November 2014 Document of The World Bank HOW TO SUSTAIN EXPORT DYNAMISM BY REDUCING DUALITY INTE DMINCAN REPUBLC A World Bank Trade Competitiveness Diagnostic gyWORLD BANK GROUP This volume is a product of the staff of the International Bank for Reconstruction and Development/The World Bank.The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent.The World Bank does not guarantee the accuracy of the data included in this work.The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part ofThe World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permis- sion may be a violation of applicable law. The International Bank for Reconstruction and Development/ The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750- 4470, http://www.copyright.com/. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org Editorial coordination: Miguel Eduardo Snchez Martin, Giselle Vel6squez y Alejandra de la Paz. Design: Florencia Micheltorena. Translation: Cristina Amiama. November 2014 Document ofThe World Bank HOW TO SUSTAIN EXPORT DYNAMISM BY REDUCING DUALITY NT OME N_ CA REPULC A World Bank Trade Competitiveness Diagnostic k WORLD BANKGROUP EXECUTIVE SUMMARY 10 INTRODUCTION AND MOTIVATION 20 1. TRADE OUTCOMES ANALYSIS 25 a. Recent Export Performance and the Role of Special Economic Zones 26 b. Changes in Sector Specialization and Excessive Market Concentration 29 c. Two-Tier Export Basket in terms of Sophistication and Quality 39 d. A Survival Analysis of Agricultural Exporters 44 e. Summary of Challenges and Potential Bottlenecks 47 II. COMPETITIVENESS DIAGNOSTIC 50 a. The Dominican Republik's Tradle Complilance Capacity in Fruits and Vegetables. 50 1. An Overview of US Agricultural Market Access Policy 51 2. US Import Alerts and Border Rejection Analysis 52 3. Institutions and Policy Options to Enhance SPS Complilance in the Dominican Republic 59 b. The role of Free Trade Zones in the Dominican Republic: Adjusting to a New Era of Competition 62 1. From Textile-led Exports to Relative Diversification: Recent Developments in SEZs 63 2. Diverse Pathways to WTO Compliance 67 3. Looking Forward: End of the SEZ Model as We Know It? 70 c. The institutional infrastructure supporting international trade 76 1. Institutional Architecture in the Dominican Republic: A Dual Structure 76 2. An Institutionally Empowered Export Promotion Agency: A Tool to Foster Competitiveness 80 REFERENCES 86 ANNEXES 91 Annex 1: Comparison Between Dominican Customs Registries and COMTRADE Data 92 Annex 2:The Margins of Trade and their Contribution to Export Growth - Decomposing the Extensive Margin in the National Regime of the Dominican Republic 93 Annex 3: Quality Ladders for Main Products Exported from SEZs 95 Annex 4: New Access to the US for CAFTA Countries 96 Annex 5: Technical Appendix to Part I, Trade Outcomes Analysis 97 Annex 6. Dominican Exporters-Results from the Enterprise Survey 100 LIST OF FIGURES Figure 1: Structure of the report 12 Figure 2: Evolution of terms of trade and real exchange rate in the Dominican Republic 22 Figure 3: Evolution of sez and non-sez real exports in dominican republic 22 Figure 4: Dominican Republic: exports, imports, and trade balance 26 Figure 5: Export performance between 26 Figure 6: Exports over GDP and GDP per capita 27 Figure 7: Exports growth decomposition 29 Figure 8: Participation of the extensive margin at different levels of development 29 Figure 9: Dominican Republic export performance 30 Figure 10: Growth orientation of products 32 Figure 11: Dominican Republic revealed comparative advantage 36 Figure 12: Export market concentration: herfindahl-hirshchman indexes 36 Figure 13 Benchmarking bilateral export relationships in Dominican 40 Figure 14:The Dominican Republic: Export sophistication and complexity 41 Figure 15: Technological composition of exports 42 Figure 16: US quality ladders for top 3 agricultural products 43 Figure 17:The Dominican Republic: survival rate of agricultural export spells to the US 47 Figure 18: Number of active hs codes as a share of total approved, by country and year 54 Figure 19: CAFTA-DR: Unit rejection rates 56 Figure 20: CAFTA-DR: Relative rejection rates 56 Figure 21: Relative rejections rates in the big four international markets 57 Figure 22: Top 10 refused products 57 Figure 23: Reasons for rejections 57 Figure 24: Overall growth in the SEZs 67 Figure 25: Free zone exports by industry 67 Figure 26: Apparel export to the US 71 Figure 27: Institutional architecture supporting trade policies in the Dominican Republic 79 Figure 28: EPA budgets per capita in LAC (in US$) 82 Figure 29: EPA budgets in LAC (in millions of US$) 82 Figure 30: EPA budgets per employee (in US$) 83 Figure 31:Total annual export value from firm-level data as a share of mirror comtrade total value (%) 92 Figure 32: Quality ladders for top 3 manufactured products 95 LIST OF TABLES Table 1: Summary of policy options aimed at improving the competitiveness of Dominican exporters 17 Table 2: Export performance by type of exporter 28 Table 3: Sectoral composition, revealed comparative advantage, and growth 31 Table 4: Sectoral composition of merchandise exports originated in SEZs 33 Table 5: Top 10 exported products in 2012 35 Table 6: Most important destinations 35 Table 7: Firm-level product and destination mix in exporting markets 39 Table 8: Relative quality of top 10 agricultural products exported to the US 44 Table 9: CAFTA countries: survival rate of agricultural export spells to the US 48 Table 10: List of approved commodities from the Dominican Republic 53 Table 11: Dominican Republic import alerts with detention without physical examination 55 Table 12: Number of Dominican exporters with border refusals 58 Table 13: Relative decline of the SEZ apparel industry, 1992-2012 64 Table 14: Complete decomposition of export growth along the margins of trade 94 Table 15: Characteristics of Dominican companies compared to the world and regions 102 Table 16: Characteristics of Dominican exporters compared to the world and regions 103 Table 17: Comparison of firm characteristics and perception of business environment in the Dominican Republic 104 Table 18: Foreign-owned firm characteristics and linkages in the Dominican Republic, compared to other regions 105 LIST OF BOXES Box 1: Dominican exports to Haiti 37 Box 2: Commitment to quality cocoa - The case of Conacado 45 Box 3: What are free zones? 63 Box 4: CAFTA-DR and free zones 66 Box 5: The history of medical device manufacturing in the Dominican Republic 68 Box 6: WTO compliance and SEZs 69 Box 7: Are Dominican SEZs less connected to domestic economy? 73 LIST OF ACRONYMS ADOEXPO Dominican Association of Exporters (Asociaci6n Dominicana de Exportadores) ADOZONA Dominican Association of Free Zones (Asociaci6n Dominicana deZonas Francas) AG EXPORT Guatemala Exporters Association ATC Agreement on Textiles and Clothing BEC Broad Economic Categories CAFTA-DR Dominican Republic- Central America Free Trade Agreement CAGR Compound Annual Growth Rate CARICOM Caribbean Community and Common Market CBTPA Caribbean Basin Trade Partnership Act CEI-RD Dominican Republic Export and Investment Center (Centro de Exportacion e Inversiones de RD) CEPIl French Research Center in International Economics CNC National Council of Competitiveness CNMSF-DR National Committee for the Application of Sanitary and Phytosanitary Measures CNZFE National Export Processing Zones Council (Centro Nacional de Zonas Francas Exportadores) CODOPESCA Dominican Council of Fisheries and Aquaculture COMTRADE UN International Trade Statistics Database CONACADO Dominican Republic Cocoa Exporters Confederation (Confederaci6n nacional de coca cultores dominicanos) CSR Corporate Social Responsibility DGA General Customs Bureau (Direcci6n General de Aduanas) DICOEX Department of Foreign Trade and Trade Agreements Administration DR Dominican Republic DWPE Detention without Physical Examination EPA Export Promotion Agency EU European Union EXPY Index of sophistication of a country's export basket FAVI R Fruits and Vegetables Import Requirements FDA US Food and Drug Administration FDI Foreign Direct Investment FTA Free Trade Agreements FTZs Free Trade Zones GAP Good Agricultural Practices GDP Gross Domestic Product GNP Gross National Product HACCP Hazard Analysis and Critical Control Point HOPE Haitian Hemispheric Opportunity through Partnership Encouragement Act HS Harmonized System Description and Codification of Merchandise IADB Inter-American Development Bank 1lBI Institute for Innovation in Biotechnology and Industry IPM Integrated Pest Management IRR Import Refusal Report ISO International Standards Organization ITBIS Tax on the Transfer of Industrialized Goods and Services JAD Dominican Agricultural Board LAC Latin America and the Caribbean LAVECEN Ministry of Agriculture Central Veterinary Laboratory LCR Latin America and the Caribbean Region LDC Least Developed Country MEPyD Ministry of Economics, Planning and Development MFA Multi-Fiber Arrangement MIT Ministry of Industry and Trade NGO Non-Governmental Organization NSF National Science Foundation PPP Purchasing Power Parity PRA Pest Risk Analysis PRODY Individual Product or Industry's Export Basket Sophistication PROMPEX Peruvian Commission for Export Promotion PVS Private Voluntary Standards RRR Relative Rejection Rate SANCO European Commission for Health and Food Safety SCM Subsidies and Countervailing Measures SEZs Special Economic Zones SMEs Small and Medium Enterprises (PYME) SPS Sanitary and Phytosanitary UNIDO United Nations Industrial Development Organization US United States USD United States Dollar USDA-APHIS U.S. Department of Agriculture Animal and Plant Health Inspection Service USAID United States Agency for International Development VAT Value-Added Tax WLD World WTO World Trade Organization The Dominican Republile Trade Competitiveness Diagnostic is the product of a World Bank team composed of Miguel Eduardo S6nchez Martin (Economist and Team Leader), Jos6 Daniel Reyes (Trade Economist, PRMTR), Swarnim Wagle (Se- nior Trade Economist, PRMTR), Aleksandra lwulska (Research Analyst, LCSPE), Rafael Van der Borght (Consultant, LCSPE), Lotta Moberg (Consultant), Marie Agnes Jouanjean (Consultant), and Cristian Ugarte (Consultant). Substantive inputs were provided by Cristian Quijadla Torres (Private Sector Development Specialist, LCSPF), Ana Margari- da Fernandes (Senior Economist, DECTI), Martha D. Pierola (Economist, DECTI), Jasmine Saerom Park (Consultant, DECTI) and Giselle Velasquez (Temporary, GMFDR). This report was prepared under the overall guidance and strategic direction of McDonald Benjamin (Dominican Republic World Bank Manager), Francisco Galrao Carneiro, (Lead Economist and Sector Leader, LCC3C), Marialisa Motta (Sector Manager, GTCDR) and Auguste Tano Kouame (Practice Manager, GMFDR). The peer reviewers are Maurizio Bussolo (Lead Economist, ECACE), Thomas Farole (Senior Economist, AFTP1), and Pavel Isa Contreras (External, Dominican academician). The Team is grateful for the collaboration of the following government counterparts and private sector representatives: Magdalena Lizardo and Juan Reyes (Ministry of Economy, Planning and Development); Rossy Fondeux, Wendy Adams, Karina Alc6ntara (Ministry of Industry and Trade); Ram6n Gonz6lez, Frank Fuentes Brito (Central Bank of Dominican Re- public); Fernando Fernandez, Gabino Jos6 Polanco, Maria Edelmira Marmolejos, Glan Matteo Flabane (General Customs Directorate, DGA); Augusto dle los Santos, Luis Miguel Gonzilez Arias (National Office for Statistics, ONE); Jean Alain Rodriguez, Rafael Samuel Sena, Maricell Silvestre (CEI-RD); Luisa Fernandez, Ebell de Castro (National Free Trade Zones Council, CNZFE); Andres Van der Horst, Laura del Castillo Savifi6n (National Competitiveness Council); Kai Schoenhals (Dominican Exporters Association, ADOEXPO); Aquiles Berm6dez (Dominican Free Zones Association, ADOZONA); Wilfre- do Ubri (Dominican Importers Association); William Malamud, Gustavo Mejfa Ricart (AMCHAMDR); Rafael Velasco Espall- lat (Ravez); Fernando Capell6n and Mercedes Capell6n dle Lama (Grupo M); Franklin G6mez (Conacado). The World Bank is indebted to the General Customs Directorate, the National Office of Statistics, and the National Council for Free Trade Zones for granting access to relevant databases that contributed significantly to this assessment. During the past decade, responding to changes in the al slowdown of 2008, results indicate that Dominican ex- global economy and international investment, the Do- ports are way below what would be expected at its stage minican Republic has transformed itself from a coun- of economic development. try dependent on resources and clothing production to one with a more diversified export base. The phas- To understand which factors could be preventing ing out of the Multi-fiber Agreement (MFA), completed at the Dominican Republic from fulfilling its potential the beginning of 2005, led to a decline in the textile-led in terms of merchandise exports, we have applied a exports. Once the MFA expired, the Dominican Republic Trade Competitiveness Diagnostics toolkit. Following was unable to compete with cheaper clothing from China, the methodology proposed by Reis and Farole (2012), we Hong Kong, Vietnam, and Bangladesh, and it lost much of first apply a Trade Outcomes Analysis, which helps den- its share in the US garment market. In the aftermath of the tify strengths and weaknesses in terms of exports perfor- CAFTA-DR, however, minerals, metals, and relatively more mance (growth and share), diversification in terms of prod- sophisticated manufacturing products such as medical ucts and markets, quality and sophistication, and entry devices, footwear, and pharmaceuticals have started to and survival. In the second part of the report, we analyze emerge. In 2010, the Haitian earthquake impacted the the weaknesses in more detail to better understand the Dominican Republic's export basket; shipments to its factors that are constraining export competitiveness. neighbor soared. E According to the diagnostic, export growth in the Spe- E However, the Dominican Republic still underperforms cial Economic Zones (SEZs) usually results from ex- similar countries in terms of trade openness and, more pansion of existing relationships; outside the zones, U precisely, share of exports per capita. In the face of the we find more dynamism in terms of new entrants and W recovery in export dynamism in the aftermath of the glob- exploration of new products and markets. The majority of export growth in the Dominican Republic's SEZs (89% America and the New Growth Poles (Brazil, China, Russia, in 2010-12) occurs by the development of established ex- etc.) are below potential. port relationships-the intensive margin of trade. This is quite common in these zones; the firms based are high- The sophistication of Dominican exports is relatively ly specialized and usually have well-established export high for a country of its level of development, but the channels that allow them to deepen export relationships. degree of complexity is intermediate', and issues of In the same period, a much larger proportion of export quality persist. Results indicate that exports are more'so- growth outside the SEZs is found on the extensive mar- phisticated'in the Dominican Republic than in peer coun- gin-36% for net entry of firms, 9% for net market di- tries, with the exception of Costa Rica. The degree of eco- versification, and 4% for new product diversification. nomic complexity is just below Costa Rica, Colombia, and However, it is worth noting that the higher dynamism El Salvador. It is worth noting the market duality in the ex- observed outside SEZs may be also due to lower survival ports basket-firms operating inside SEZs tend to export rates among companies, an issue we later discuss. Over- goods with some level of technological transformation, all, the analysis suggests that after the slowdown of 2009, while exporters operating outside of them focus on pri- Dominican exports have grown because of push factors mary and resource-based products. In terms of quality, the (increased productivity), but demand in partner countries main products exported from the SEZs seem to lag other has not contributed significantly. regional competitors, suggesting the Dominican Republic is not climbing the quality ladders as fast as other Central The Dominican Republic has been relatively success- American countries. When we look at the main agro ex- ful in diversifying products. As mentioned above, the ports, Dominican cigars exhibit high quality, but rum is Dominican Republic's comparative advantage in textiles perceived as of low quality compared to the products of progressively eroded over the past decade, but the coun- other CAFTA-DR countries. Over the past decade, Domin- try has experienced a noticeable expansion in exports of ican cocoa beans have seen a noticeable improvement in foodstuff, minerals, chemicals, and plastics-despite the quality, although room for further improvement remains. fact that the comparative advantage in these products is relatively small. The Dominican Republic has been also Finally,the survival rate of Dominican exporters seems able to build up comparative advantages in footwear and to be average for the CAFTA-DR area, although it is sig- medical equipment. These changes have been mainly nificantly lower among non-SEZ exporters. When we 0 driven by dynamics in SEZs, although the importance of analyze survival rates for agricultural exporters, the prob- non-SEZ exports has increased substantially over the past ability of a firm-product export relationship surviving the four years. first year of operations is only 42% among non-SEZ firms, compared to 53% for SEZ exporters. By the fifth year, these At the same time, we observe the persistence of a high probabilities have declined to 4.4% outside SEZs and 21% degree of concentration in destination markets. The US inside them. E 0 share of Dominican exports declined from 87% (2003-04) to 55% (2012). However, the United States and Haiti still In sum, the Trade Outcomes Analysis presented in part account for 70% of total exports, making the Dominican I suggests Dominican exports fare well in terms of per- Y Republic the least diversified among selected regional formance, sophistication, and survival in SEZs. Three peers (El Salvador, Jamaica, Costa Rica, and Colombia). This is caused by SEZs. Non-SEZ exports are much more diversi- is cuse bySEZ. Nn-SZ eport ar muh mre ivesi- 1 Sophisticated products are defined by Hausmann, Hwang and fied in terms of destination markets. Overall, analysis using Rodrik (2006) as those produced in advanced economies, whereas cmplex products would be those with higher levels of embedded a gravity model suggests that Dominican exports to South knowledge (Hausmann et al, 2012). e Structure of the report Part 1: Trade Outcomes Analysis. Dimensions analyzed Diversification Entry and Survival Grothn d are (Extensive margin) Sophistication (=) similar to CAFTA (+) in products and Quality ( ( low in national regime (-) in markets Part II: Trade Competitiveness Diagnostics. Challenges identified c) High concentration b) Two-tier export basket a) High level of rejection faced on US and Haiti (duality SEZs) by DR agro exports Proposed policy options National promotion Rm v prilgsIncreased information capacity strategy exports Hrznapoies .and control Agency with more capacity (EPA) Ade au i gae evcs Support for the entire value chain Source: authors'elaboration. main challenges are identified: quality issues, the ex- linkages, technology transfer, demonstration effects, etc.). port basket's duality, and excessive concentration in At the same time, SEZs present a notable fiscal cost for the terms of markets. Dominican Republic, with tax exemptions estimated to be close to one percent of GDP. First, the perceived quality of certain agricultural products is lower than regional competitors. Even if the Dominican Third, the Dominican Republic has been relatively success- Republic is well-positioned to take advantage of favor- ful in diversifying export products, but diversification of able US market access, the exploitation of these oppor- destination markets has been more limited. The heavy re- tunities remains conditional on the capacity to comply liance on the US arises mainly because the majority of SF7 with the US market's stringent sanitary and phytosanitary production is directed to the US market. Preliminary find- standards (SPS). ings suggest that the Dominican Republic could expand exports to China and Brazil by exploring sectors outside Second, the future of free zones will depend on how their the extractive industries-in particular, pharmaceuticals, accumulated strengths are leveraged in a new era of com- plastic products, and medical equipment. petitiveness built on fundamentals rather than transient policy distortions. As already noted, a duality characteriz- Policy options aiming at improving the outcomes in these es the Dominican export basket: SEZs ship products with a three areas are discussed in the remainder of this section certain degree of transformation, while non-SEZ exporters and summarized in Table 1 at the end of this Executive focus on primary and resource-based products.The poten- Summary. E tial problem for the Dominican Republic is that SEZs may constitute "enclaves" that are relatively isolated from the The Competitiveness Diagnostic in part 2 discusses in rest of the economy, reducing the potential for generating more detail some of the constraints and bottlenecks positive externalities that could benefit non-SEZ exporters hampering the competitiveness of Dominican exports, CU and other domestic companies (backward and forward defining a series of challenges: tions would be missed. In the short run, as SEZs adapt to the end of export performance-based subsidies, authori- First, as discussed in section L.b, the Dominican Repub- ties could try to set up safeguards to prevent a transition lic has developed a Two-Tier Export Basket in terms of markedly influenced by entrepreneurial groups and insti- Sophistication and Quality (Figure 15). Exporters from tutions strongly interested in preserving the status quo. outside SEZs basically sell minerals, agricultural goods, Giving into these forces could endanger compliance with and other primary products, while the more sophisticat- the WTO and is unlikely to bring the dynamism needed to ed transformation processes take place in SEZs (clothing, build export competitiveness. shoes, medical instruments, plastics, etc.). This duality is likely to persist or even to become more pronounced be- In the medium and long term, Dominican stakehold- cause SEZ companies have weak linkages to the rest of the ers could consider the progressive homogenization economy, limiting the potential for technology transmis- of benefits offered to companies inside and outside sion, learning by doing, and other positive externalities SEZs, which would help level the