, P., C. Saborowski, E. von Uexkull (2010). What Explains the Low Survival Rate of Developing Country Export Flows? World Bank Economic Review 24 (3), 474–499. Bristow, G. (2010). Resilient Regions: Re-‘place’ing Regional Competitiveness, Cambridge Journal of Regions, Economy and Society, 3: 153–167. Brülhart, M. (2011). The Spatial Effects of Trade Openness: A survey, Review of World Economics, 147: 59–83. Cadot O., L. Iacovone, M. Pierola, and F. Rauch (2013). Success and Failure of African Exporters, Journal of Development Economics 101:284–296 Carballo, J., C. Volpe Martincus (2008). Is export promotion effective in developing countries? Firm-level evidence on the intensive and the extensive margins of exports. Journal of International Economics 76, 89–106. Carneiro, P. and J.J Heckman (2003) Human Capital Policy, IZA Discussion Papers 821, Institute for the Study of Labor (IZA). Cebeci, T., A. Fernandes, C. Freund, and M. Pierola (2012). “Exporter Dynamics Database,” World Bank Policy Research Working Paper 6229. ________, D. Lederman, and D. Rojas (2013). “The Structure of Exports across Destinations and Labor-Market Outcomes: An Empirical Case Study of Turkey” (unpublished manuscript). World Bank, Washington, DC. Chankseliani, M. (2013). Rural Disadvantage in Georgian Higher Education Admissions: A Mixed-Methods Study, Comparative Education Review, 57 (3): 424–456. Cheshire, P.C. and I.R. Gordon (1998). Territorial Competition: Some Lessons for Policy, Annuals of Regional Science, 32: 321–346. Chinitz, B. (1961). Contrasts in Agglomeration: New York and Pittsburgh, The American Economic Review, 51 (2): 279–289. Ciccone, A. (2002) Agglomeration Effects in Europe, European Economic Review 46: 213–227. Clerides, S., S. Lach and J. Tybout (1998). “Is Learning by Exporting Important? Micro-dynamic evidence from Colombia, Mexico, and Morocco.” The Quarterly Journal of Economics, 113(3): 903–47. 106 | References Draft 5.indd 124 22/12/14 15:19:16 Georgia: Seizing the Opportunity to Prosper Cunha, F. and J.J Heckman (2006) Investing in Our Young People, NBER Working Papers 16201, National Bureau of Economic Research. Dabla-Norris, E. (2006) The Challenge of Fiscal Decentralization in Transition Countries. Comparative Economic Studies, 48: 100–131. Dunn, J. and D. Wetzel (2000). Fiscal Decentralisation in Former Socialist Economies: Progress and Prospects, World Bank: Washington, D.C. Duranton, G. and D. Puga (2001). Nursery Cities: Urban Diversity, Process Innovation, and the Life Cycle of Products, American Economic Review, 91 (5): 1454–1477. _______. and D. Puga (2004). Micro-Foundations of Urban Agglomeration Economies, In Henderson, V. and Thisse, JF. (Eds.), Handbook of Regional and Urban Economics, 4: 2063–2117. Ebel, R. D. and S. Yilmaz (2002). Concept of Fiscal Decentralization and Worldwide Overview, World Bank: Washington, DC Ellison, G. and E. L. Glaeser, (1999). The Geographic Concentration of Industry: Does Natural Advantage Explain Agglomeration? American Economic Review, 89 (2): 311–316. Expert Group on Future Skills Needs (2007) “Tomorrow’s Skills: Towards a National Skills Strategy”, 5th Report. Ezcurra, R. and P. Pascual (2008) The Link Between Fiscal Decentralisation and Regional Disparities: Evidence From Several European Union Countries, Environment and Planning A, 40 (5): 1185–1201. Farole, T., A. Rodríguez-Pose, and M. Storper (2011). Cohesion Policy in the European Union: Growth, Geography, Institutions, Journal of Common Market Studies, 45 (5): 1089–1111. Feldman, M.P. (1994). The Geography of Innovation, Kluwer Academic Publishers: Boston. Felzenshòtain, D. and B.A. Portnov (2005). Regional disparities in small countries, Springer. Fitzpatrick, P. and