playing field, reduc- that could benefit domestic companies. ing distortions. In the medium term, the Dominican Re- public would need to progressively reduce the number of To comply with WTO rules, the Dominican government "strategic sectors;' perhaps temporarily maintaining those is considering to grant strategic industries benefits with notable value added and the potential to become similar to those currently enjoyed by SEZ companies, growth engines for the economy. According to Hausmann most likely resulting in a missed opportunity to ad- et al. (2011), for example, the cluster of furniture, domes- dress the duality inside and outside SEZs. WTO rules tic appliances, and construction materials entails relative- require countries to phase out subsidies contingent on ex- ly sophisticated activities that are close to the Dominican port performance in 2015 (Box 6).To comply, the Domini- Republic's current productive structure-but it remains can Republic is considering to declare most of the sectors unexploited. currently producing in SEZs to be"strategic"-a pragmat- ic approach to prevent rapid changes in the SEZ sectors. In the long run, these"strategic sectors"would eventually A few tentative steps have already been taken towards a not be longer needed, and the country should be moving strategic sector-based model. In 2007, textile, shoe, and toward policies that are "horizontal" in nature and would leather companies all obtained SEZ-equivalent benefits benefit all kind of exporters. Overall, exporters outside 0 because they were singled out as "strategic;' eligible for and inside SFZs should be receiving support from the benefits regardless of their location (inside or outside export promotion agency (FPA) in the form of commer- SEZs) or export performance. Abolishing the minimum cial actions, technical assistance, identification of export export requirement for SEZ companies has been another opportunities, etc. At some point, as tax exemptions are important step. However, since the current legal frame- reduced, the country could be able to further boost ex- C. work in the DR still favors exports over domestic sales, this porters'competitiveness by lowering and homogenizing E 0 is unlikely to be sufficient for the Dominican Republic to corporate taxes and import tariffs (as has been done in comply with WTO rules. Should the Dominican Republic China and Mauritius, among others). It is duly acknowl- continue introducing changes in the legal framework to edged that following this path would require from a long fully opt for a "strategic sector" approach, it would be im- and sustained process of national dialogue, leading to an a portant to engage in multi-stakeholder discussions for the agreement among public and private sector actors on C: definition of strategic sectors, in order to prevent a gener- the strategy to be adopted. It may be argued that reduc- alization of exemptions that would result in an important ing incentives would result in missing opportunities for 0 fiscal cost. The opportunity to streamline existing distor- attracting FDI, multinationals abandoning the DR, and problems to compete with other countries in Central and South America. To avoid undesired outcomes, a first step would be conducting a thorough cost-benefit analysis of Second, the problems associated to this dual struc- Special Economic Zones, as well as surveys to companies ture are patent in the agricultural sector, with surviv- established in SEZs, to understand their strategic motiva- al rates much lowerfor non-SEZ exporters (Figure 17). tions and inform the debate. This is likely to be the result of limited exchange of in- formation between companies inside and outside SEZs, Only by climbing up the value chain and increasing insufficient institutional support for non-SEZ exporters, quality and productivity will Dominican SEZ export- and an overall lack of capacity to comply with US SPS ers be able to remain globally competitive in in the measures. The Trade Outcomes Analysis shows that the medium and long term. It has been argued that special perceived quality of Dominican agro exports is mediocre benefits and exemptions are needed for competitive- (Figure 16), and that rejection rates at the US border are ness because Dominican SEZ exporters can no longer higher for Dominican fruits and vegetables than those compete on the basis of low wages. However, the com- from other CAFTA countries.The Dominican Republic has petitive advantages derived from low tariffs, fiscal incen- historically received favorable US market access due to tives, and wage restraint were not enough to prevent strong economic ties and favorable CAFTA-DR trading the decline of the textile sector in the 2000s. This sug- rules; however, its ability to comply with US SPS regula- gests it would be crucial to begin building fundamental tion is low. In addition, evidence suggests that Domini- competitiveness, rather than relying on transient prefer- can products may be subject to more frequent and strin- ential taxes or tariffs. For this to happen, the institutional gent inspections due to a past compliance difficulties architecture supporting Dominican trade and compet- (Jouanjean et al, 2012). Unless challenges in this area are itiveness policies might need to be adjusted to better tackled, the country is unlikely to fulfill its potential for support exporters and attract more investments. This agricultural exports. can be done by adding value to the products through vertical integration, offering "bundles" of manufactur- This calls for a revision of the national capacity to meet ing and services activities to foreign investors willing to international standards in agro and food products, locate in SEZs, and betting on innovation. Examples of including improvements in physical and institutional leading Dominican companies that have gone through infrastructure, better coordination between interna- some of these processes, and are discussed in this re- tional bodies and the private sector, and increased port, are Grupo M and Conacado. transparency along the value chain. There is a need to create standardization bodies, system certifications, con- Finally, to foster backward linkages from SEZs to the formity assessments, and accreditation mechanisms. Ex- rest of the economy, the EPA can support capacity isting institutions in charge of the surveillance of animal building in local producers. Policies to increase local health and control of plant pests could also benefit from firms'volume and quality will help them become reliable capacity building. In addition, vaccination campaigns and suppliers to companies in SEZs. In addition, efforts could wider dissemination of risk-mitigation methodologies focus on attracting FDI in sectors more likely to bring could be organized. positive externalities to the rest of the economy-metals, machinery, and other manufacturing, among others. More The Dominican Republic also needs to expand its physical research would be needed to understand the current na- infrastructure, including refurbishing existing laboratories ture and status of those linkages and better tailor policies to enhance analytical testing capabilities and creating U aimed at strengthen them, new inspection warehouses at several border points. al rae4uhlwrfrnnSZepres(iue1) The relevant Dominican agencies should participate in in- entrepreneur support efforts. The National Council for ternational SPS standard setting institutions as well as the Free Trade Zones and CEI-RD are de facto in charge of WTO SPS committee. This would ensure that the Domini- investment attraction inside and outside SEZs. A similar can support infrastructure conforms international norms duality is found in private exporter associations. In spite and prompts updates of regulations on SPS and pesticides, of commendable coordination efforts, duplicities are helping the country doing a better job of preventing entry likely to appear, and achieving economies of scale in sup- of banned products and reducing the risk of unintended porting exporters would be more difficult. The overlaps inter-crop contamination. If they have not already done among the different institutions in which the Ministry of so, authorities may want to create and frequently update a Industry and Trade participates are also likely to result in a list of banned pesticides and enact a law setting maximum suboptimal use of public resources (Figure 27). Legal attn- residue levels for fruits and vegetables. butes and responsibilities among the institutions do not seem clear. Finally, a set of actions could be aimed at supporting producers and exporters associations along the val- Under its current structure, CEI-RD may not be suffi- ue chain, as well as at improving the coordination of ciently empowered to efficiently perform its mandate public and private stakeholders. In Peru and Guate- to promote exports and attract investment. Its budget mala, export promotion agencies have been essential to per capita slightly lags the median for Latin America. How- promoting good agricultural practices, by ensuring the ever, the budget per employee is the lowest in Latin Amer- transmission of information between producers and im- ica,indicatinga heavy staffing burden.This limits space for porters. The EPAs provide information to producers on commercial support and technical assistance as well as its quality requirements and promote quality products in ex- presence abroad. Most Dominican trade representatives port markets.They are also essential for implementation of abroad are in fact diplomats belonging to the Ministry of public-private partnerships and supporting regulations. In Foreign Affairs who devote only part of their time to ex- the Dominican Republic, CEI-RD has been initiating similar port promotion. actions on ad-hoc basis,2 and this kind of support could be systemized and extended to diverse agricultural products. An institutionally empowered EPA might be an option A related policy option involves carefully supporting pri- for promoting export competitiveness and market di- vate certifications schemes. Experience by various donors versification: it is fully consistent and complementary 0 shows that it is important to adapt the certification need- to the previously identified policy options. First, this .2 ed to the market channel and the specificities of demand EPA would need to address significant information short- in the destination market. ages and focus primarily on increasing Dominican exports to previously untapped markets where exports seem be- low potential, such as the South American countries and the New Growth Poles. The more important trade barriers2 0 are, the more efficient the EPA will be. Second, this EPA Third, as discussed in section Ill.c, The institutional would need to offer bundled services and a tailored ap- infrastructure supporting international trade, also proach to ensure efficiency. As noted above, EPAs'support presents a marked duality and fragmentation, which of agricultural producers by promoting good agricultural a is likely to result in suboptimal export promotion and practices and increasing the capacity to comply with SPS C: norms has proven efficient in Peru and Guatemala. In the 2 CEI-RD has been accompanying producers of red pepper to up- Dominican Republic, EPA's support in increasing agricul- grade their production process and ensure compliance with SPS 0 norms. tural exporters' capacity to comply with SPS norms might be very useful because of the relatively low quality of cer- supply, continue improving the quality of education at all tain agricultural products. Third, counting with a unified levels and provide technical formation that is suitable to and strenghtened EPA promoting exports and offering the needs of exporters and manufacturers. (iii) Continue efficient support to all Dominican companies may help improving several aspects of the business framework, in- in shifting from discriminatory fiscal policies and "special cluding burdensome procedures that result in informality regimes" toward more horizontal support that offers gen- and perceived unfair competition, and strengthen the for- eral, stable, and predictable assistance. This would include eign investor protection framework, both inside and out- bundled services consisting of several actions and a tai- side SEZs. In addition, according to private sector actors, a lored approach with help in forming consortia of SMEs in higher real exchange rate (compared to the nineties) may exporting sectors, financing activities, and identifying po- be also a drag for export competitiveness. tential buyers and partners. Further research will be needed to better understand To reach that point, it will be necessary to clarify the how to achieve a higher growth potential by increas- role of the different institutions dealing with export ing the linkages between SEZ companies and the rest promotion, streamline support processes and reduc- of the Dominican economy. We have been able to estab- ing overlaps. A first step in that direction would be a di- lish on a preliminary basis the lack of backward linkages alogue among private and public stakeholders to achieve between SEZs and foreign-owned companies and the rest a national strategy for promoting exports and attracting of the economy (Box 7). However, we are far from under- investment.This strategy would facilitate the alignment of standing the nature of existing linkages and the position of strategic plans, the setting of rules to coordinate attracting the Dominican Republic in global value chains, We would investment (reducing the scope for favoritism), and the like to better understand the characteristics of Dominican defining of support actions. Moreover, the strategy would exporters, comparing them to non-exporting firms, as well need to identify how authorities are planning to move as the determinants of successful exporters (those with from generalized fiscal incentives in "strategic sectors" to- longer survival rates). A follow-up analysis based on en- ward more "horizontal" and non-discriminatory schemes terprise surveys and a deeper look at companies that are of support, ultimately eliminating the existing duality be- both exporters and importers would be needed to address tween exporters inside and outside SEZs. Policies to pro- these pending questions. mote productivity and encourage innovation would need to be part of that strategy. In conclusion, the Dominican Republic is at a junc- ture where inter-institutional and multi-stakeholder Finally, it is worth noting that Dominican exporters discussions are needed to benefit from an opportu- perceive unreliable electricity, relatively complicated nity to reform incentives and infrastructure to sup- procedures to pay taxes, informality, and the limited port exporters and fulfill foreign trade potential. The availability of skilled workers as constraints on their Caribbean Growth Forum initiative3 may constitute the ability to operate. Although a deep assessment of these ideal space for debate among public and private actors. hurdles is beyond the scope of this report, some policy Drawing from the dialogue, Dominican authorities could options emanating from recent research and country di- try to formulate a national strategic plan for promoting E alogue are: (i) In the electricity sector, reduce the price exports and attracting investment, which would help E of generation by minimizing arrears and diversifying the coordinate efforts in these areas and make them more energy mix, while continuing to improve efficiency and efficient. This plan could envisage a series of short- and enhance monitoring and evaluation on transmission and p 3 For more information, access http://caribgrowth.competecarib- tdistribution (World Bank, 2014). (ii) On qualified labor bean.org. medium-term actions aimed at maximizing the potential of policies in a moment of transition. The discussion of the agrarian sectors by increasing the capacity of Do- on SPS compliance in fruits at vegetables will be used to minican exporters to meet international SPS standards. shape a World Bank loan aimed at increasing efficiency in In addition, the plan could contemplate the transition of agricultural value chains, currently under formulation. In SEZs in a way that reduces the dual treatment of com- addition, this report is expected to contribute to ongoing panies outside and inside SEZs, and moves the Domin- discussion about WTO compliance and a new customs ican Republic into compliance with WTO rules. In the law because it highlights the limitations of the current long run, support could be granted through a series of SEZ regime and the competitiveness constraints suffered non-discriminative"horizontal" policies implemented by by non-SEZ exporters. Moreover, the report signals how a better empowered EPA. Dominican exporters could explore diversification into new markets (China, Brazil) that demand certain products This Trade Competitiveness Diagnostic expects to fa- currently produced by the Dominican Republic (iron and cilitate well-informed decisions and better targeting steel, medical equipment). Diagnosis area CalnePolicy solution a) Survival rates of agricultural Increase capacity to comply with Building capacity of institutions for the surveillance and exports similar to other CAFTA-DR SPS measures inspection of plant pests and animal health, including countries, but lower perceived creation of standardization bodies, system certifications, quality and higher levels of rejection conformity assessments. Organize vaccination campaigns at the US border and disseminate risk mitigation methodologies Improve laboratory infrastructure, analytical testing capabilities, create new inspection warehouses at entry points (terrestrial borders, ports, or airports) Ensure participation of Dominican SPS standard-setting institutions in the WTO SPS committee to quickly align national standards with international norms 0 Create list of banned pesticides and enact regulations dealing with maximum residue levels Survival rates for agricultural Supporting producers and exporters associations along exporters are significantly lower the value chain of the different agro products. The Export outside than inside SEZs Promotion Agency can be very useful to facilitate the w transmission of information both ways, linking producers E and importers and to set public-private partnerships in the sector Carefully target support for private certifications schemes v to encourage them to adapt to market channel and demand C. O 0 Diagnosis area halnePolicy solution b) A duality is observed: SEZs export Dominican SEZ exporters can no To remain competitive in the current global context, products with a certain degree of longer compete in the basis of low Dominican exporters need to continue adding value to transformation, whereas non-SEZ wages their products by vertically integrating new processes, exporters focus on primary and offering "bundles"to foreign investors, innovating, trying resource-based products to climb the value chain, and working to increase the quality and productivity Enhance backward linkages Promote backward linkages between foreign companies between SEZs and the rest of the in SEZs and local suppliers by introducing incentives economy to attract foreign investors to sectors that rely more on domestic inputs (metals, machinery, and other manufacturing); in addition, the EPA and other relevant institutions have a role in enhancing the capacity and capabilities of local producers, so they become reliable SEZ suppliers Seize the opportunity of the need In the short run, enact a series of safeguards to prevent to comply with WTO ban on export export-promoting policies from being captured by anti- subsidies to pursue country-wide reform coalitions reform Conduct a cost-benefit analysis of SEZs, considering job creation, contribution to growth, and fiscal costs On the basis of the cost-benefit analysis, in the medium term, consider combining nationwide export promotion policies (regardless of the export regime) with enhanced selectivity in terms of strategic sectors (only a couple of growth engines) In the long run, progressively level the playing field between strategic and non-strategic sectors, with lower taxes and import tariffs extended to non-strategic companies; enact export promotion policies that are "horizontal" in nature c) Recently, the Dominican Republic Most of the commercial agents Clarify the roles of the various institutions dealing with has been relatively successful in abroad are diplomats belonging export promotion to avoid overlaps and reduce demanding exporting new products, but not in to the Ministry of Foreign coordination needs under the actual structure, ultimately selling to new destination markets affairs, while specialized trade enhancing the efficiency of the export promotion budget promotion officers would be more and activities effective; CEI-RD has a budget slightly below the Latin American average, but the payroll burden leaves it with scarce resources compared to other EPAs (U W X LU Diagnosis area halnePolicy solution c), Reety theDoinca c) Recently, the Dominican Export promotion efforts are dispersed Enact a national strategy for promoting exports and Republic has been relatively across multiple stakeholders, which may attracting investment that aligns existing strategic successful in exporting new decrease the efficiency of resources and plans, establishes standards in attracting investment products, but not in selling to the effectiveness of actions (avoiding arbitrary benefits and unequal treatment), new destination markets and signals key sectors for export promotion (instead of generalized support). Include specific policies to promote productivity and innovation Promote export market destination Foster a well institutionally empowered EPA, which diversification to reduce reliance on the would focus on (i) addressing significant information US market shortages to increase exports to relatively unexplored markets in South America and among the New Growth Poles; (ii) increasing the capacity of agricultural producers to comply with SPS norms; (iii) offering bundled services and a tailored support approach with help in forming consortia of SMEs in exporting sectors, financing activities, and identifying potential buyers Use difference-in-difference and matching methodologies (Volpe Martincus, 2010) to assess the effects of current EPA activities (promotion, identification of partners, technical assistance, etc.) on companies that benefited from support, comparing them to those that did not benefit d) Other obstacles in the Relatively expensive and unreliable Reduce the price of generation by minimizing arrears business environment electricity supply hampers the and diversifying the energy mix; continue improving competitiveness of Dominican exporters efficiency in distribution (World Bank, 2014) The limited availability of skilled human Continue improving the quality of education at all levels capital may be a constraint in the future and provide technical formation suitable to the needs of exporters and manufacturers Y 0 The Dominican Republic has not made Continue working in facilitating several aspects of the significant gains in the Doing Business business framework, including guarantees over movable rankings lately, which may negatively assets, investor protection, insolvency regulation, and affect the perception of potential procedures to pay taxes foreign investors and the operations of domestic firms Source: World Bank staff elaboration.oa 0 U -S a 0 0 Although the Dominican Republic underwent a suc- In the past decade, the country saw a change in for- cessful structural transformation over the past five tune. The phasing-out of the Multi-Fiber Arrangement decades, high growth did not translate into poverty (MFA), completed at the beginning of 2005, resulted in the reduction. Beginning in the late 1960s, the Dominican decline of Dominican exports. For 30 years, the MFA had Republic started to diversify its economy, bringing to an regulated global trade in textile and apparel products. It end centuries of dependence on agriculture, especially was notable mainly for permitting major apparel import- sugar production. By the 1980s, the country's economic ers, such as Canada, the US, and the European Union, to focus had shifted to mining, tourism, and manufacturing. set quotas on the amount of foreign-made apparel and Successive governments helped build up the country's textiles they would import from any specific producing manufacturing sector by creating Special Economic Zones country. The main impact of these import quotas was to (SEZs) or Free Trade Zones, which offered tariff exemptions restrict the amount of apparel from China and some other and a series of tax concessions to foreign companies. The countries in the Far East, allowing the signing countries, main activity in the SEZs was garment assembly for export such as the Dominican Republic, to meet the unfulfilled mainly to the market. The dynamism of this sector boost- demand. Once the MFA expired, the Dominican Republic 0 ed exports and helped the Dominican Republic to achieve was unable to compete with the cheaper clothing from impressive GDP growth rates in the 1990s and early 2000s. China, Hong Kong, Vietnam, and Bangladesh, and it lost 0 After that, economic growth continued, but it was no lon- much of its share in the US garment market. Between 2004 ger led by exports.4 and 2006, US garment imports from China surged by 65%, c and in 2006 China accounted for 30% of the US market, 0 with other Asian countries accounting for an additional 4 GDP growth averaged 5.7 percent per year between 1991 and 25%. In 2000, the Dominican Republic had a 4% share of 2013, making the DR one of the top performers in the Latin America and Caribbean region (LCR). US garment imports; by 2006, it had fallen to 1.6%. Deteriorating terms of trade and overvaluation re- cussed above, the sharp contraction in textile exports over spect to the nineties may have also impacted the com- the past decade led to a decline in SEZ economic activity petitiveness of Dominican exports. The financial sec- and employment. In the aftermath of the global economic tor crisis in the Dominican Republic resulted in inflation slowdown of 2009, SEZs have shown signs of revival. They reaching 42 percentage points in 2003, which was tem- have been able to expand production of a more diversified porarily compensated with a sharp depreciation that re- bundle of goods, including surgical equipment, shoes, and suIted in a rebound in economic growth. However, in 2005 plastics and chemicals. In addition, some companies have real exchange rate levels were already higher than in the initiated vertical integration processes, or have started to pre-crisis situation, and have since remained much higher offer complementary services, to increase their products' than in the nineties. Private sector actors in the Dominican value added. Furthermore, non-SEZ exports have shown a Republic often quote exchange rate overvaluation as one markedly greater dynamism in recent years, thanks to the of the main causes behind the decline in export compet- vitality of agricultural exports and the beginning of gold itiveness. By looking at the terms of trade index, it is pos- export activity. Despite reasons for optimism, the Domin- sible to observe certain deterioration since 2002, meaning ican Republic continues to lag Costa Rica, Honduras and that the unit value of exports is on decline if we compare other DR-CAFTA countries and other small islands in ex- with the unit value of imports. port growth. Authorities reacted to this new external environment SEZs have been often characterized as "enclaves; and by joining ongoing negotiations to a free-trade agree- their relative isolation from the rest of the economy ment between the Central American region and the could be one explanation for the limited inclusive- US, but the Dominican foreign trade remains below ness of the Dominican growth model. Research on the potential.The Dominican Republic was a latecomer to the Dominican Republic's economic development has often Central America Free Trade Agreement (CAFTA), formally highlighted-but not thoroughly proven-the lack of the accepted in 2004. The agreement was renamed CAFTA- backward linkages between SEZs and the rest of the econ- DR, and it entered into force in the Dominican Republic in omy (Kaplinsky, 1993; Willmore, 1995; Snchez-Ancoe- 2007, one year later than for the other member countries.' chea, 2006). The result is limited opportunities for technol- In addition to this agreement, the Dominican Republic ogy and knowledge transfer between firms outside and had also signed trade pacts with the Caribbean commu- inside SEZs, constraining the opportunities for upgrading 0 nity (CARICOM) in 1998 and with Panama in 2003. Since of Dominican production processes. As discussed in sec- .2 October 2008, it has been a member of the Economic tion l.c,, even when SEZs have been successful in gener- Partnership Agreement (EPA) between the EU and the Ca- ating employment and economic growth, they have also ribbean. In spite of tariff declines, Dominican exports and originated a two-tier exports basket as well as dual stan- imports asa share of GDP is lower than in other economies ards in taxation and labor markets. It seems legitimate with the same level of income per capita, suggesting inter- to question the current role of SEZs, and how they should E 0 national trade is below potential. evolve in the future, especially now that the country faces a December 2015 deadline to comply with WTO rules by Exports have recovered some dynamism after 2009, phasing out of export subsidies (see section l.b.2). U and the Dominican Republic has been able to diversify a in terms of products, although the export value in real Together with SEZs, tourism has been since the nine- C: terms remains far below the level of the 1990s. As dis- ties the other engine for the Dominican economy. At 5 With the exception of Costa Rica, where CAFTA-DR entered into a effect in 2009. of travel services remain as an essential component in Evolution of terms of trade and real Evolution of SEZ and non-SEZ exchange rate in the Dominican Republic real exports in Dominican Republic 120 3000 100 2500 80 2000 60 1500 40 1000 20 500 01990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 01993 1995 1997 1999 2001 2003 2005 2007 2009 2011* -- Terms of trade (2000=100) N National territories Free trade zones Exchange Rate Index (REER, 2000=100) *Indicate partially revised figures and preliminary figures subject to rectification Source: World Development Indicators. Source: WB staff calculations based on Central Bank data. the balance of payments, to compensate for a wide trade counterparts in the Ministry of Industry and Trade, and deficit (averaging 14.9 percent in the same period). They the Ministry of Economy, Planning and Development have also constitute an important source of foreign currency, expressed their concerns about the lack of solid analytical which helps mitigating exchange rate fluctuations and work to guide public policies on trade issues. This report ensuring external sustainability. However, it is worth no- tries to contribute to filling this gap by applying the Trade ticing that travel receipts have declined in relative terms, Competitiveness Diagnostic methodology (Reis and Farole, if we compare to the pre-crisis period (as they averaged 2012) to systematically assess Dominican goods exports. 11.9 percent of GDP in 2005-08). As we will see, this re- port does mainly focuses in exports of goods, given that In light of this history and motivation, this report's the customs transaction database we analyze does not objective is to thoroughly assess export performance, record exports of services; nevertheless, it is important to identify the main bottlenecks for export competitive- acknowledge the importance of tourism in the Domini- ness, and come up with suitable policy options to ad- can economy, as well as to emphasize the need for further dress them. In pursuit of this objective, the report tries to research about the topic. answer these questions: How did export of goods evolve in the last decade and how did the Dominican Republic corn- The idea of conducting a Trade Competitiveness Di- pare to other LAC countries? Were changes in exports driven agnostic originally emerged during a series of e-semi- by intensive or extensive margins? To which extent is export nars on Free Trade Agreements (FTAs) held in Septem- concentration a source of vulnerability? Is there space to ber 2011.6 In a series of follow-up meetings, government improve trade relationships with other countries? How has 0 trade with the new growth poles (BRICS) evolved? What is the 0 6 During the Annual Meetings 2010, President Fernandez, in dis- o cussion with President Zoellick, voiced his interest in learning from presents the lowest rate ofexport survival? international experiences in FTAs, to take better advantage of the existing CAFTA-DR. Following this request, the World Bank organized two e-conferences called "International Experiences on Free Trade 0 Agreements, Lessons for the DR'"They took place in September 2011, connecting experts in Santo Domingo, Washington, and Singapore, including the CEO of the Chamber of Commerce of Singapore and two Dominican ministers.The case of Singapore, successfully becom- agnostic. Part I assesses Dominican Republic trade out- ing an Asian hub for biotech and pharmaceutical industries thanks to theUSFTA,wasbroughtup,alongwiththeexampleofChile. comes over the pasttwo decades, compared to its CAFTA peers and some other international peers. Three main Finally, we acknowledge the importance of other challenges emerge from the Trade Outcomes Analysis: factors that may have a bearing on export compet- low compliance with international norms in agricultural itiveness but, for different reasons, are beyond the exports, the economic duality emerging from SEZs, and scope of this research. As abovementioned, the report the fragmented institutional support for exports. These focuses on three bottlenecks for export competitive- challenges have also been highlighted as topics of interest ness: agro exporters' difficulty in complying with SPS that, to our knowledge, have not been thoroughly inves- standards, a two-tier export basket with limited link- tigated so far. They are analyzed in detail the in the Trade ages between SEZs and the domestic economy, and Competitiveness Diagnostic section. fragmented institutional support efforts. In addition, other factors such as the quality and cost of produc- Section li.a of the Trade Competitiveness Diagnostic tion inputs (affected by energy prices, taxes or tariffs), studies the capacity of Dominican exporters to comply the factors of production (functioning of capital and with US sanitary and phytosanitary (SPS) measures in labor market) and available technology (often import- thefruits and vegetables sector. It first provides an over- ed foreign technology) are also likely to influence the view of the US market access policies and reviews the cur- competitiveness of Dominican exports. A detailed rent provisions specific to the Dominican Republic. It then analysis of these factors could not be included in this employs US Food and Drug Administration import alerts assessment, since some of them entail questions well and import rejection information to track Dominican trade beyond a trade competitiveness diagnostic (electricity compliance and benchmark it against other CAFTA-DR sector and labor market factors), and others have been countries. Looking at compliance is crucial because the in- already covered in a number of studies by the IADB ability of exporters to meet developed markets'stringent and USAID (transport and logistics). Preliminary anal- SPS measures often hinders their export survival and ham- ysis using World Bank Enterprise Surveys reveals elec- pers their integration into the global economy. tricity, tax administration, informality, and insufficient availability of skilled labor as perceived by Dominican Section li.b looks at the current competitiveness of exporters to be more of an obstacle than in other parts Dominican SEZs and their broader achievements and of the world. Results from this preliminary analysis are failures to date. It diagnoses key factors that might be presented in more detail in Annex 6, Dominican Export- associated with these successes and failures, to finally ers-Results from the Enterprise Survey. 0 suggest a forward-looking policy course, drawing partly on cross-country international experience of maturing or The report focuses on 2002-12 and uses General graduating SEZs in the region and beyond. It is also worth Customs Directorate merchandise export transac- noting that we also include some preliminary evidence tion database as its main source. The export trans- about backward and forward linkages between foreign action database is thoroughly assessed in part I Trade C. owned companies and the rest of the economy (Box 7). Outcomes Analysis. Section l.a employs newly available E 0 data on US import refusals to determine the Dominican Section li.c assesses the Dominican Republic's institu- agricultural exports'rejection rate by US border author- tional infrastructure to support export promotion and ities. Section Il.b complements the analysis on the role attracting investment. First, it analyzes existing institu- of SEZs by looking at Central Bank and CNZFE data and a tions and bodies in charge of these functions as well as their by drawing from in-depth interviews with stakeholders. C: interactions. Second, it discusses how export promotion Section Il.c, put together in collaboration with CEI-RD agencies in the Dominican Republic join efforts to achieve officials, discusses the role of export promotion agen- 0 moreefficientoutcomesand useofpublicresources. cies in lightof regional and international experience. The main audience for the report will be public and private stakeholders in the foreign trade area. As men- tioned above, the idea for this report came out of discus- sions with the Ministry of Economy, Planning and Develop- ment and the Directorate for Foreign Trade at the Ministry of Industry and Commerce. In addition, the audience for this report would include other public sector bodies (CEI- RD, CNZFE, the National Competitiveness Council), private sector associations (ADOZONA, the Dominican Associa- tion of Importers, the Dominican Association of Exporters, etc.), entrepreneurs, civil society organizations, and mem- bers of academia. The Trade Competitiveness Diagnostic is one of the first building blocks in developing a new agenda of inclusive growth and enhanced competitiveness in the Dominican Republic. This document has been developed in the context of country priorities established by the Na- tional Development Strategy 2010-2030 (Strategic Axis 3, "Desarrollo Econ6mico y Competitividad"). It also re- sponds to Objective 2 under the new World Bank Country Partnership Strategy with the Dominican Republic-"im- proving the investment climate and fostering private sector development;' aimed at achieving sustained and inclusive growth.This analytical piece accompanies ongo- ing World Bank technical assistance efforts aimed at im- proving the Dominican business climate. In addition, the results of this report, especially section Il.a, with title The Dominican Republic's Trade Compliance Capacity in Fruits and Vegetables, are going to contribute to the design of an upcoming World Bank operation aimed at enhancing value chains in the agricultural sector. 0 0 0 2 7 rai a -2o h atoa eeomn taey CHAPTER I This chapter reviews the country and firm level perfor- outside the zones are mostly primary or resourced based, mance of Dominican exports along various dimensions with low value added and low quality in international that, taken together, give a fairly comprehensive pic- markets. Section d looks at survival rates of agricultural ture of trade competitiveness. The analysis is intended exporters and finds them low but no different from other to guide a systematic generation of hypotheses about the CAFTA countries. Finally, section e summarizes the main country's export performance, prospects, and challenges. challenges to export competitiveness. During the analysis, we benchmark the Dominica Republic with other CAFTA countries and, in some cases, with se- Through the analysis, we are able to identify three 0 lected peers outside the region. The chapter is organized main challenges to export competitiveness. First, the as follows. Section a reviews the evolution of export per- survival rates for Dominican agricultural exports are not formance and the role of SEZs in shaping export dynam- different from other CAFTA countries, but the perceived ics over the past decade. Section b discusses the changes quality of certain products is lower, providing room for in sector specialization caused by changes in preferential potential for enhancing quality. Second, there is a duality access in the US market, the Dominican Republic's main in sophistication of exported products: non-SEZ export- E 0 export destination. It also highlights the narrow product ers focus on primary and resource-based products; SEZs and destination mix of Dominican exporters. Section c re- export products with a certain degree of transformation, veals the existence a two-tier export basket in terms of so- becoming "enclaves" with limited positive externalities for phistication and quality: products from SEZ firms involve the rest of the economy. Third, the Dominican Republic some technological transformation, while products from has been relatively successful in diversifying export prod- ucts, but it has done less well in finding new destination markets. These challenges are assessed in more detail in 8 The Trade Outcomes Analysis was prepared by Jos6 Daniel Reyes, 0 with contributions by Cristian Ugarte. part of this report, Trade Competitiveness Diagnostic. Dominican Republic: exports, imports, Export performance between and trade balance (Billions of USD) 2000-12 (2000=1) 15 MFA CAFTA-DR expires enters into effect 10 8.6 9.14 5.7 6.7 5.5 6. 5 o3 -5 2 -10 9 12 -15 201 151 82000 2002 2004 2006 2008 2010 2012 -20 1 5 1 200020012002 2003 20042005 2006 2007 2008 2009 201020112012 - Dominican Rep. - Colombia El Salvador * Exports Imports - Trade Balance Mauritius Costa Rica Honduras Note: This figure shows the evolution of export and imports for merchandise trade in Note: This figure compares merchandise exportgrowth in the Dominican Republic with the Dominican Republic. Source: Authors'calculation using Central Bank information. selected peer countries. Source: Authors calculation using World Development Data for 2013 is preliminary. Indicators. A. RECENT EXPORT PERFORMANCE-AND THE the country. Many nations with similar reliance on MFA ROLE OF SPECIAL ECONOMIC ZONES benefits also had to adjust to the new international rules. Countries in the same region also negotiated preferential Swapping global preferences in certain sectors (MFA) trade agreements. Different export strategies were put to for preferential market access in the US market (CAFTA- the test. Unfortunately, the Dominican Republic seems to DR) did not payoff automatically, butthere are some en- trail similar countries in export performance. Figure 5 com- couraging signs. Figure 4 shows the trends in exports, im- pares export trends in the Dominican Republic and Mauri- ports, and trade balances in current USD dollars. Between tius, Costa Rica, Honduras, Colombia, and El Salvador. 