K. McQuinn (2005). Measuring Bank Profit Efficiency, Research Technical Paper, 3, Economic Analysis and Research Department, Central Bank and Financial Services: Ireland. Fujita M. and J.F. Thisse (2002). Economics of Agglomeration, Cambridge University Press: Cambridge Fujita, M., P. Krugman, and A.Venables (1999). The Spatial Economy: Cities, Regions, and International Trade, MIT Press: Cambridge. Geroski, P.A. (1995). What Do We Know about Entry? International Journal of Industrial Organization, 13: 421–440. Ghemawat, P. (2007). Redefining Global Strategy: Crossing borders in a World Where Differences Still Matter, Harvard Business Press. Glaeser E., H. Kallal, J. Scheinkman, and A. Shleifer, (1992). Growth in Cities, Journal of Political Economy, 100: 1126–1152. _______, and Mare (2001). Cities and Skills, Journal of Labor Economics, 19 (2): 316–342. _______, and W. Kerr (2009). Local Industrial Conditions and Entrepreneurship: How Much of the Spatial Distribution Can We Explain? Journal of Economics and Management Strategy 18 (3): 623–663. Görg, H., R. Kneller and B. Muraközy (2012). What makes a successful export? Evidence from Firm-Product- Level Data. Canadian Journal of Economics Canadian Economics Association 45 (4), 1332–1368. Government of Georgia (2010) State Strategy for the Regional Development of Georgia 2010–2017, Government of Georgia: Tbilisi. | References   107 Draft 5.indd 125 22/12/14 15:19:16 Georgia Country Economic Memorandum | December 2014 Grävingholt, J., B. Doerr, K. Meissner, S. Pletziger, J. von Rümker, and J. Weikert (2006). Strengthening Participation through Decentralization: Findings on Local Economic Development in Kyrgyzstan, Studies, German Development Institute: Bonn Hanushek, E., L. Woessmann and L. Zhang (2011) “General Education, Vocational Education, and Labor-Market Outcomes over the Life-Cycle”, NBER Working Paper 17504. Hausmann, R., J. Hwang and D. Rodrik (2007). “What You Export Matters.” Journal of Economic Growth, 12(1): 1–25. __________, D. Rodrik (2003). Economic development as self-discovery. Journal of Development Economics 72, 603–633. Hayek, F.A. (1945). The Use of Knowledge in Society, The American Economic Review, 35 (4): 519–530. Hayek, F.A. (1978). Competition as a Discovery Process, In Hayek, FA. (Eds.). Studies in Philosophy, Politics and Economics, Chicago: 66–81. Henderson, J. V. (1988). Urban Development: Theory, Fact, and Illusion, Oxford University Press: New York. Hess, W. and M. Persson (2010). The Duration of Trade Revisited: Continuous-Time vs. Discrete-Time Hazards. Lund University, Department of Economics working paper 2010:1. Hollweg, C., and E. Ruppert Bulmer. 2014. “Testing the Link between Exporting and Labor-Market Outcomes” (unpublished manuscript). World Bank, Washington, DC. Huber, M. (2004). State-building in Georgia: Unfinished and at Risk?, Clingendael Institute, February, The Hague. Jackson, P. (2004). Ethnicity, Decentralisation and the fissile state in Georgia, Public Administration and Development, 24 (1): 75–86. Jacobs, J. (1969). The Economy of Cities, Vintage: New York. Jakubowski, M., H.A. Patrinos, E. E. Porta, and J. Wisniewski (2010) “The Impact of the 1999 Education Reform in Poland”, Policy Research Working Paper Series 5263, World Bank, Washington, DC. Japaridze, D. (2010). Post-Crisis Challenges of Regional Economic Development in Georgia, Spatial Economy for Society Conference: Poznan Kaplan, D. S., D. Lederman, and R. Robertson (2013). “Worker-Level Adjustment Costs in a Developing Country: Evidence from Mexico” (unpublished manuscript). World Bank, Washington, DC. Keating, M. (1998). The New Regionalism in Western Europe, Edward Elgar. Kirn, T. and E. Khokrishvili (2009). Challenges of Decentralisation in the Republic of Georgia, In King, L. and Khubua, G. (Eds.), Georgia in Transition, Peter Lang: Frankfurt. Klugman, J. (1994). Decentralisation: a survey of literature from a human development perspective, United Nations Development Programme, Human Development Report Office: New York. Krugman, P. (1991). Increasing Returns and Economic Geography, Journal of Political Economy, 99: 483–499. _______, (1997). Pop Internationalism, MIT Press: Cambridge. Levinsohn J. and A. Petrin (2003) Estimating Production Functions Using Inputs to Control for Unobservables. The Review of Economic Studies, Vol. 70, No. 2, pp. 317–341 108 | References Draft 5.indd 126 22/12/14 15:19:16 Georgia: Seizing the Opportunity to Prosper Levinsohn, J. and A. Petrin (2003). Estimating Production Functions Using Inputs to Control for Unobservables, Review of Economic Studies, 70: 317–341. Losaberidze, D. (2007). Kyrgyzstan and Georgia—Perspectives Based on Georgia’s Decentralization Experience, in Toft, D. (Eds.), The challenges of decentralization in the Kyrgyz Republic, Fiscal Decentralization Initiative for Central and Eastern Europe: Budapest. Malmberg, A., B. Malmberg, and B. Lundeqvist (2000). Agglomeration and Firm Performance: Economies of Scale, Localization, and Urbanization among Swedish Export Firms, Environment and Planning A, 32: 305–321. Marshall, A. (1890). Principles of Economics: An Introductory Volume, MacMillan: London. Martínez-Vázquez, J. and R. McNab (2003). Fiscal decentralization and economic growth, World Development, 31: 1597–1616. Natsvlishvili, I. (2011) “Entrepreneurship Attitudes in the Context of Post-Soviet Transformation (Case of Georgia)”, Tbilisi State University and George Washington University, Washington, DC. Oates, W.E. (1999). An Essay on Fiscal Federalism, Journal of Economic Literature, 37 (3): 1120–1149. Odero, K.K. (2004). PRSPs in decentralized contexts: Comparative lessons on local planning and fiscal dimensions, World Bank: Washington, DC OECD (2010). “The High Cost of Low Educational Performance: The Long-Run Economic Impact of Improving PISA Outcomes”, Organization for Economic Co-operation and Development ,Paris. Ottaviano, G.I.P. and J.F. Thisse, (2004). Agglomeration and Economic Geography, In Henderson, J V. and Thisse, J.F. (Eds.), Handbook of Regional and Urban Economics, 4, North-Holland: Amsterdam, 2563–2608. _______, and J.F. Thisse, (2005). New Economic Geography: what about the N? Environment and Planning A, 37 (10): 1707–1725. Ö̈zsoy, I. and M. Kubetova, (2006). The Development of Business Sector in Georgia, IBSU Scientific Journal, 1 (1): 1–14. Prud’homme, R. (1995). The dangers of decentralization, World Bank Research Observer, 10: 201–220. Rappaport, J. and J.D. Sachs (2003). The United States as a Coastal Nation, Journal of Economic Growth 8 (1): 5–46. Rauch, J. and J. Watson (2003). “Starting Small in an Unfamiliar Environment,” International Journal of Industrial Organization, 21, 1021–1042. Reyes, J.D. and G. Varela (2013). “Trade Competitiveness Diagnostic—The Case of Georgia”, World Bank mimeo. Roberts, M.J. and J.R. Tybout (1997). The decision to export in Colombia: an empirical model of entry with sunk costs, The American Economic Review, 87 (4): 545–564. Rodden, J. (2002). The Dilemma Of Fiscal Federalism: Grants and Fiscal Performances Around the World, American Journal of Political Science 46 (3): 670–687. Rodríguez-Pose, A. (2002). The Role of the ILO in Implementing Local Economic Development Strategies in a Globalized World: ILO. Rodríguez-Pose, A. (2012). Trade and regional inequality, Economic Geography, 88 (2): 109–136. Rodríguez-Pose, A. and N. Gill (2005). On the ‘economic dividend’ of