2000 and 2012, imports almost doubled, partly due to de- pendency on foreign energy and increasing international The Dominican Republic's below-average export per- oil prices, while exports increased by just 60%. This dyna- formance also holds when compared against all coun- mism increased trade deficits over time, reaching 11.9% of tries of similar per capita income. Exports over GDP and GDP in 2013.1 The 2008-09 global financial crisis certainly per capita income have a concave relationship: countries affected trade performance, but exports bounced back to tend to export more as a share of the GDP as incomes rise their 2008 level in 2010. In 2011, exports displayed an im- but at a decreasing rate. Figure 6 show the location of pressive annual growth rate of 26%, led by a 45%surge in the country along these two dimensions for 2005-07 and non-SEZ exports. Exports then grew at a more moderate 2010-12. The results indicate that Dominican exports are 5% in 2012 and 6% in 2013, with non-SEZ exports growth far below the norm for its level of economic development rate (12%) outperforming the SEZ growth rate (2%). (denoted by the band in each figure). In fact, the Domini- Ln can Republic is the worst performer among other Caribbe- The global conditions that the Dominican Republic an islands (labeled by their ISO country codes) as well as had to sail over the past decade are not specific to among peers outside the region. Other dots indicate the E rest of countries in the world.10 0 9 This deficit has been traditionally financed by (i) tourism related C revenues, (ii) remittances, and (iii) FDI. According to preliminary cen- 10 These figures do not take into account services, which are espe- tral bank data released in 2013, tourism related revenues account- cially important in the Dominica Republic.When theyare considered, ed for 6.43% of GDP and remittances for 5.49% of GDP. FDI reached however, the country still underperforms the norm for its develop- 3.28% of GDP ment stage. Exports over GDP and GDP per capita Average 2005-2007 Average 2010-2012 120 *U 120 **GUY 100 *SUR *110 100 *1TO SUR 8 80 *BLZ 80 BLZ 0 *HND JAM *HND *LCA 4 60 60 VCT 0 0 C 6 7 8 9 10 11 6 7 8 9 10 11 Log of GDP per capita (PPP) Log of GDP per capita (PPP) Note: The curve shows expected export levels os oashore of GDP given eoch country economic development. The bond represents the 95% confidence intervol. The verticol dotted line is the ave roge per capita income of countries in the sample. Source: Authors'colculotion using World Development Indicotors. Over the past decade, export growth has been more SEshsbeacopndbyatayicrsente dynamic in non-SEZ areas than in SEZs-although nme fepres orh h rdc cp nme SEZs remain a main driver of export performance. o i-ii oe)i uhbodrofrsoeaigot Highly detailed data allows us to review the export perfor- sieSZthnEZxprr. mance through the lens of the firms. Table 2 separates the export performance of companies inside and outside SEZs Tesuc fepr rwhas ifr ral,wt between 2002 and 2012." An in-depth discussion about recent developments and challenges SEZ will be facing W eops oa xotgot ttefr-ee ln ahead is provided in section II.b, The role of Free Trade tes-aldmriso rd'thslvlo nlss x Zones in the Dominican Republic: Adjusting to a New Era of pr rwhcncm rmfu ancmoet.Frt Competition. There are four main patterns worth noting:of e p t First, the share of export value originated from SEZ firms Eto h sae m arienseamarin"rae Se has consistently decreased from 80% in 2002 to 54% in o x exporters may troduc o e prod-er 2012. Second, the number of firms has not followed suit, sid, the m ent er or etwm s an out suggesting a change in specialization within SEZs from ne firs ay En exportrkf the clothing sector to more services-oriented activities .e Third, the increase in exports from firms operating outside of trade. Fiu 7he source of export growth by recet deeloment an chalengs SZ wil b fa ty e ofcexporter (Dtals ont gth a ecomposition aonxpr 11 Data from 2007 and 2012 was provided by the Dominican sus- grwth aloe margins of trade us frel aa ae- toms authorities (Direcci6n General de Aduanas, DGA hereafter) for proe in cor rtin fopartial ye r ts the purpose of this study. Data for 2002-09 was previously collect- ed by the World Bank by Molina et al. (2010). When possible, we use inA exSscon1)TemjrtofxptgowhnD- both datasets, but we cannot track firms between the two databases because firms' identifiers are incompatible. The DGA data provides incan exsors by eveong esthe eotre- customs-level information of all export transactions recorded by the shse itsive margin"of trade. e0 DGA between 2007 and 2012. The database also identifies if an ex- port transaction originated in an SEZ. Export transactions are avail- n ndd 89%sfith exportroh in irms po- able daily for the period of reference. For the purpose of this analysis, we aggregate transactions at the firm-product-country-year-month ed is es come fro the n tensive margin level. A product is a six-digit HS codeTotal exports from the DGA data fetre que omonin thesone as fir baedthere mate of exporter s. Fourt the d poDu ti (soee (nre 1 got six-dgthe)i muchin bfrader forn firmleoperat out-U 11 ~ ~ ~ ~ ~ ~ ~~ sd SEZs than SEZ7 exporters.vde y heDmiianss-a the~~~Th souree of expor growth also difer greatly, witheiulyclec-1 ed y heWoldBak y Mlia t l.(210. henposile w ue n nn-eZ fcis morTe jri ty the exotsv margthin.o both~W decompose, tota exor growth atac them fbrm-leve alongatbae because~ ~ ~ ~ ~ ~ ~ ~ ~~~~~h so-'ietfesae noptbe h GAdt rvds mncalle "mgis o f tade"Atpin tbisev of or anlsi,ex- cusom-lve ifomaio oalxpotrasatinsreoredbyth ltonhesm e arkesthe"minsv manof trade. S ec- 01 por tanacio oignaedinanSEZ Epot rasatinsar aai-ond2, istn 9 oh exporter ma gnroce or rmv poda- ableuts third, they may enterd or exitece new marets andos fourth,nlyis we ggegae ranacion atefrmprouc-conty-yarmo toethsie thEZse three areown ase th"etensive margin " levl.A rouc isa ixdiitHScod. otl xprt frmheDGdt fatre. Fuigue 7mdepict thes sores osfieprt growdtheby 0 minica SEZsl occur byt developin estblihe expor re-e 1) U Exporters located outside Special Economic Zones Exporters located in Special Economic Zones Year Total Growth Firms HS6 Destinations Total Growth Firms HS6 Destinations Export (%) Products Export (%) Products 2002 847.7 6.7 1,909 917 109 4,317.3 -3.7 495 455 94 2003 1,064.0 25.5 2,137 1,016 114 4,406.8 2.1 513 519 106 2004 1,250.7 17.5 2,076 996 112 4,685.2 6.3 525 523 115 2005 1,395.1 11.5 2,054 943 112 4,749.6 1.4 543 522 113 2006 1,931.4 38.4 1,708 954 101 4,678.8 -1.5 496 586 112 2007 2,635.1 36.4 2,151 1,407 118 4,525.1 -3.3 591 948 118 2008 2,393.4 -9.2 1,877 1,396 115 4,354.1 -3.8 510 962 127 2009 1,689.3 -29.4 2,169 1,457 122 3,793.6 -12.9 544 971 121 2010 2,536.1 50.1 2,139 1,487 118 4,217.4 11.2 496 994 112 2011 3,678.0 45.0 3,007 1,527 124 4,814.1 14.1 641 1027 117 2012 4,129.0 12.3 4,168 2,468 122 4,940.1 2.6 431 1334 133 Note: This table shows the main characteristic of exporters operating inside and outside SEZs in the Dominican Republic. Export share is the percentage of total exports coming from each group. Source: Authors'calculation using information from the Central Bank and the DGA. are highly specialized and usually have well-established to peer countries, Figure 8 compares its share of export export channels that allow them to deepen export rela- growth from the extensive margin with other countries tionships. Non-SEZ exporters have been able to expand across a range of development levels and geographical lo- evenly along the intensive and extensive margins. Particu- cations, subject to data availability." It shows that the role larly important is the contribution of the net entry of firms of the extensive margin significantly increases along the into exporting markets, which explains 36% of total non- development path. In fact, the role of the extensive margin SEZ export growth between 2010 and 2012.12 Finally, the in the Dominican Republic is well above what would be analysis also indicates that within firm market diversifica- expected at its income level; it is comparable to relatively tion played a role in export growth for non-SEZ exporters. more competitive countries, including Colombia and Mex- ico. Unfortunately, data availability precludes comparisons The emerging role of the extensive margin as a driver with other Caribbean countries. Annex 3 discusses the de- of export performance for firms outside SEZs is a pos- composition of the extensive margin for non-SEZ Domini- itive indication because growth on extensive margins can exporters in more detail. is usually associated with a healthy dynamic of export relationships. More competitive firms emerge, product In examining recent Dominican export performance, (a c and to a lesser extent sector diversification becomes possi- the final consideration is the role of international prices ble, and firms move beyond local and traditional markets. in export growth. True competitiveness gains come when E 0 To see how the Dominican Republic fares with respect export increases are accompanied by a surge of the quanti- O 0 < 12 It is worth noting that net firm entry is very important for non- 13 Firm-level cross-country comparison are possible using the data SEZ export growth, but these firms face lower rates of export survival available in the World Bank Exporter Dynamics at http://go.world- tin comparison to their SEZ counterparts (see Molina et al 2010). bank.org/KZGM1 20470. grwhfo2h8xtniemri it te onre 7.Exports growth decomposition Non-SEZ Exports SEZ Exports (2010-2012) (2010-2012) Net firm entry Net firm entry Change in established Change in established export relationships export relationships 51% Net market 89% diversification diversification 2% 9% Net product diversification Net product 1% diversification 4% Note: These figures decompose export growth into the part that occurs within established export relationships-the intensive margin-and new export relationships-the extensive margin. For the intensive margin, the share is the net effect of changes of export values within established relationships. For the extensive margin, the shares are the net effect of entrants and exiting firms on export growth. Source: Author's calculations using DGA data. ties a country ships abroad. Figure 9 shows the evolution of Participation of the extensive margin Dominican export growth in both value and volume from at different levels of development 2006 onwards. The figure also shows the deviation of ob- (Latest available year) served export growth with respect to the world. The data .8 come from Gualier, Santoni, Taglioni, and Zignago (2013), COD MEX ~E .6CL who also provide information about the sectoral and geo- MKD graphical composition of exports along with other factors R specific to the exporting country. Sectoral effects indicate .2 whether export performance is driven by sectors that are I KEN PER buoyant in international markets. Geographical effects indi- .0 SLV CHL cate whether export performance is affected largely by main 7.5 8 8.5 9 9.5 trading partners'economic conditions. Finally, the authors Log ofGDP per capita, PPP provide a measure of competitiveness (the so-called push Note: The analysis takes into account all exporting firms in a given country, regardless 0 effect) that isolates the impact of changes in international of their location in SEs. Due to data availability constraints, data for each country are aaverage across different years as foilows: Iran: 2006- 10 Chile 2003-09, Colombia _2 conditions (demand and a change in composition) from the a conitins dem ndanda cang inco postio) fom he 2007-09, Ecuador 2002-0, Jordan 2003-10, Macedonia 1998-2010, Pakistan L changes due to other determinants of export performance. 2002-10, Peru2000-09, Mexico2000-09, SouthAfrica2001-09, Nicaragua2002-11, Turkey 2002-11I; The Dominican Republic.:2007-12. 6DP perrcapita is also the average over the corresponding time period. The chokce of countries is dirt ated by the availabi- lity of customs transaction level data through the World Bank8 Exporter Dynamics After the global economic crisis of 2009, Dominican Database. Source:Authors' Calculations. exports grew in both value and volume. While value E 0 'U growth has been similar to the overall world rate, export B volume grew faster than the world average since 2011. AND EXCESSIVE MARKET CONCENTRATION This performance was mainly driven by push effects, U suggesting an improvement in competitiveness and per- Over the past decade, the Dominican Republic has formance by Dominican exporters. Geographical effects transformed itself from a nation dependent on re- C: seem to be minor for export performance, indicating that sources and clothing production, becoming a more di- main Dominican exports' destination markets have not verse economy where minerals, metals, and relatively 0 been increasing demand. sophisticated manufacturing products have started Dominican Republic export performance Export Value Export Volume 2 .2 *1 * * 00 0 0 .4 -.3 * 2006q3 2008ql 2009q3 2011ql 2012q3 2006q3 2008ql 2009q3 2011ql 2012q3 N Push Effect Geographical Effect Sectoral Effect * Deviation World Export Growth ..--. Export Growth Note: These figures show the year-to-year growth for export value and volume between 2006 and2012. Three main sources of exports are identified: (i) geographical effects, which indicates the extent to which export performance is pulled by the growth in the main destination markets; (ii) sectoral effects, which indicates the extent to which export performan- ce is pulled by high international demand in key export sectors; and (iii) push effect, which ore country specific factors driven by export performonce. The push effect is on indicator of the level of competitiveness of a given country. Source: Gualier, Santoni, Taglioni, andZignago (2013). to emerge. The decade brought a long-term shift in the that after declining over the last part of the 2000s, the tex- engines of export growth, which is linked to the ability to tile sector made a positive contribution of 3.3 percentage appropriate gains from international trade and translate points in 2012. In fact, it was the third best performing them into economic growth and development. Table 3 sector behind only mineral oils at 5.4 percentage point gives an overview of this change in Dominican sectoral and chemicals at 4.8 percentage points. This uneven per- specialization by showing the evolution exports, sector formance reflects a restructuration of the sector from shares, revealed comparative advantages, and contribu- large-scale sewing to more just-in time production, short tions to export growth over the past decade. To under- series, and full-package solutions (discussed in detail in stand the impact of changes in market access policies section ll.b.1 on how SEZs are adjusting to a new era of (mainly with the US) on Dominican export performance, competitiveness). we decomposed the decade into four periods. First, we look at the average export performance in 2003-04, when While the nominal value of clothing exports remained the clothing industry was protected under the US quotas at best flat over the past decade, the country almost defined by the global Multi-Fiber Agreement (MFA). Sec- doubled the nominal value of merchandise exports, ond, we look at the average export performance in 2006- building on the good performance of agricultural 07, a year after the MFA expired but before the country products (vegetables and foodStUff4) and the emer- signed the CAFTA-DR. Third, we look at the post-CAFTA- gence of metals, mineral oils, chemicals, and plastic DR, post global financial crisis scenario by examining ex- and rubber products. The advent of these sectors was port performance in 2010-11. And fourth, we show the not simultaneous. For instance, the surge of metals' ex- sectoral indicators of export performance for the latest ports occurred between 2003 and 2006, when the sector year for which data is available. contributed more than half of total export growth. It is im- portant here to distinguish exports of ferronickel and gold The elimination of export quotas in textiles greatly from those of scrap metal." Ferronickel exports were par- impacted the Dominican Republic. While the sector is E 14 Ic is worth noting that toodstuff includes tobacco exports trom o still predominant in the overall export basket, contribut- companies that are mainly located in special economic zones. ing 15.9% of total exports, the nominal value of clothing 15 Growth in exports of metal (trom 2008-2009 until 2011) was 0 mainly caused by exports ot scrap metal (HS heading 7404). Metal exports in 2012 ($1.1 billion) is slightly lower than it was thett (in particular manhole covers, phone and electric wires, pieces structures of bridges, fences, etc.) for the purpose of exporting due to in 2003-04 ($1.2 billion). It is important to note, however, high international prices was a public scandal 3 years and pushed for De - -r - 0 C>' 0, 00 c - C) 6 l al * -1 r4 00 00 C> -c 00 '.0 CN4 rN - < N * L n~ c)< r-4c l 00 C e~ 1ý - -2 -i 0 n cý O r 0 , ,ý- r a r-4 C3 00 r - c 1 * 1 0r -ý C4 m C3 r co4 n -- -F--- en O N O eN r! 60ir. 1 MM m CR o '-55 OJ --e <=1 r - C rl~ 0 Lnl l.R 49 C ý C fl! C! 1,0 fl t~ ~I 00 ON \00 In 00 m 0200 g/m2 721410 Forged iron 1.5 711319 Articles of jewellery & parts thereof 0.5 Total 34.4 19.4 Note: This table shows the top 10 products (six-digits HS codes) in terms of export value for both firms operating both inside and outside SEZs in 2012. Source: Authors' calculation using DGA data. Average 2003-2004 Average 2006-2007 Average 2010-2011 P Market Share Market Share Market Share Market Share United States 87.2 United States 69.8 United States 53.9 United States 55.0 0 Europe-27 5.3 Europe-27 11.2 Haiti 15.9 Haiti 14.4 Haiti 1.4 Haiti 6.4 Europe-27 10.4 Europe-27 7.0 Brazil 1.1 China 2.1 China 3.0 China 5.0 Canada 0.7 Korea 1.3 Nigeria 1.3 Guatemala 1.9 Cuba 0.4 Japan 0.9 Venezuela 1.0 Honduras 1.6 0 Honduras 0.4 Canada 0.7 Jamaica 0.9 Korea 1.2 - Jamaica 0.4 Taiwan 0.7 Taiwan 0.9 Peru 1.0 , Switzerland 0.3 Honduras 0.5 Honduras 0.9 Canada 0.9 Japan 0.3 Jamaica 0.5 Ecuador 0.9 Nigeria 0.9 97.4 94.2 89.1 88.9 Note: This table shows the export shares by main destination in the Dominica Republic. Exports to Haiti take in consideration only official data. Unreported exports o to Haiti are estimated to be as important as recorded exports. Sources: Authors'computations using DGA data. F2gure Dominican Republic revealed comparative advantage and Brazil and China import shares (2012) 2 2 * Medical or surgical instruments and apparatus * Medical or surgical instruments and apparatus * Iron and steel * Iron and steel * Pharmaceutical products S Plastics and articles theref Plastics and articles thereof aJ 0~~ 76 0 2- -20 -4 -4 1 4 7 10 13 16 19 1 4 7 10 13 16 19 Brazil Import Share China Import Share Source: Authors'calculations using COMTRADE data. centration is only an issue for SEZ firms. In terms of market level generated by a theory-grounded gravity model. concentration, firms outside the zones are not very much We use this framework to see how "natural" agricultural different from similar firms in peer countries. export relationships are and to identify untapped export opportunities at the market level for firms operating out- In the concentration of exports among served desti- side SEZs (Figure 8, Section 2). Figure 13 plots actual Do- nations, Dominican firms outside SEZs are much like minican bilateral export relationships (y-axis) against their exporters in similar countries. However, these firms potential values (x-axis). The model is estimated for a set of do have a smaller product and destination mixes in 213 countries, but for presentation purposes we only show export markets. Table 7 reports the median and the av- values pertinent to Dominican exports. The first panel (up- erage of the number of products (six-digit HS codes), per left corner) focuses on bilateral export relationships destinations, and products-destination combinations by between the Dominican Republic and European countries exporters in the Dominican Republic and peer countries. (labeled by their individual ISO codes). If an observation For Dominican firms outside the SEZs in 2012, the median is above (below) the 45-degree line, the observed export was one for both destinations and products-the lowest relationship is more (less) than what the gravity model among these countries. Looking at averages, Dominican exporters reached six product-destination combinations, efd rcman ineesa(2012 well below the 17.6 for Dominican SEZ firms and 10 for peer countries. The large difference between medians and averages tells us the distribution of the variable of interest (destination, markets, products-destination) differs greatly DOM SEZ 05 from a normal distribution. In fact, it closely resembles to a D ll Pareto distribution, where very few big players determine COL 2 the average value. JAM CRI E DOM NAC 0.12 o Given that Dominican firms outside SEZs are con- LKA centrated in agricultural products (Table 4), we now 00 0.1 02 0.3 0.4 0.5 0.6 O benchmark observed bilateral export relationships in pSource: Authors'calculations using COM TRADE data. The Dominian Republic Data is the agricultural sector with respect to the predicted from the DGA. Dominican Exports to Haiti The trade relationship between the Dominican Republic and its only neighboring country has always been intense, and it grew in an asymmetric way during the past decade. Dominican shipments to Haiti have increased from 3% of total exports at the beginning of the 2000s to around 16% in 2012, with Haiti becoming the second most important destination for Dominican exports.20 The bilateral trade balance is positive for the Dominican Republic. On the other hand, Haitian exports to Dominican Republic have not increased substantially; as a result, imports from Haiti represented less than 6% of Dominican exports to Haiti in 2012. This trade structure has also created asymmetric macroeconomic interdependencies. Haitian imports from the Dominican Republic represent more than 30% of total imports (9.9% of Haitian GDP), while Dominican imports from Haiti are relatively insignificant compared to the size of the economy (0.1% of Dominican GDP). Not surprisingly, evaluating the trade potential of both countries with a gravity model yields simulations indicating Dominican exports to Haiti are near their potential and Haitian ones are well below potential (World Bank 2012). Dominican exports to Haiti have boomed, rising 16-fold from the decade's beginning to 2012. All exported products showed gains, although increases were relatively smaller in food products and metals, the most important sectors in the early 2000s. Table B.2.1 shows the export values and shares by sector for this bilateral trade relationship. Table B.2. 1. Sectoral composition of Dominican exports to Haiti Sector 2003-2004 2006-2007 2010-2011 2012 Value share Value share Value share Value share 01-05 Animal 0.5 0.9 5.3 1.6 22.2 1.9 11.1 1.1 06-15 Vegetable 4.8 8.6 29.7 8.8 125.3 10.8 97.1 9.9 16-24 Food products 12.4 22.3 30.2 8.9 114.8 9.9 98.8 10.0 25-26 Minerals 2.3 4.1 8.8 2.6 50.1 4.3 63.7 6.5 27 Mineral Oils 1.9 3.4 3.9 1.2 19.2 1.7 13.4 1.4 28-38 Chemicals 4.3 7.7 12.6 3.7 55.0 4.7 46.8 4.8 39-40 Plastic / Rubber 3.2 5.8 17.2 5.1 76.3 6.6 76.0 7.7 Y 41-43 Hides, Skins 0.0 0.0 0.1 0.0 1.7 0.2 0.7 0.1 0 44-49 Wood 3.2 5.8 11.4 3.4 20.8 1.8 38.5 3.9 .2 50-63 Textiles, Clothing 6.7 11.9 182.4 559.0 410.4 64-67 Footwear 0.9 1.7 3.2 0.9 4.2 0.4 6.9 0.7 68-71 Stone / Glass 1.0 1.7 1.5 0.5 2.4 0.2 3.5 0.4 72-83 Metals 13.5 26.4 7.8 85.7 7.4 98.3 10.0 C 0 84-85 Mach/Elec 0.6 1.1 3.2 0.9 10.3 0.9 8.2 0.8 U 86-89 Transportation 0.2 0.3 1.7 0.5 5.9 0.5 4.5 0.5 2 90-97 Miscellaneous 0.4 0.6 1.8 0.5 6.8 0.6 8.1 0.8 U Total 55.8 100.0 339.3 100.0 1,159.7 100.0 985.9 100.0 Source: Authors'calculations using DGA data.. 0 20 For the elaboration of this box, we take into consideration data from DGA that differs from the Central Bank data. 0 : Dominican Exports to Haiti (cont.) In the aftermath ofthe January 2010 earthquake in Haiti, SEZ firms fueled exports'rise of 79.8% in 2010 and 21.2% in 2011. Non-SEZ exports to Haiti were traditionally more important, but SEZ exports became predominant in 2010 and represented around 55% of total exports to Haiti in 2012. While SEZ exports are very concentrated on textile and footwear, non-SEZ exports are more diversified. In 2007-12, textiles, clothing, and footwear made up 86% of SEZ exports to Haiti, with vegetables and food products representing another 8%. Food products and vegetables are the most popular non-SEZ exports, but they represent only 25% of total exports. Metals (13%), minerals (10%), and plastic (9%) are also among the top export categories. Recently, controversy erupted when Haiti temporarily banned eggs and chicken imports from the Dominican Republic. Interestingly these products are not at the top of bilateral exchanges in terms of value traded. According to official statistics, poultry products, a sub-category of food products, does not represent more than 10% of total Dominican exports. Nonetheless, this controversy momentarily deteriorated bilateral relations. It, was resolved in February 2014 by a high-level agreement reached in the framework of broader bilateral negotiations. Textiles and clothing made the most important contribution to the increase in Dominican exports to Haiti between 2003 and 2012, accounting for more than 40% of aggregate export growth. Vegetables, food products, and metals are responsible for 9-10% of the increase. These sectors have similar weights (10%) in the current export basket to Haiti. Increases in export values for such sectors as plastic and rubber, minerals, and chemicals explain the remaining export growth. The share of intermediate goods in SEZ exports to Haiti has been constantly increasing since 2007, reaching 95% of SEZ exports in 2012. This increase is very likely a consequence of the Haitian Hemispheric Opportunity through Partnership Encouragement (HOPE) Act, passed by the US Congress in 2006 and amended in 2008. It provides more flexible rules of origin than the CAFTA-DR in qualifying for duty-free entry into the US. One of the side effects might have been a diversion of intermediate Dominican textile products through Haiti, where they are processed to incorporate at least 50% of Haitian inputs, becoming eligible under HOPE. Final products are either directly exported to the US or sent back to the Dominican Republic to be finalized and re- exported to the US (see chapter II.b and the case of Grupo M, which operates its companies on a binational model throughout the Hispaniola Island). An analysis using customs data requires an important caveat. Although formal trade has increased, the porous border between the two countries sharing the Hispaniola Island is characterized by a high level of informal trade that remains unrecorded and could invalidate some of the elements presented here. Types of the BEC-classified exports and imports to/from Haiti (2007-12) 400 350 0 2007 300 2008 - 250 0 2009 200 ~ 2010 2:1 100 50 02012 0 final intermedio primario final intermedio primario final intermedio primario final intermedio primario 0trade zne National territories Free trade zone National territories ** Exports Imports 0 Source:Authors'oalulations using DGA data, according to BECclassifi cation. - 1201 d p p-d d p p-d Nicaragua 1 2 3 2.5 6.9 9.2 Costa Rica 2 2 3 3.2 5.8 10.3 El Salvador 1 2 3 2.4 7.2 10.8 Guatemala 1 2 3 2.5 8.6 12.4 Colombia 1 2 3 2.9 5.0 9.4 The Dominican Republic 1 2 2 1.9 5.7 7.4 National Exports 1 1 2 1.7 4.8 6.0 Special Economic Zones 2 5 8 4.1 12.3 17.6 Note: This table shows the average and median number of products (p), destinations (d), and product-destination combinations (p-d) by firm. Data availability precludes making comparisons for the same year. Data for Nicaragua is 2011, for Costa Rica is 2007, El Salvador is 2009, Guatemala is 2010, Colombia is 2009, and the Dominican Republic is 2012. Sources: Authors'calculations using the Exporters Dynamics Database of the World Bank. predicts and the exporter is said to be over-trading (un- C. TWO-TIER EXPORTBASKETIN TERMS OF der-trading) with its trading partner. If the observation is SOPHISTICATION AND QUALITY above (below) the band parallel to the 45-degree line, the exporter is said to be significantly over-trading (under-trad- The products countries produce, and how they pro- ing). Subsequent panels in the same figure examine Domin- duce them, matter for export-led growth. All else equal, ican bilateral exports with other CAFTA countries (second goods that embody greater value-added in terms of inge- panel, upper right corner), with South American countries nuity, skills, and technology fetch higher prices in world (third panel, bottom left corner), and with the so-called markets. Upgrading product quality, therefore, can be an New Growth Poles (fourth panel, bottom right corner). important source of both export growth and economic growth. Moreover, a significant strand of research sug- While Dominican agricultural exports to the US and Eu- gests countries that produce goods that are more sophis- rope are roughly in line with the predictions of a gray- ticated than the norm at their income levels tend to see ity model of international trade, the country is trading higher rates of future economic growth. below its potential with some large markets in South America and with China. In particular, agricultural ex- An ongoing debate rages over whether export competi- C. ports to Colombia, Venezuela, Brazil, and China were well tiveness is best achieved by an evolutionary process of E 0 below benchmark in 2011. Helping agricultural exporters upgrading-selling lower quality goods to regional mar- find and deepen export relationships in these markets kets and building capabilities before moving into more is a pending policy objective. As protectionist pressures competitive, sophisticated global markets-or by leap mount worldwide, the agenda of market diversification is frogging immediately to sophisticated goods. Hausmann, especially challenging for small developing countries. A Hwang and Rodrik (2006) argue that exporting more so- more thorough look at this issue is presented in Section phisticated products leads to faster growth because of the II.c, which discussed the role of export promotion agencies prospect of benefitting from greater knowledge spillovers 0 in developing economies, of and technology transfers. A product largely produced Benchmarking bilateral export relationships in Dominican agricultural products (HS 01-24) (2012) Europe CAFTA 20 20 15 15 GBR ESP USA BELO % 10 CHE * 10 HND 5 V *PRT 5 0 NIC LV 0 0 4 6 8 10 12 14 4 6 8 10 12 14 Log of Predicted Exports Log of Predicted Exports South America New Growth Poles 20 20 15 15 z 10 *CHL z 10 *VEN us AR E * URY U *COLo 5ZAF *BRA *CHN 0 0 4 6 8 10 12 14 4 6 8 10 12 14 Log of Predicted Exports Log of Predicted Exports Source: Author's Calculations using COMITRADE data. by rich countries--say, an internal combustion engine- edge embedded in the set of products they export (Haus- is revealed to be'rich'and sophisticated (see Annex 5.3, Ex- mann et al. 2012). This measure is also strongly correlated port Sophistication and Complexity -Measurement and Ca- with economic growth. veats). By contrast, low-income countries dominate coffee bean production worldwide, so that product is classified The sophistication and complexity of the Dominican as having low sophistication.The sophistication of a coun- export basket is comparable to countries at a higher try's export basket (denoted as 'EXPY') derives from the stage of development. Figure 14 presents the evolution sophistication of the individual products in it (denoted as of the export sophistication indicators as well as the eco- 'PRODY'). Hausmann, Hwang and Rodrik (2006) show that nomic complexity index for the Dominican Republic and (a C high-EXPY countries tend to have higher future growth some peer countries over the past decade (see Annex 5, rates, supporting the idea that countries 'become' what section 3 for details on export sophistication and com- E o they export by converging to the income level implied by plexity measurements). Exports are more sophisticated their export baskets. The authors subsequently developed in the Dominican Republic than in the peer countries- a more comprehensive measure, called Economic Com- with the exception of Costa Rica. Elimination of the MFA plexity, which also ranks countries in terms of the knowl- temporarily reduced Dominican export sophistication, u The Dominican Republic: Export sophistication and complexity 9.9 .5 *CRI 9.8 0 - -LUL EDOM E Chamomile Hyacinth bean Peruvian carrot Tuna (Prickly pear fruit) Chervil Indigo Pigeon pea Turnip 0 U Chickpea Jackfruit Pineapple Water chestnut* Chinese amaranth Jicama Pinguin Watercress 1- Chinese cabbage Lambsquarter Pokeweed greens Yam -g Chinese water chestnut Lemon Pomegranate* Yard-long bean Note: This table lists all approved products that can be exported to the US by Dominican producers. Asterisks indicate products authorized regardless the country of origin. The source of the information specifies what part of the plant is accepted for entry as well as the authorized port of entry. To reduce clutter in the table, c wejustpresent the name of the product. Source: Authors collection using the US Fruits and Vegetables Import Requirements (FAVIR) database (http://wwwaphis. usda.gov/favir/). o Number of active HS codes as a share of total approved, by country and year 0.8 2002 0.8 2005 0.7 0.7 0.6 0.6 0.5 0.5 0.4 0.4 0.3 0.3 1 1 . 1 1 0.1 0.0 DOM CRI GTM SLV HND NIC 0.0 DOM CRI GTM SLV HND NIC 2008 2011 0.7 0.7 0.6 0.6 0.5 0.5 0.4 0.4 0.3 0.3 0.2 0.2~ I i 0.11 1 1 1 10. 0.0 DOM CRI GTM SLV HND NIC 0.0 DOM CRI GTM SLV HND NIC Note: These figures show the number of products (six-digit HS codes) that are exported to the US as ashare of the total number of products that are authorized by the USDA. Source: Author's calculations using information from FAVIR and Codes of Federal Regulation. pure, wholesome, safe to eat, produced under sanitary list-an automatic process known as Detention Without conditions, and contain informative and truthful label- Physical Examination (DWPE). Once a firm becomes sub- ing in English." In addition, exporters of various specific ject to DWPE, the FDA will detain all shipments for as long products must provide proof that they followed specific as the company remains on import alert, even after cor- production processes-from good agricultural practices rective actions have been put in place to bring products (GAP) for the control and management of microbial food into compliance with FDA regulations. For each detained safety to hazard analysis and critical control point (HACCP) shipment, the firm must contact the FDA with evidence standards for fish products. the shipment is now in compliance and request that it be allowed to proceed into the US.To stop the DWPE process, The FDA inspects around 1% of all food import ship- a formal petition must be submitted to FDA. This petition ments at the port of entry (Buzby, et al. 2008). When- must clearly detail the corrective actions taken to ensure ever a country/product/firm is found repeatedly violating that the problems that caused the items to be placed on 0 US regulations, however, the FDA issues alerts and imple- import alert have been eliminated and will not recur." ments procedures that result in a 100% sampling process. w An import alert authorizes FDA inspectors to detain all 31 Once problems have been corrected and the FDA has cleared C shpet rmcmaisinlddo h lr' e at least five shipments have for entry into the U.S., a petition may .be drafted showing the FDA documentation and evidence that the a) products are no longer at risk of being in violation of U.S. regulations. CL The review process can take several months before a final decision is 30 http://www.fda.gov/Forlndustry/ImportProgram/ImportPro- reached. Until that decision, shipments will continue to be subject to U gramOverview/default.htm. a higher level of potential controls. Import Alert Name Cheese Due to Microbiological Contamination Frozen Raw and Cooked Conchmeat Seafood Products Due to the Presence of Salmonella Tamarind Products (Fresh and/or Processed) from All Shippers from All Countries Due to Filth Coconut Due to the Presence of Microbiological Contamination Coumarin in Vanilla Products Extracts Flavorings Imitations Guanabana (Soursop) Guidance of Foods Containing Illegal and/or Undeclared Colors Cosmetics Containing Illegal Colors Black Hair Cream From The Dominican Republic Skin Whitening Creams Containing Mercury Unlicensed Botulinum Toxin Unapproved New Drugs Promoted In The U.S Manufacturers Of Low Acid Canned Foods And Acidified Foods Raw Agricultural Products for Pesticides Processed Foods for Pesticides Raw Agricultural Products for Pesticides Food Products Containing Sulfites Raw Fresh Fruits And Vegetables Due To The Presence Of Pathogenic Contamination Note: This table presents the current list of import alerts with DWPE affecting Dominican exports to the US. Source: Authors'collection using information from the FDA. The information was accessed in April 2, 2014 (http://www.accessdata.fda.gov/cms_ia/country D.html). Y 0 Currently, the Dominican Republic is under country- ing to Jouanjean (2012), these factors raise the average wide alerts with DWPE for all raw agricultural prod- unit prices of exports to the US. Jouanjean and Le Vernoy ucts due to illegal pesticide residues. Product-specific (2010) shows that there is indeed a negative relationship alerts with DWPE include tamarind products, coconuts, between unit prices and reputation (refusals) on export C. and guanabanas. A firm-specific alert with DWPE covers markets.The fact that countries complywith stringentSPS E 0 raw agricultural products shipped from M.S. Export C.X.A, regulation is related with products with higher quality (e.g. a Dominican firm whose exports of fresh Indian long organic products) which tend to fetch higher prices. squash were found to be contaminated with salmonella. U Table 11 lists the current US import alerts with DWPE for The FDA is also entitled to detain any regulated prod- a the Dominican Republic. Because the burden of proof is uct that appears to be out of compliance, even if not C: transferred to the exporter, these detentions add to ex- under an import alert. The FDA district office will then porters'cost of demonstrating compliance and the time it issue a Notice of FDA Action specifying the nature of the 0 takes for products to reach the final consumers. Accord- violation.The ownerorconsignee is entitled to an informal CAFTA-DR: Unit Rejection Rates CAFTA-DR: Relative Rejection Rates .05 .10 .04 .12 . . . 1.14 * * .02 *.' * .01 .16................... .................. ................ ............. ......... T ..18 0 2002 2004 2006 2008 2010 2012 .20 - Average CAFTA without DR. -- Dominican R. * Costa Rica Guatemala * Honduras Nicaragua * El Salvador Note: This figures show the unit and relative rejection rates for fresh fruits and Note: This figures show the unit and relative rejection rates for fresh fruits and vegetables. Source: Authors' calculations using information from FDA Import Refusal vegetables. Source: Authors' calculations using information from FDA Import Refusal Reports. Reports. hearing to provide testimony regarding the admissibility share of total imports. For example, if the Dominican Re- of the product. If the owner fails to submit evidence that public share of rejections in the US is 5% and the share of the product is in compliance or fails to submit a plan to Dominican export in total US imports is 5.7%, then the RRR bring the product into compliance, FDA will issue another is -13.1 [ln(0.05/0.057)]. Notice of FDA Action, refusing admission to the product. The product then has to be exported or destroyed within Dominican exports of fruits and vegetables faced a 90 days and this information gets automatically recorded higher rejection level at the US border than the same in the FDA import-refusal database. The FDA provides ac- products from other CAFTA-DR countries. Figure 19 cess to this database online through their Import Refusal presents the URR, and Figure 20 shows the RRR. Although Report (IRR). For each rejected consignment, the report its level of SPS compliance improved after the CAFTA- provides information about the manufacturer's name, DR agreement took effect in 2007 (blue vertical line), the country of origin, product code (according to FDA's own Dominican Republic is still above the regional average. industry classifications), product description, refusal date, The Dominican Republic underperforms every CAFTA-DR and the violation reason. We use the IRR to examine the country in both measures. pattern of import refusals from the Dominican Republic and other CAFTA-DR countries in the fruits and vegetables When compared with all countries exporting to the sector. We map FDA product codes to six-digit HS codes US, the Dominican Republic is catalogued as a poor over the period 2002-12.32 We use two metrics, proposed compliance performer. Figure 21 shows the Dominican by UNIDO (2012), to analyze the importance of import re- Republic's RRR in the Big Four international markets: Aus- 0 fusals and to infer possible trade compliance issues. The tralia, Europe, Japan, and the US. This figure divides the .2 first one is the Unit Rejection Rate (URR)-the number of RRR into three equal groups to create a tercile distribution. rejections per US$ 1 million of exports. The second one is Countries in the highest tercile, labeled "high:'have a rela- the Relative Rejection Rate (RRR)-the natural logarithm tively poor compliance performance. Those in the middle of the ratio of a country's share of total rejections to its tercile are labeled "medium;' and best performers in the CL bottom tercile are labeled "low."The source of this analysis 32 This analysis covers 2002-12. It extends the data used in Jouan- U jean (2012)andJouanjeanetal (2012). is UNIDO (2012). In addition to the poor performance in Relative Rejections Rates Top 10 refused products URR in the Big Four international markets (2002-12) 10 Papaya,fresh 1.3 Cucumber, fresh 3.2 DR Juice of Tamarind or Soursop 12.5 5 Pepper, fresh 116.9 R Squash and Gourds fresh 121.1 -1 --..R.Pigeon peas 0 35.8 0 - DR Fresh and chilled beans l 135.3 Other frozen vegetables 158.6 -5 Tamarind and bitter melon 255.4 Eggplant, fresh 652 Australia European Union Japan United States 0 100 200 300 400 500 600 700 M RRRI - First tercile M RRRI - Second tercil RRRI - Third tercil Note: Each product has the following associated HS code: other frozen vegetables: .Median - 5th y 95th percentile e Outliers 071080; fresh or chilled beans: 070820; eggplant: 070930; papaya: 080720; squash and gourds: 070990; pepper: 070960; tamarind and bitter melon: 081090; and cucumber: 070700. Source: Authors'calculations using information from FDA Import Source: Trade Standards Compliance Footprint, UNIDO (2012). Refusal Reports. the US market, Dominican exports to Australia also faced Reasons for rejections (2002-12) high refusal rates, but in Europe they seem to perform around the median. UNIDO data does not report refusal of Costa Rica m Dominican exports in Japan, most likely because there are Dominican R - very few Dominican shipments to that country. El Salvador Guatemala- Rejections of Dominican products are mainly due to Honduras problems with the use of pesticides. The products that Nicaragua face the highest rate of rejection at the US border are 0 20 40 60 80 100 eggplants, tamarind, bitter melon, frozen vegetables, and U Filthy Listeria Pesticides 0 Shigella Insanitary chilled beans (Figure 22). These products have historical- 0 Misbranded U Poisonous substance N No Firm information a Labeling / Perm it/registration m Salmonella ly been subject to high levels of scrutiny by US border ly eensubec tohig leel ofscrtin b USborer Source: Authors'calculations using information from FDA Import Refusal Reports.0 agencies, mainly due to problems with application of pes- ticides." Historically, the heavy use of chemicals, includ- into the country's production system over time. However, ing pesticides, came as a response to the US program in many of these pesticides are banned by the US Environ- the 1970s for the development assistance to agricultur- mental Protection Agency (EPA). As early as in 1987, more al diversification in the Dominican Republic. It brought than 12% of the Dominican shipments tested by the FDA new production technologies and new demands for aes- were in violation of the pesticicide standards (Murray and E 0 thetic and grade qualities. The use of pesticides was not Hoppin, 1992). The issue was particularly important for controlled, and a large number of them were introduced two products for which the Dominican Republic was the largest exporter to the US-eggplants and snow peas. ThisY 33 According to FDA alert 99-05, consignment detention was due led in 1987-88 to the issuance of a countrywide alert for to (i) residues of a pesticide specifically banned for that product; (ii) a) residues of a pesticide not registered for that product; (iii) pesticide pesticide residues and a DWPE. Moreover, the over-use residue levelsthat exceed the maximum residue level allowed forthat pesticide; (iv) failure by an exporter under an FDA detention warning to provide evidence that the products to be exported are compliant etables. The combination of both issues resulted in trade by submitting a pesticide test certificate from a national laboratory 0 within the required 1R0-day period. disruption and in a reduction of areas under cultivation in Year Number of Exporters Number of Exporters with % of Firms period to the US refusals with refusals 2002-2004 373 66 17.7 2005-2007 804 124 15.4 2008-2010 541 78 14.4 2011-2012 439 36 8.2 Note: This table counts the total number of Dominican exporters in the fruits and vegetables sector and the number of firms facing refusals at the US border. Source:Authors collection using information from the FDA Import RefusalReports and tirm-level export information from the Dominican Republic CustomsAgency. the Dominican Republic. As a consequence, the Domini- port refusals in this sector. Since the Dominican Republic can Republic, which was in the 1980s among the leading has had problems with pesticides in the past, US border providers to the US market for snow peas, eggplant, canta- agencies always assume that Dominican shipments are loupe, and some other fruits and vegetables, lost its place not compliant, and the probability of being inspected and to the benefit of other regional competitors, such as Gua- rejected increases (Jouanjean et al, 2012). temala. The Dominican Republic is not the only country in the Salmonella contamination is the second main reason CAFTA-DR region with a history of trade compliances for rejection. Dominican exporters also have had prob- challenges in the US market. Honduras and, in particular, lems with permits or registrations as well as with some Guatemala have had problems with pesticides residues. In shipments determined to be insanitary or filthy. Problems the early 1990s, 27.3% of non-traditional Guatemalan ag- with pesticides residues are common in CAFTA-DR coun- ricultural shipments were detained at the US border due tries (Figure 23). However, rejections due to salmonella to pesticide residues. Many initiatives of Integrated Pest contamination are specific to the Dominican Republic and, Management (IPM), supported by the US Agency for Inter- to a lesser extent, to Costa Rica and Guatemala. national Development (USAID), have tried to address this issue. The program reduced the occurrence of refusals for The Dominican Republic's compliance difficulties with Guatemala to a small extent. (Murray, 1994). Nonetheless, US SPS measures for fruits and vegetables is a sec- the Guatemalan snow peas chain survived the pesticides tor-wide problem, not an issue specific to a small num- and pest outbreak crises, regaining its 1991 pre-crisis ex- ber of exporters. By merging the import refusal informa- port volume in 2002. Few