devolution, Regional Studies, 39 (4): 405– 420. | References   109 Draft 5.indd 127 22/12/14 15:19:16 Georgia Country Economic Memorandum | December 2014 _______,. and A. Krøijer, (2009). Fiscal decentralization and economic growth in Central and Eastern Europe, Growth and Change 40 (3): 387–417. Scott, J. (1988). Social Network Analysis, Sociology, 22 (1): 109–127. Seker, M. (2010). “Difficulties in Job Creation and Exporting.” Enterprise Note Series No. 12, Enterprise Surveys, The World Bank, Washington, D.C. Shankar, R. and A. Shah, (2003). Bridging the Economic Divide Within Countries: A Scorecard on the Performance of Regional Policies in Reducing Regional Income Disparities, World Development, 31 (8): 1421–1441. Strumpf, K.S. (2002). Does Fiscal Decentralization Increase Policy Innovation?, Journal of Public Economic Theory, 4: 207–241. Tatar, M. (2010). Estonian local government absorption capacity of European Union structural funds, Administrative Culture, 11 (2): 202–226. Timm, C. (2013). Economic regulation and state interventions: Georgia’s move from neoliberalism to state managed capitalism, PFH Forschungspapiere/Research Paper No. 2013/03: Göttingen. Tondl, G. and G. Vuksic, (2003). What makes regions in Eastern Europe catching up? The role of foreign investment, human resources and geography, ZEI working paper No. B 12-2003. Treisman, D. (2000). Decentralization and the Quality of Government, UCLA. UNDP. (2005). Fiscal Decentralisation in Transition Economies: Case Studies from the Balkans and Caucasus, UNDP: Bratislava Regional Centre. Van Beveren I. (2012) Total Factor Productivity Estimation: A Practical Review, Journal of Economic Surveys Volume 26, Issue 1, pages 98–128. Véron, R., G. Williams, S. Corbridge, and M. Srivastava (2006). Decentralized Corruption Or Corrupt Decentralization? Community Monitoring of Poverty-Alleviation Schemes in Eastern India, World Development, 34 (11): 1922–1941. Volpe Martincus C. and J. Carballo (2009) Survival of new exporters in developing countries: does it matter how they diversify? IDB working paper series; 140. Von Braun, J. and U. Grote (2000, November). Does decentralization serve the poor? International Monetary Fund Conference on Fiscal Decentralization, Washington, DC, November. Wetzel, D. and J. Dunn (2001). Decentralisation in the Transition Economies: Challenges and the Road Ahead, PREM Unit Europe and Central Asia, World Bank: Washington, DC. Williamson, J.G. (2005). Winners and losers over two centuries of globalization. Wider perspectives on global development, UNU-WIDER, Palgrave Macmillan: New York, 136–174. World Bank. (1994). Infrastructure for Development, World Development Report 1994, Oxford University Press: New York _______. (2009). Reshaping economic geography, World Development Report 2009, World Bank: Washington, DC _______. (2010). Doing business 2011: Georgia - making a difference for entrepreneurs: comparing business regulation in 183 economies, World Bank: Washington, DC. _______. (2013a) “Doing Business 2014: Understanding Regulations for Small and Medium-Size Enterprises”, Washington, DC: World Bank Group. _______. (2013b) “Fostering entrepreneurship in Georgia”. 110 | References Draft 5.indd 128 22/12/14 15:19:17 Georgia: Seizing the Opportunity to Prosper _______. (2013c). Georgia Rising: Sustaining rapid economic growth, Country Economic Memorandum, PREM Unit Europe and Central Asia, World Bank: Washington, DC World Bank. (forthcoming). Public expenditure on a frontline: Strategic issues and reform agenda, World Bank: Washington, DC. | References   111 Draft 5.indd 129 22/12/14 15:19:17 Draft 5.indd 130 22/12/14 15:19:17 Draft 5.indd 131 22/12/14 15:19:17 Draft 5.indd 132 22/12/14 15:19:17