studies look at the pesticides res- tion with firm-level export information, Table 12 shows idue issue in the Dominican Republic. However, according that 36 exporters saw their shipments refused at the US to FTAA (2004), the National Integrated Pest Management border between 2011 and 2012.They represented 8.2% of (IPM) Program for Pesticide Violations and Residue, imple- 0 the total number of exporters to the US in the fruits and mented by the State Secretariat for Agriculture (Secretaria vegetables sector. While the fraction of firms with compli- de Estado de Agricultura - SEA), in 1989 was unsuccessful ance problems is still considered high, the share has de- because of changes in directives and a lack of capacity to creased from 17.7 % in 2002-04. Reasons for the high level fulfil its responsibilities due to insufficient funds, among of refusal are twofold. First, Dominican firms keep using other factors. This situation and the inability to address forbidden pesticides for the treatment of their fruits and SPS concerns are highly detrimental to the country's rela- 0 pvegetables. Second, a path dependency leads to more im- tive competitive position. According to the WTO (2010), the Dominican Republic lacks of institutional capacity to comply with the WTO SPS agreement requirements. Moreover, it seems that This section looks at the existing public and private SpS the country still lacks SPS infrastructure, and exporters related institutions34 and initiatives to better understand have to rely onUS laboratories, Third-party certification is what is currently happening in terms of SPS capacity even more important. Many US supermarkets, rather than building in the Dominican Republic. relying on brokers importing products in bulk, directly im- port from foreign exporters and require third-party certi- In 2005, Dominican public and private institutions fication from laboratories with internationally recognized created the National Committee for the Application quality control systems, such as Primus LabS31 or NSF Davis of Sanitary and Phytosanitary Measures (CNMSF-DR). Fresh TechnologieS16. The committee brings together public and private insti- tutions and organisations relating to SPS regulations in In November 2010, the European Commission (SANCO, the Dominican Republic: on the public side, the Ministries 2010) conducted an analysis to evaluate controls of of Agriculture, Public Health, Commerce and Industry, pesticides in food of plant origin intended for export Foreign Affairs, and Environment Centre for Export and to the EU. The analysis mentioned progress, particularly Investment; one the private side, the JAD (Dominican in following up with increased traceability to the produc- Agribusiness Board) and CODOPESCA (The Dominican er on consignments refused entry in the EU. However, Council of Fisheries and Aquaculture). The committee's the field evaluation still highlighted the poor coverage of objective is to provide advice on the creation and imple- official supervision of small growers exporting to the EU, mentation of SPS-related public programs and regula- with little adoption of good agricultural practices (GAP). tions and to promote public-private interaction and co- A monitoring program for pesticides residues seemed to ordination to comply with SPS standards. To pursue this be considered, but it had no plans for official sampling of objective, it participates in international SPS institutions pesticide residue due to the lack of analytical capability. and WTO committees, disseminates Dominican laws and policies regarding SPS, participates in trade negotiations, Three laboratories operate in the Dominican Republic. and coordinates its work with official and private nation- The Ministry of Agriculture Central Veterinary Laboratory al institutions related to and/or interested in SPS matters (LAVECEN), a laboratory run by JAD (Junta Agroempresari- 0 (CNMSF, 2005). Since 2005, the committee has produced al Dominicana), and 11B3 laboratory (institute for Innovation training materials on SPS measures and formed a delega- in Biotechnology and Industry, governed by an Advisory tion that participated in the WTO SPS committee meet- Council presided over by the Ministry of Higher Science ings (IDB, 2012). and Technological Education). LAVECEN was restructured in 2008 with the intention of becoming the National Ref- erence Laboratory for analysis under the monitoring plan E 34 Existing public SPS-related institutions and initiatives are: The 0 Offices of Plant and Animal Health of the Ministry of Agriculture (SEA); for agrochemicals and veterinary residues in food.The JAD the Food and Beverages Risk Control Department and the Contact laboratory, which has been testing for pesticide residue Point of the Codex Alimentarius of the Ministry of Public Health and H Social Assistance (SESPAS); Department of Foreign Trade and Trade since the FDA DWPE alert in 1992, invested in new facili- Agreements Administration (DICOEX); General Bureau of Standards and Quality Systems (DIGENOR) of the Ministry of Industry and Trade ties with GC-MS to be operational in 2011. The EU consid- (SEIC); Dominican Secretariat Environment and Natural Resources CL (SEMARENA); Department of Food Safety within the SEA serving as ered the facilities adequate. However, it pointed out that a permanent link between the SEA, the SESPAS, the SEMARENA and the SEIC; Department of Environmental Standards of the Ministry of Environment and Natural Resources. Private sector institutions are: Dominican Agribusiness Board (JAD) and CODOPESCA (The Domin- 35 http://www.primuslabs.com/. 0 ican Council of Fisheries and Aquaculture). 36 http://www.nsf.org/.e according to internationally recognized quality assurance a tenfold reduction in refusals from 514 with a total export systems and were not accredited to ISO 17025. In 2011, value US$80.8 million in 2002-07 to 54 with a value of 84.2 the USAID reported on Dominican SPS capacity building million in 2008-10. However, in spite of the improvements activities, acknowledging significant improvements in the supported by the USDA/USAID program, the Dominican quality infrastructure. However, the report emphasized Republic still lags other CAFTA countries in compliance. that more efforts should be made toward developing pesticide residue monitoring tools and regulations. It also 3.1 Policy options to enhance SPS compliance highlighted a lack of institutional capabilities and infra- structure for plant health surveillance systems. Consistency and transparency along the value chain builds competitive advantage and "client loyalty" from importers Overall, various initiatives support Dominican com- because it reduces the risk of trade disruptions due to er- pliance with SPS regulations. The two most important ratic and irregular quality. By reducing risks, it also allows initiatives were assistance programs to support imple- for investment in value chain development, upgrading, mentation of regional trade agreements-the first with and value addition. A sustainable agricultural value chain the US within CAFTA-DR negotiations since 200611 and requires supporting regulatory and non-regulatory mea- the second within the EPAs negotiations with the EU since sures, such asgood agricultural practices, post-harvest han- 201 3. 8 The USDA/USAID PAPA program (2005-11) aimed dling and treatment, good manufacturing practices, and a at improving public-sector SPS infrastructure (systems Hazard Analysis and Critical Control Point (HACCP) system. and personnel) and harmonizing agricultural statistics The implementation of such measures requires both public (production and prices) in CAFTA-DR member countries. It and private capacities as well as strong institutional coor- mainly focused on meat, poultry and dairy, although some dination to promote a business environment conducive to technical assistance was offered in such areas as container increasing the competitiveness of agricultural exporters in inspection, maximum residue levels training, and labora- international markets. In the following, we suggest various tory diagnostics. policy options that could improve the Dominican Repub- lic's SPS compliance for fresh fruits and vegetables. These initiatives seem to have been relatively success- ful because the number of detentions and refusals of Institutions and regulations Dominican products was 10 times lower in 2010 than at the beginning of the decade. The USAID reports that Buildcapacityininstitutionschargedwithplantpest detentions of fresh produce exported to the United States and animal health surveillance, inspection, and san- decreased from over 4,000 in 2007 to less than 500 in 2010. itary standards. Such institutions keep track of pests It is not clear which products were considered and wheth- and diseases present in the country and develop early er detention is equivalent to a refused consignment. A re- warning systems. They can organize vaccination cam- duction of detention might be the consequence of an in- paigns and support the dissemination of knowledge creased access to laboratories certification, allowing firms about risk assessment, mitigation methodologies and 0 to get off the DWPE list. Our own refusals database shows technologies, good agricultural practices, and integrat- ed pest-management techniques that reduce reliance 37 USDA/USAID PAPA (Participating Agency Program Agreement), on pesticides. Some of these efforts have been recently Initiative for improved and harmonized Agricultural statistics and a) Sanitary-Phytosanitary regulatory infrastructure in Central America undertaken by CEI-RD, an institution that has been pro- (2005-2011). 38 The Institutional Support Program for Regional Integration (IS- PRI) and the 10th EDF Program titled "Support to the Forum of Carib- has motivated a noticeable decline in export refusals 0 bean States in the implementation of the commitments undertaken U undertheEconomicPartnershipAgreement(EPA).s at the US and EU borders between 2012 and 2013. * Strengthen the capability and reliability of infra- Exporters'and producers'associations are essential to structure for the control of SPS. Possible actions the consistency of the value chain. Latin America pro- could focus on laboratory infrastructure and appli- vides many examples of exporters' associations play- cation of the principles of internationally recognized ing a key role in the survival and upgrading of fresh quality assurance techniques (such as ISO 17025), op- fruits and vegetables exports. The Peruvian Commis- erations and analytical testing capabilities, and facili- sion for Export Promotion (PROMPEX) and the Gua- ties such as inspections warehouses located at entry temala Exporters Association (AGEXPORT) have been points (terrestrial borders, ports, or airports). essential in promoting of good agricultural practices in their countries. Such associations are help ensure * Adopt International Standards. Dominican insti- the transmission of information both ways, linking tutions would benefit from participation in SPS stan- producers and importers. They provide information to dard-setting institutions and the WTO SPS committee. producers on quality requirements and promote qual- Adopting national standards in line with international ity products on export markets. They are also essen- norms reduces importers'information asymmetry. tial for setting up public-private partnerships and in supporting the development of regulations. A strong * Update the regulation on pesticides. Preventing the producer and exporter association is also an essential entry and use of banned products in export markets tool for the rapid reaction to a pest or food security will reduce the risk of unintended inter-crop contam- outbreak, reducing the potential impact on the coun- ination. This could involve compiling a list of banned try's reputation overseas. pesticides and establishing regulations for maximum residue levels. Coordinate stakeholders through public-private partnerships. Building consistency in the value chain Coordination entities requires the commitment and participation of all stakeholders-from regulatory and government agen- * Support producer and exporter associations with cies to farmers, producers, food business operators, careful identification of the value chain governance and intermediaries. This requires effective informa- structure. Potential support strategies as well as the tion exchange, collaboration, and cooperation among costs and SPS risks in the value chain vary according these entities to ensure that they have access to the 0 to the type of stakeholders involved. For instance, ex- necessary knowledge, skills, and capacities. Public-pri- porters can supply raw materials from smallholders' vate partnerships are a way to coordinate action. In production sold on the spot market, from coopera- resource-scarce countries, public-private partnerships tives, through contract farming, or full vertically inte- can also be a way to alleviate the capacity constraints. grated production. In the US market, exporters can sell products in bulk or sell directly to supermarkets. Such Various examples of food-safety outbreaks demon- E 0 structures should be studied to find the proper scale of strate that one careless exporter can disrupt markets. action, the relevant costs, and potential winners and Joint action and coherence of all stakeholders in the losers." supply chain is a tool to prevent free-rider behavior. 7 39 Governance of the value chain is one reason for Guatemala's dif- However, integration can be detrimental to marginal smallholders ficulties in managing its pesticides residues related import refusals in who rely on the spot market. In such circumstances, supporting co the US. Snow peas exporters, for instance, still rely heavily on prod- operatives and producer groups could decrease the burden of the ucts supplied from the spot market, where supply chains and agri- logistic and management costs of third-party certification. However, cultural practices are suspect. The problem is two-sided: Supporting this is not a panacea. In a weak institutional environment, producers' the integration of the production could help ensure tractability and incentives to fulfil their contract obligations fade when market prices 0 reduce the costs of capacity building in good agricultural practices. are more advantageous. In Peru, for instance, the public and private sectors tion of a demand-driven approach to the provision of collaborated to implement important programs for quality certifications (linking to potential suppliers be- standards harmonization that have proven to be the fore the intervention). basis for success in exporting high quality fruits and vegetables. In Guatemala, joint public-private institu- B. THE ROLE OF FREE TRADE ZONES IN THE tions have been created to support the monitoring of DOMINICAN REPUBLIC: ADJUSTING TO A producers and exporter for compliance with SPS mea- NEW ERA OF COMPETITION40 sures. However, success of a public-private partnership requires a range of conditions, such as governance and The Dominican Republic has been recognized as a glob- transparency, ownership and trust, common interest, al pioneer of free trade zones, generically known as alignment of expectations, and strong leadership. special economic zones (SEZs). With a program ongoing for more than 40 years, the country is home to world-class Targeted interventions to build public and private zones and industrial parks that attract quality investment capacity in manufacturing or outsourced business-processing ser- vices. SEZs have encouraged a shift away from commod- Carefully focus support to private certification ities, with the manufacturing sector growing from about schemes. Various donors' experience in supporting 18% of GDP in the 1970s to 27% by the 2000s. As discussed certification schemes showed the importance of adapt- in part I, Trade Outcomes Analysis, the SEZs went through a ing to the the market channel and demand. Many proj- period of sharp decline overthe past decade, although they ects, either from donors or NGOs, have supported the were able to recover some of their dynamism after 2009. certification of smallholders groups for various private voluntary standards (PVS). However, research shows The future of Dominican free zones will depend on that most producers gain little price premium from how the accumulated strengths are leveraged to face certification. Therefore, producers often have few in- a new era of competitiveness built on fundamentals centives to maintain their certification status once the rather than transient policy distortions. The private external support ends (De Battisti et al., 2009; Supervie sector played a key role in creating and managing SEzs, and Vagneron, 2013). Finally, the fact that producers and organizing itself to engage in policy discussions with are certified does not ensure access to a business rela- the government. Sustainable SEZ development requires tionship with an exporter or a supermarket chain. competitiveness grounded on fundamentals: trained and productive workers, reform of the business climate and The US market is still very heterogeneous in terms high-quality infrastructure, This will gain importance with of quality requirements and private certifications. In the upcoming deadline to align SEZ incentives with the the US, each and every retail company has its own WTO Agreement on Subsidies and Countervailing Mea- food-safety standards that usually focus more on sures (SCM). Furthermore, linkages with the domestic good manufacturing practices (GMP-transformation economy are difficult to forge without incentives to source 0 and elaboration level) and less on good agricultural inputs locally, which limitsthe possibility for learning and .U practices (GAP-production at the farm). They usually upgrading within the zones and beyond. rely on third-party certification. Although many fruits and vegetables exporters sell directly to supermarket This section begins discussing the recent economic chains, many products are still sold in bulk through achievements and policy decisions in SEZs. It then as- CL brokers. Therefore, greater efficiency calls for careful 0 U evaluation of the needs of the supply chain and adop- 40 This section was prepared by Lotta Moberg and Swarnim Wagle. What are Free Zones? Free zones or special economic zones (SEZs) are geographical areas where the rules of business differ from the rest of the country. They usually offer a more attractive environment for local and foreign businesses by: (i) alleviating constraints related to accessing industrial land, quality utilities, and infrastructure; (ii) providing a special customs regime that facilitates duty-free import of inputs; (iii) helping set up and license companies through streamlined regulatory procedures; and (iv) often providing an attractive fiscal regime with reduced taxes and labor regulations. Zones vary in scope from industrial parks and export processing zones to wide area SEZs and free ports. In the Dominican Republic, locating in SEZs opens the doorto a variety of duty-free mechanisms and fiscal exemptions, including:41 - Full corporate tax exemptions; - Full exemption from local VAT (ITBIS) or tax on assets; - Exemption from any import tax, tariffs, and export taxes; - Exemptions are granted for 15 years-20 years if the company is located near the border-but the period can be extended by the National Free Zones Council, the free-zone regulator. The main aim of most free zones and SEZs is to augment foreign earnings and create jobs by facilitating trade and attracting foreign direct investment. Many zones also play a dynamic role by helping to generate knowledge spillovers, pilot economic reforms, and foster development of lagging regions. From about 175 in less than 50 countries in the mid-1980s, free zones have grown to more than 3,500 in about 130 countries. This proliferation suggests something powerful in the idea of resorting to "second best" means of relaxing constraints when the ideal solution of undertaking wider national reforms is not possible for political or financial reasons. The performance record, however, is uneven. While some zones have played a transformative role, especially in East Asia in the 1980s and the 1990s, many fail to live up to initial promises and some end up being wasteful misadventures. Source: Farole and Akind, eds. (2011);Authors. sesses what the compliance with WTO SCM means for fu- emptions, Dominican SE7 exporters will be able to remain 0 ture economic policy, partly drawing on what other coun- competitive in the global context in the medium and long tries have done in recent years. Finally, it looks forward to term only by climbing up the value chain and increasing what the possible evolution of the Dominican SEZ model quality and productivity. Finally, the progressive homoge- could be, including a discussion of a few pending challeng- nization of the benefits enjoyed by companies inside and es. We make three broad conclusions-all that in the con- outside SEZs would help leveling the playing field, reduc- text of compliance with the phasing out of subsidies con- ing distortions. E 0 tingent to export performance (Box 6), to be implemented by WTO countries in 2015. First, the Dominican Republic is 12 likely to declare most of the sectors currently producing in SEZs to be "strategic" This may be a pragmatic approach a to prevent rapid changes in the SEZ sectors. However, tar- 1.1 The Decline of the Textile Industry and Recent C: geting strategic sectors also risks locking in the prevailing Changes in the Policy Landscape system of discriminatory policies that has helped create 0 a dual economy. Second, regardless of benefits and ex- 41 Chapter 7 of Law 8-9. No.offimsNo. of employees Exports (real US$ Share in Free Zone _rMllionin%201nvalue exports(% 1992 272 100,437 1994 301 115,440 1996 284 107,867 2,437.70 56.4 1998 293 135,634 3,142.10 57.3 2000 275 141,945 3,235.90 53.6 2002 262 118,652 2,699.50 51.6 2004 281 131,978 2,448.10 45.3 2006 198 79,365 1,876.00 37.1 2008 143 49,735 1,180.60 26.8 2010 120 41,882 983.9 23.3 2012 111 40,666 1,213.90 25.6 Source: CNZFE. Note: Data from CNZFE slightly differs from the DGA data we use in the first chapter. This might be explained by the fact that SEZ sales to the domestic market are computed as exports by CNZFE but not by DGA, resulting in the lower level of exports reported by DGA. In the Dominican Republic'sfree zones, the most prom- remained and kept investing in the Dominican Republic; inent changeover the past decade has been the relative many manufacturers that specialized in sewing have dis- decline of the textile industry. Since peaking in 1993, appeared since 2003. An American brand that needs a the number of textile firms has been on a steady decline manufacturerto design and make clothes for their specific (Table 13). The number of people employed in the sector niche would demand several rounds of samples to deter- fell sharply from more than 141,000 in 2000 to less than mine its range of products. In this regard, an Asian firm is 41,000 in 2012. The fall of the Dominican textile industry at a disadvantage, located a two weeks'shipping distance is discussed in detail in section I.b Changes in Sector Spe- from the final destination on the eastern coast of the US. cialization and Excessive Market Concentration. With quotas Distribution centers, all of which were previously located abolished at the end of the Multi-Fiber Arrangement (MFA) in the US, are also opening in the Dominican Republic. and the Agreement on Textiles and Clothing (ATC) in 2004 (Wagl6, 2005), newly unrestrained Asian competitors had The Dominican Republic is competitive in just-in-time a particularly negative impact on Dominican companies production, lean manufacturing, and full-package specializing in labor-intensive production, especially sew- solutions. While both Dominican and Asian exporters are ing. Major buyers in North America now found it profitable increasingly offering full-package solutions (Gereffi et al. 0 c to outsource sewing to China rather than the Dominican 2006), the Dominican Republic's location gives it an edge Republic and other Latin American countries. over Asian competitors. Grupo M, one of the largest Do- minican clothes manufacturers in SEZs, exemplifies this The fate of most textile companies after the end of comparative advantage. The company designs, cuts, sews, 2 the ATC primarily depended on their place in the in- decorates, and launders clothes. It has also diversified CL E ternational value chain. Companies producing for their into new product lines, such as synthetics. Grupo M has 0 U own brand, such as Hanes and Gildan in clothing, have reduced its Dominican workforce from 14,000 to 3,600 in the past decade, but it has added 7,000 employees in Hai- import tariffs on domestic sales at the rate that applies ti, where it has outsourced labor-intensive sewing. Besides to other WTO members selling to the domestic market." Grupo M, few companies have outsourced their labor-in- They must also comply with a 3.5% tax on gross domes- tensive stages of production to Haiti, despite the apparent tic sales and the national 18% VAT.' These taxes rough- benefits in investing across the border (Box 1 in Part 1). ly offset the advantage that SEZ companies have by not paying corporate income taxes. Companies and officials In response to the textile industry's shrinking, the alike seem to agree that domestic producers are at a dis- Dominican government initiated regulatory changes. advantage compared to SEZ companies when selling in In the midst of the adjustments required by CAFTA-DR the domestic market. Yet, the previous 20% limit does not implementation (Box 4), a 2007 law proclaimed textiles, seem to have been binding. According to a Central Bank clothing, shoe, and skin manufacturing a national priority. survey (2014: 23), less than 2% of SEZ goods were sold in All companies in these industries were granted the same the domestic market in 2011. Another 10% went to other benefits as SEZ companies, regardless of their location or SEZ companies, while 88% was exported directly. The data target market. The aim was to help companies in these suggest that most SEZ firms are unlikely to start selling in industries absorb the newly unemployed textile workers the domestic market, since many of them are mainly ex- from SEZs. In addition, the government offered subsidized port oriented. loans to 32 qualifying SEZ textile companies in 2008. The move was widely criticized for creating unfair competition. 1.2 Relative Diversification and Recent Rebound Only half of the companies that received the loans are still in the SEZs in operation. Diversification of economic activities undergirds the The government also moved to contain the general de- recent SEZ rebound. Figure 24: Overall growth in the SEZs cline of free zones. Between 2004 and 2007, SEZ employ- indicates SEZs have experienced a recovery in terms of ex- ment fell from 190,000 to 128,000. In an attempt to incen- ports and employment after 2009. The number of jobs in tivize all free zone companies to keep their workers, the the zones has increased from 113,000 in 2009 to 134,000 state offered a temporary subsidy for each worker of 30% in 2012, and exports have also significantly increased. The of the Dominican minimum wage (around $50 per month) number of firms stands at an all-time high of 584, while between 2007 and 2008. The policy may have stemmed the yearly value of export per SFZ employee has risen to 0 the pace of mass layoffs, but it provided only temporary $35,300 in 2012, up from $28,500 (in real USD) in 2004. As .2 relief. In 2008, the SEZs lost only 3,500 jobs. However, as of 2012, 41% of the companies were of American Origin, the effects of the wage subsidy abated and combined with and 35% had Dominican ownership, with the remaining the impact of the global financial crisis, the decline contin- distributed in small shares around the globe. The revival ued with further job losses of 11,500 in 2009. in activity can only be partly attributed to a tepid resur- gence of the textile industry. Because other sectors have E 0 Requirements placed on SEZ companies' proportion of grown too, the SEZs are increasingly diversified into new domestic sale versus export have also been adjusted. sectors (Figure 25)." One is the growth of high-technolo- Previously, a free zone company had to export at least 80% gy industries, as defined by the OECD" to include medical of its production.42 In 2011, those restrictions were lifted, 43 Law 139-11, Article 11, available at http://www.suprema.gov.do/ a and SEZ companies can sell everything they produce on documentos/PDF/novedades/NovedadLeyNo139-1 1.pdf. the domestic market. However, SEZ companies must pay 44 Law 253-12, Article 31, available at http://www.dgii.gov.do/leg- islacion/leyesTributarias/Documents/253-1 2.pdf h 45 For the discussion on diversification of exports and DR's compar- ative advantage see part t. S0 42 Id.: Article 17. 46 OECD: "Isic Rev. 3, Technology Intensity Definition: Classification CAFTA-DR and Free Zones During the textile downturn in 2007, the Dominican Repubic joined the Central America Free Trade Agreement (CAFTA) between the US and five Central American countries. The Dominican Republic was already part of the Caribbean Basin Initiative (CBI) and the Caribbean Basin Trade Partnership Act (CBTPA), and the CAFTA-DR basically helped consolidating existing preferences (see Molina et al, 2010). Before the agreement, close to 80% of Dominican exports entered the US duty-free; now, free access has been extended to most products (except for some agriculture products restrained by quotas-for example, sugar). In addition, CAFTA-DR requires reciprocity in the exchange of tariff concessions, while the CBI was a unilateral preferential program, implying increased competition from member countries, including the US. Overall, CAFTA-DR made permanent trade preferences that had to be re-negotiated under previous agreements, ending uncertainty and restoring some advantages. In particular, the rules of origin were made more flexible, but it is unclear to what extent these changes provide an advantage over Asian competitors not bound by the so-called "yarn-forward" rule, which requires the use of yarn and fabric sourced only from FTA partners. Before 2007, the Dominican producers had to buy most of their inputs from the US to enjoy duty-free access to the US in apparel. Exporters can now source inputs from other CAFTA- DR members. In the case of woven apparel, this practice, known as "cumulation,"also applies to Mexico, with inputs sourced from that country qualifying for duty-free access in the US. This provision is expected to integrate production in the region and help US companies with investments in Mexico. However, it remains unclear the extent to which these provisions negate the advantages of Asian competitors, who pay higher tariffs on their exports to the US but can use inputs in an unfettered manner from the cheapest source anywhere in the world. In other words, these competitors from other regions not yet party to a US free trade agreement enjoy a cost advantage over CAFTA-DR members because the"yarn-forward" rule of origin does not apply to them. It is also worth noting that the US has a product-specific safeguard that it can use to impose tariffs on textile items when it determines imports have surged. While it is often claimed that SEZ companies enjoy an advantage over domestic firms in the domestic market, other CAFTA-DR countries may face lower total costs than both SEZ and non-SEZ companies. This is because they are largely exempted from tariffs that SEZ companies must pay when selling in the Dominican market. However, it is unclear how much this matters across industries. Honduran cigars may, for instance, have a slight cost advantage over Dominican ones now that tariffs have declined. Yet, Dominican shops and hotels sell Dominican cigars, not Honduran. This suggests other border costs, hurdles, and delays could still render competing foreign goods in the CAFTA area unprofitable. equipment and pharmaceuticals, products with certain skilled and well-trained workforce and management. Only complexity.47 By 2012, there were 25 SEZ medical equip- two Dominican companies currently produce medicines, ment companies, with almost 18,000 employees, a signif- but the pharmaceutical industry is expected to growth. icant increase from the 6,650 employees of 1996. Medical The sector employed 1,254 people in 2012 in the country equipment is mainly exported to the US, where Dominican as a whole, rising from only 504 people in 2010. Medicines products must meet the criteria of the FDA. This requires a made in the Dominican Republic are primarily exported to W other Latin American countries. of manufacturing industries into categories based on R&D intensi- ties" OECD Directorate for Science, Technology and Industry, Eco- a) nomic Analysis and Statistics Division, 7 July, 2011, available at http:// www.oecd.org/sti/ind/48350231.pdf 0 47 See The Observatory of Economic Complexity: http://atlas.me- grading and adding value by vertically integrating U dia.mit.edu/. See also Figure 14. Overall growth in the SEZs Free zone exports by industry 150 50 125 40 -.p 30 8100 1 1 20 75 10 50 0 - Firms Employees Exports - Clothing - Medicals Electricals - Jewelry Tobacco - Shoes Source: CNZFE. Source: CNZFE. new production processes. In shoemaking, primarily Diversification has brought a surge in the number of leather cutting and sewing once took place in the Do- specialfreezones.Singlecompaniescanformspecialfree minican Republic. Now, the country makes finished zones (zonasfrancasespeciales) if they cannot locate in an shoes, some of the for well-known global brands like industrial park for such reasons as the need to be close to Timberland and Jeff Bains. At one time, clothing was pre- a particular natural resource. Much chemical manufactur- dominantly shipped to the US in boxes. Now, Domini- ing take place in special free zones. Call centers can often can apparel producers are increasingly making finished obtain special zone status by claiming that they need to clothes and lingerie, on hangers with the price tag on, be in the city to access an English-speaking pool of labor. that can be shipped directly to American stores. Since Some agriculture businesses are also in special free zones. early-stage integration is often energy intensive and ex- As these industries have grown, so has the number of spe- pensive, Dominican textile manufacturers have tended cial free zones, doubling to reach 140 between 2007 and to import yarn. Some textile manufacturers have, how- 2012. However, continuation of his trend is uncertain be- ever, invested in their own mills and can import raw cot- cause a 2012 law suspended the creation of new special 0 ton, primarily from the US. Some lingerie manufacturers free zones, allowing currently operating special free zones work both with synthetics and with molding plastic de- to remain.41The suspension can be understood as a way to tails. As a result, as with many shoe manufacturers, the limit fiscal leakages derived from exemption. only pre-manufactured components they import are metal parts. 0 The service sector now includes the highest number Under the WTO Agreement on Subsidies and Coun- of SEZ firms. It also accounts for the fourth largest share tervailing Measures (SCM), granting special benefits of employment after textiles, tobacco, and medical equip- to companies for exporting is prohibited. By Decem ment and pharmaceuticals. According to a Central Bank ber 2015, all WTO members, except countries the United survey (2014), around 17% of service exports are telecom- munications (primarily call centers), followed by security 48 Article 36, Law 253-12,2012. "Ley 253-12 sobre el Fortalecimien- to de la Capacidad Recaudlatoria del Estado para la Sostenibilidad services, information and IT, back-office functions, ac- FiscalyelDesarrolloSostenible.G..No.10697de1 deNoviembre counting services, distribution, and warehousing. cse 2012;' available at http://www.dgii.gov.do/legisaion/eyesTribu- tarias/Documents/253-12.pdf The History of Medical Device Manufacturing in the Dominican Republic The growth of medical device manufacturing in the Dominican Republic SEZs is intimately tied to Puerto Rico. In 1976, the US gave significant tax incentives to American companies to locate in Puerto Rico, which until the mid-1980s made the American territory the largest offshore manufacturer of both medical devices and pharmaceutical drugs used in the US.The 1983 Caribbean Basin Initiative (CBI) gave fiscal incentives for companies to invest in other CBI countries. By then, labor costs were also higher in Puerto Rico than most Latin American countries. Between 1987 and 1990, some medical devices companies based in Puerto Rico, including Baxter Healthcare, Eli Lilly, Johnson & Johnson, and Abbott Hospitals, started operations in the Dominican Republic, based on the "twin plants" concept. The labor intensive and simpler assembly work took place in the Dominican Republic and the more complex and capital intensive plastic molding was performed in Puerto Rico. Much production at the time was in intravenous sets for administering solutions into patients'veins. It was soon clear that the Dominicans could perform increasingly complex manufacturing processes. Baxter was among the first to set up a stand-alone Dominican plant in 1994, with both plastic molding and packaging done in the country. Several companies soon followed their example. Between 1996 and 2000, Puerto Rico gradually phased out the fiscal incentives that it had relied on, prompting medical- equipment production to move to Latin American countries, such as the Dominican Republic, Costa Rica, and Mexico. Today, complex medical equipment is made in the Dominican Republic, including biopsy needles, blood-therapy products, drainage products, and surgical sutures. Plastic molding, metal grinding, and sterilization all take place in the Dominican Republic. Not only workers are hired locally, but so are managers and executives. Because of the growing skill base and capital investments in the country, the Dominican Republic is likely to remain important for medical-equipment manufacturing. Source: US GeneralAccounting Office, June 1993, Puerto Rico and the Section 936 Tax Credit, GAO/GGD-93-109; CNZFE; the Authors. Nations officially recognizes as least developed coun- A second option is to grant non-SEZ firms the same tries (LDCs) and countries with a Gross National Product fiscal benefits offered to SEZ companies, removing (GNP) per capita of less than $1,000, must comply with the special status of exporters. This may create a level the SCM (Box 6). The deadline applies to the Dominican playing field but at a tremendous fiscal cost in countries Republic. like the Dominican Republic that are already struggling to increase their tax bases.The value of Dominican tax breaks With a view to ultimate compliance, developing coun- offered to SEZ companies is estimated at US$540 million tries have been reforming their SEZ provisions by fol- for 2014, or approximately 0.9% of GDP (the top sector in 0 lowing one among a few possible strategies. One op- terms of tax expenditure, holding 25% of the total).' The tion is to comply and remove the subsidies that come in country is unlikely to follow this path because it could en- C) m the form of tax exemptions-a de facto tax increase. It is dagrtestinblyofp lcfnne. w a politically difficult approach because it can mean reneg- Sing on a country's compact with foreign investors, and it 49 "Gastos Tributarios en Rep6blica Dominicana. Estimaci6n para el L can have a negative effect on the country' attractiveness Presupuesto General del Estado del aido2014" Comisi6n Interinstitucion- 0 t ua Coordinada por la Direcci6n General de Politica y Legislaci6n Tributar- to future investors. ia del Ministerio de Hacienda, Septiembre 2013. WTO Compliance and SEZs Article 3.1(a) of the SCM Agreement prohibits"subsidies contingent, in law or in fact, [... ] upon export performance." SEZ fiscal incentives qualify as subsidies conditioned on export performance because most SEZ statutes require companies to export most or all of their products as a condition for receiving the statutory incentives. Thus, all WTO members that maintain SEZs with fiscal incentives and de jure or de facto export requirements are in prima facie breach of their WTO obligations. Article 27 of the SCM Agreement, however, grants certain developing nations an exemption from Article 3.1(a)'s prohibition on export subsidies. Annex VII states that the countries must either be recognized by the United Nations as a least developed country (LDC) or have a Gross National Product (GNP) per capita of less than $1,000. When a non-LDC's Gross National Income (GNI) per capita surpasses the $1,000 threshold for three consecutive years, it "graduates" from the list. Even when a developing country neither held LDC status nor had GNI per capita of less than $1,000, as is the case ofthe Dominican Republic, the WTO Committee on Subsidies and Countervailing Measures (SCM Committee) often granted ad hoc extensions to SEZs. In 2007, the SCM Committee decided to phase out its practice of granting extensions and set a final deadline of December 31, 2015, for making SEZ statutes in these countries WTO compliant. Some developing countries subject to the 2015 deadline have tried unsuccessfully to reform their SEZ statutes. Others have succeeded in enacting SEZ reforms, but whether they have fully complied with Article 3.1(a) remains unclear. Source: Waters (2013). A third option is to condition SEZ fiscal incentives on ally moving toward incentives available horizontally to all standards of corporate social responsibility (CSR). It companies in select industries or regions: has been argued, mainly by Waters (2013), that condition- ing fiscal incentives on issues like labor standards, work- China abolished preferential tax rates for SEZs in 2008. ers' unions, and ethical business practices could provide It extended all incentives that were available to foreign countries a novel way of branding their competitiveness investors in the SEZs to domestic investors, lowering while also complying with the WTO. It is unclear whether tax rates from 33% to 25%. China also shifted to a mod- 0 any developing countries would go this route. el where subsidies were primarily provided to specif- ic industries and regions, especially in the interior. To Finally, a fourth option is attractive to a government forewarn investors and give legal certainty, the gov- that wants to promote exports, preserve tax revenues, ement announced its intentions to remove preferen- and protect domestic producers from competition, tial incentives for FDI as early as 2000. C. and has been adopted by a number of developing E 0 countries. Because the WTO does not prohibit subsidies Vietnam removed its export-contingent incentives in to companies per se, a government can give special fiscal 2007, the year it acceded to the WTO, with a four-year benefits to designated strategic sectors or regions, rather phase-in period. Its SF7 regime was integrated with than to the country's exporters. Targeting strategic sec- industrial zones that targeted domestic investments. a tors is probably closest in spirit to a modern best practice Taxes were reduced from 28% to 25%. Vietnam also in- C in SEZs, with an emphasis on quality infrastructure and troduced new incentives for high-technology sectors a sound business environment, not fiscal incentives. As and economically lagging regions. To avoid reneging 0 Farole (2013) and Waters (2013) show, countries are gracIu- on initial promises to foreign investors, Vietnam al- Iowed existing investors to maintain incentives until SEZ-equivalent benefits because they were singled out as they were originally intended to expire. strategic, eligible for benefits regardless of their location (inside or outside SEZs) or export performance. Abolishing Mauritius previously had tax-free zones, while the do- the minimum export requirement for SF7 companies has mestic income taxes had a top rate of 25% for corpo- been another important step. However, since the current rate and 30% for personal income. In July 2007, a flat legal framework in the DR still favors exports over domes- tax of 15% was introduced for all firms and individuals. tic sales, this is unlikely to be sufficient for the Dominican All sector-specific exemptions were removed with a Republic to comply with WTO rules. Should the Domini- three-year phase-in period. can Republic continue introducing changes in the legal framework to fully opt for an "strategic sector" approach, Costa Rica decided in 2010 that income tax exemp- it would be important to engage in multi-stakeholder dis- tions for SEZ companies would expire on December cussions for the definition of strategic sectors, in order to 31, 2015. It established a new category of "process- prevent a generalization of exemptions that would result ing companies" that do not need to satisfy any ex- in an important fiscal cost. port requirements. Costa Rica also shifted incentives to regions outside the main metropolitan area. It es- tablished a new 10% fiscal credit for new processing companies that reinvest their earnings. In addition, Costa Rica only pledged fiscal incentives for firms that The reform from export targeting to sector targeting invested at least $10 million and employed at least 100 looks more radical on paper than it may turn out in workers. practice. TheDominican Republic may finally opt for se- lecting its export industries-today's SE7 beneficiaries- El Salvador, like Costa Rica, seeks to condition its in- as strategic. For many firms, fiscal incentives may there- centives either on location in backward regions or on fore not change much with new reforms. The Dominican a minimum investment of $500,000 and a job-creation Republic's small domestic market implies a clear divide requirement of 50 permanent jobs per business. The between the export and non-export sectors. Most SF7 in- country will also extend the period of real estate tax dustries do not have their equivalent producers outside exemptions for companies that double their initial in- the zones that serve the domestic market (discussed in vestment after their first year of operation. section Lc Two-Tier Export Basket in terms of Sophistication and Quality. In terms of demand, no large domestic buy- Panama eliminated fiscal incentives selectively. It ers exist for medical device equipment or components for removed the income tax exemption for manufactur- electronic devices.The Dominican market is only marginal ing but allowed tax-free importation of production for SF7 companies producing clothing and shoes. Cigar equipment. manufacturers sell to the domestic market, but no cigar producers are outside SEZs, so granting all cigar manufac- 0 The Dominican Republic has taken a few tentative turers SF7 benefits would result in much change. In addi- stepstowards thisfourth approach, butthis may not be tion, tobacco producers are in practice already exempted sufficient to comply with WTO norms. The government from paying tariffs when selling in the domestic market. selects specific sectors as eligible for benefits currently 2 enjoyed by SEZ companies. A few tentative steps have al- The strategic sector approach is likely to mimic the ready been taken towards a strategic sector-based model. current SEZ regime. Most companies that enjoy SF7 E 0 In 2007, textile, shoe, and leather companies all obtained benefits would become strategic sectors and keep their "srtgc;eiil foUeeisrgrls fterlcto benefits while continuing to export. This would allow the Apparel export to the US government to keep assisting the country's exporters. Many industries that currently do not exist in the Domini- 394 can Republic are also likely to be christened strategic. If a company offers to invest in car production, for example, that company would likely obtain tax and tariff benefits by virtue of representing a new industry. Besides grant- 12 ing benefits to strategic sectors, the Dominican Republic 3 38 is likely to adjust its area-specific tax benefits regime cur- 030 8 .61.6 rently in place along the lagging region along the Haitian border. On the other hand, it is worth noting that continu- ing to grant benefits to multiple sectors could constitute a missed opportunity to define a clear strategy of export-led Source: UN Comtrade. Note 1: Apparel includes SIC Division 84. Note 2: Shares growth, which would require much more selectivity. More basedonimporterrecordsaveragebetween2001-02and2011-12. importantly, under this approach tax expenditures would be likely to increase, should all industries represented in Producersofplasticdetailsformedicaldevicesorelectron- SEZs be granted with a "strategic sector treatment" To ics locate near these industries. Packaging companies and prevent a potentially serious problem for public financ- warehouses are also located in the same zones as many es, some compensatory measures, such as a progressive of their customers. SEZs thus benefit companies in ways and/or partial phasing out on corporate tax exemptions beyond the government's fiscal benefits. Since private free for companies that have been established in SEZs for more zone managers are unlikely to experience much change than fifteen years. with the SEZ reforms, we do not expect much change in their business models either. The definition of "strategic sectors" would not affect in principle activity in existing special economic zones. Apart from the needed changes in legal framework, we Strategic companies will still need to locate within SEZs highlight next three sets of broad challenges that the to access eased transactions and complementary services. SEZs are likely to face going forward. Meanwhile, non-strategic industries are unlikely to be al- 0 lowed to locate in SEZs. Out of the 53 free zones, 35 are pri- Challenge 1: International competition vate, and already present a high degree of occupancy. In contrast to public zones, they generally offer much more The Dominican Republic can no longer compete just to companies than fiscal benefits, attracting firms with a on low wages. The intensity of Asian competition in the variety of services.The Dominican Republic electricity grid textile industry confirms that competitiveness built only is unreliable, which prompts businesses to buy their own on low costs is not sustainable. In higher-valued goods, E 0 generators. SEZ developers can invest in large generators the Dominican Republic's main competitors are from to offer companies a reliable energy supply. SEZs also within the Latin American region. Many of them have ex- house their own security arrangements and customs of- port-promotion schemes similar to those in the Domini- U fices, which saves companies time and money when deal- can Republic. They also share the Dominican comparative Z C) ing with customs. In addition, several private zones have advantage in full-package textile production. At its height their own fire department, clinics, and training centers. in the 1990s, the Dominican Republic had a US market SEZs also generate agglomeration economies, with many share in clothing of around 5%.This share dropped to 3.4% 0 textile companies, for instance, located in close proximity. in 2001-02 before plummeting to 0.8% in 2011-12. Mexico, Honduras, El Salvador, and Guatemala all exported more than in price. The cases of Conacado (see Box 2) and Gru- clothing to the US in 2012 (Figure 25). Since 2011, even po M provide examples of value enhancing activities that Haiti has taken over the DR in clothing exports to the US, some top Dominican companies of different sectors have benefiting from the HOPE Act of 2006 which is designed to been undertaking, and could be inspirational for others. As promote Haitian garment industry (Hamlin 2011). discussed in Annex 6. Dominican Exporters-Results from the Enterprise Survey however, Dominican entrepreneurs The medical equipment and pharmaceutical industry perceive the lack of adequate workforce as a constraint holds promise for the Dominican Republic, but other to operations, which could eventually make it difficult to countries in the region are also doing their best to at- climb up the value added ladder. tract new investment. The global growth of the medical equipment industry is estimated at 7% annually between Challenge 2: Limited linkages and spillovers to the rest of the 2010 and 2015, and the US constitutes half the world economy demand."' In 2011, the US imported 32% of its medical equipment, and the market is set to continue to grow. 52Be- To foster inclusive growth, backward linkages between tween 2003 and 201, the Dominican Republic increased SEZ companies and domestic firms need to be en- its electro-medical apparatus exports from US$474,000 hanced. A lack of linkages between its SEZ companies and to US$690,000.13 During the same period, however, Costa the rest of the economy has been a long-term problem for Rica increased exports of these products from US$467,000 the Dominican Republic (Kaplinsky, 1993; Willmore, 1995; to US$996,000, and Mexico went from US$2.2 million to S6nchez-Ancochea 2006; Burgaud and Farole, 2011). At US$4.7 million. For the Dominican Republic, it suggests present, SEZ companies are buying a disappointingly low that multinational medical equipment companies have share of raw materials in the domestic market (S6nchez-An- other attractive places to choose when locating in Latin cochea 2012). In 2011, only around 7% of prime materials America. purchased by SEZ companies came from the domestic market outside free zones, while 81% of prime materials Dominican exporters are increasingly looking at add- were imported (Central Bank 2014:20). Competition from ing more value to export products. The Dominican other CAFTA-DR members only partly explains this. Be- strength in the future could lie with relatively skilled tech- cause Dominican companies are exempt from import tar- nicians, specialists, and managers who oversee and devel- ifs on materials for goods that they sell to SEZ companies, op operations at a lower cost than high-wage countries the playing field between them and other CAFTA-DR mem- but with greater sophistication than in low-wage countries bers is even. Domestic companies must obtain a permit as (China, Vietnam, etc.). To remain competitive, Dominican an exporter to sell to SEZ firms, which some claim imposes exporters need to continue to climb the value chain and burdensome bureaucratic procedures. However, several increase their quality and productivity. They would need companies doubt this is an obstacle for exporting. In inter- to rely increasingly on product and process innovations views,mostSEZcompanyrepresentativessaythatthemain to distinguish their exports as niche, high-value goods reason they do not buy more locally is the lack of domes- 0 whose demand is more responsive to changes in quality tic suppliers that meet the quality and standards of their foreign counterparts. Some also stressed that the training 50 Trade statistics from UN Comtrade: http://wits.worldbank.org/ and education of Dominican producers are inadequate wits. 51 http://www.reportlinker.com/ci02249/Medical-Devices.html. and that the poor delivery record of Dominican suppliers 52 This is not least so due to the new US healthcare law. See http:// www.prnewswire.com/news-releases/the-medical-device-mar- ket-usa-1 52980685.html. could have a role in overcoming info asymmetries, as well 53 http://www.trademap.org/tradestat/Country_SelProduct_TS.as- pX. as providing with capacity building on the supply side. Are Dominican SEZs Less Connected to Domestic Economy? Existing research on Dominican foreign investment often characterizes SEZs as"enclaves"that are relatively isolated, reducing the potential for positive externalities and spillovers in the rest of the economy (knowledge transfers, forward and backward linkages, etc.). When looking at specific industries within FTZs, for example, Senderowitsch and Tsikata (2010:28) conclude that only SEZ food production enterprises have relatively strong backward linkages to local suppliers. Textile and other manufacturing seem to have weaker than expected backward and forward linkages, something more typical in such sectors as mining, construction, and utilities. According to Manzano et al. (2013), one reason for the observed lack of linkages owes to the fact that most SEZ industries during the past two decades relied on maquiladoras, requiring low-skilled labor and exporting products of low value-added. Recently, SEZs started to export higher value-added products such as medical equipment and pharmaceuticals, but these sectors import most of their inputs and have built few supply arrangements with domestic suppliers (Snchez-Ancochea, 2012). Schrank (2003) provides an interesting interpretation of the dual economy in the Dominican Republic: Industrialization through export succeeded in East Asia not only because of the presence of FDI but also because governments exhibited coherent efforts to take advantage of the destination market while adapting to requirements of the destination market (e.g. by devaluating domestic currency) and educating the domestic workforce. According to the author, this model failed in the Dominican Republic because of the system of patrimonial rule that prevails in the country. In Costa Rica, by contrast, selective-targeting of FDI in certain sectors (telecommunications) coupled with investments in education and health resulted in higher competitiveness, greater linkages, and a rise in the value-added of the export basket (Snchez-Ancochea, 2006). The lack of existing data to measure externalities forces researchers to rely mainly on anecdotal evidence in discussing the absence of backward linkages and other spillovers from foreign-owned companies and SEZs in the Dominican Republic. In this box we attempt to proxy the backward and forward linkages of the Dominican FDI companies (defined as those with a percentage of foreign ownership above 10% of social capital) by using World Bank-IFC Enterprise Surveys.54 The Dominican survey sample consists of only 57 observations; hence, the results should be interpreted with caution. How strong are the backward linkages between the FTZs and domestic economy? The surveyed Dominican FDI enterprises (some of which are located in FTZs) import almost 70 percent of their inputs, compared to 49 percent in the Caribbean, 58 percent in Central America, 43 percent in South America and Mexico, 54 percent in East Asia Pacific, and 52 percent in Europe and Central Asia (see Annex 6, Table 18). Dominican companies buy an unusually small share of their inputs from local suppliers, suggesting backward linkages may be particularly low compared to other regions. This seems to be confirmed by data from national sources; according to the DR Central Bank, around 81 percent of the companies in free trade zones obtain their inputs abroad, 12 percent buy from other SEZ companies, and only 7 percent use suppliers outside SEZs. Y 0 C On the other hand, forward linkages (imperfectly proxied by the percentage of indirect exports to sales) seem to be much larger _) in the in the Dominican Republic (13.8%) than in the Caribbean (6.6%), Central America (7.5%) and South America and Mexico (2.97%). Alternatively, the Dominican Republic's high ratio of indirect exports to sales may be the result of companies moving to the SEZs and selling part of their output to another SEZ entity (a trend described in this section). Differences in the ratio of direct exports to sales (16%), a possible proxy of vertical (market-oriented) FDI, are not statistically significant between the Dominican Republic and Latin America and the Caribbean (Table 18). If we add up the Dominican Republic's direct and indirect exports, the remaining 70% are sales in the local market-a share similar to other Caribbean and Central American countries, higher than 0 in East Asia (58%), and lower than in South America and Mexico (83%). That means that Dominican Republic and the Caribbean have a relatively balanced mix of FDI motivations, whereas FDI companies in South America are mainly "market seeking" (or 2 horizontal, meaning they mainly serve the domestic market) and those on East Asia are mainly"export oriented"(or vertical). One U caveat we have to acknowledge is that the World Bank Enterprise Survey does not allow us to distinguish between FDI companies C. inside and outside SEZs. As a result, this rough approximation of forward and backward linkages would be valid for overall FDI companies in the Dominican Republic, but there may be differences between the two regimes. 0 54 For data and metadata, please access http://www.enterprisesurveys.org/. Are Dominican SEZs Less Connected to Domestic Economy? (cont.) As an alternative approach, we have also drawn from customs data to try to proxy backward linkages as well as to better understand the nature of the production processes in Dominican SEZs. Looking at the customs database, we observe that 70% of the registered firms both import and export, indicating that SEZ companies obtain part of their inputs abroad. For example, the value of clothing imports covers the 51.5% of the value of exports in firms that both import and export. The share is 41.2% in shoes and 39.8% in surgical products. If we take into account that value adding activities take place mostly in SEZs, this would mean that the value of the SEZ inputs acquired in the domestic market would be relatively small, sustaining the intuition backward linkages are limited. When we further disaggregate imports according to broad economic categories (BEC),11 we can distinguish between final, intermediate, and primary products. A clothing export company's average composition of imports is 65% in intermediate products, 10% in primary products, and around 25% in final products (figure below). It is not surprising that they have to acquire primary products abroad, given that the Dominican Republic does not produce cotton. In the case of surgical products, the share of final products over total imports is close to 60%, whereas imports of primary products are below 2%. This could indicate that the Dominican Republic enters the value chain of surgical products relatively late, mainly assembling already finished pieces. By contrast, the role of clothing and shoe firms established in Dominican SEZs would be more transformational within the global value chain for those products. These results are tentative. More research is needed on questions about the intensity of backward and forward linkages in the Dominican Republic, the value added generated by the SEZs, and the Dominican SEZs positioning in global value chains. But, overall, these preliminary findings would tend to confirm the relative weakness of backward linkages in the Dominican Republic, compared to other countries in Latin America and other regions in the world. It could be argued that this is due to the fact that the Dominican Republic present similar ratios of foreign inputs to those in other small islands; however, with a population of around 10 million and rich natural resources, the Dominican Republic should not be in principle constrained by size to develop scale economies that would allow efficient production of local inputs at reasonable quality and cost, which could feed into the production processes of foreign companies in special economic zones. Import breakdown of major exports in SEZs (2007-12) 100% - 90% 80% 70% 2 60% - 50% 40% 30% 1- 0% E o 0% Clothes Shoes Surgical Products 0 - Final Intermediate Primary - N/A Sources:Authors'calculations using DGA data. o C 0 U55 http://unstats.un.org/UNSD/cr/registry/regcst.asp?CI=10O&Lg=l. The lack of linkages between SEZ and non-SEZ firms Challenge 3: Transition to a newSEZ scheme can be explained in part by domestic companies moving into the zones once they become SEZ sup- Should the Dominican Republic finally opt for a "Stra- pliers. They can do so because sales to SEZ companies tegic sectors"approach to complywith WTO rules, pol- count as exports. It is also practical for a firm to locate icies will likely be open to changes and may become near its SEZ customers. The Central Bank (2014: 20) re- increasinglyvulnerableto pressuresfrom the business ports that around 12% of prime materials in SEZ pro- community. Newsectorswill need to be listed asstrategic duction come from other SEZ companies. Previously, while old ones may be removed. This flexibility could en- all SEZ companies were foreign-owned; now, Domini- courage interested companies and industry organizations can firms constitute around 36%. The increasing num- to try to influence the political process. Companies will ber of Dominican SEZ companies shows that domestic have an incentive to lobby policymakers to be included producers are indeed taking advantage of SEZ benefits. among strategic sectors, The main argument is likely to be Thus, in this context, and as Dominican suppliers move preserving employment. As a country that relies heavily to SEZs, it is not surprising that we do not observe on exports, the Dominican Republic will always be vulner- stronger backward linkages.. Nevertheless, it is worth able to fluctuations in international demand and compe- noticing that, as we discuss and prove in Box 7, foreign tition from foreign suppliers. While exporters can become owned companies in the Dominican Republic buy a sig- dominant job creators, they can also cut jobs quickly as nificantly lower proportion of their inputs abroad, if we partofatransitiontohighervalue-addedproductionoras compare with other countries in Latin America or other a response to lower-cost competition. Such adaptation is regions in the world. This would in principle imply that inevitable for the Dominican Republic to stay competitive. Dominican companies (at least outside SEZs) would There is a risk, however, that such changes will be stymied have less chances to learn from the more sophisticated by policies providing incentives in the long run. Policies foreign companies, and benefit from knowledge spill- that favor job creation must take into account the need to overs and other externalities. upgrade efficiency; if not, they risk achieving fragile devel- opment outcomes at a high fiscal cost and will not provide In the absence of complementary initiatives, linkages the right incentives to foster long-run job creation. and spillovers to the rest of the economy are likely to remain weak as the country moves to the new"strate- In the medium term, a more suitable way forward for 0 gic sector" scheme. As exporters who enjoy fiscal bene- the Dominican Republic would be to aim for a combi- fits become increasingly sophisticated, the gap between nation of nationwide "horizontal" export promotion the quality that strategic companies demand of their policies with enhanced selectivity in defining just a suppliers and what domestic producers and suppliers couple of strategic sectors. In this sense, as discussed offer will likely widen. The Dominican Republic could try in section Il.c., the empowerment of export promotion C. to promote backward linkages between foreign-owned agencies (EPA) would be desirable; they are able to bring E 0 SEZ companies and local suppliers by introducing in- efficient support to the export processes ofall Dominican, centives aimed at attracting foreign investors to those regardless of their legal regime. This could be combined sectors that rely more on domestic inputs (metals, ma- with a progressive reduction in the number of strate- Y chinery and other manufacturing). A more demanding gic sectors granted fiscal benefits that takes place in the but more rewarding alternative would be to use policies framework of a consensus national strategyfor export pro- aimed at enhancing the capacity and capabilities of do- motion, where a couple of"champion sectors" are select- mestic producers, so they can become reliable suppliers ed.These champions would ideally be dynamic and young 0 for foreign subsidiaries established in SEZs. sectors in the Dominican Republic, with relatively sophis- ticated production processes, notable value added, and targeted supply-driven policies in the form of institution- large potential to become growth engines in the future. al support to help firms diversify into new markets, start Nonetheless, as it has been noticed by authorities, the exporting, or maintain a commercial relationship. The Do- elimination of benefits beyond exports subsidies (such as, minican Republic could deepen exports to China and Brazil among others, corporate tax exemptions) would need to by exploring sectors outside the extractive industries-in be carefully considered, since it may have negative impli- particular, pharmaceuticals, plastic products, and medical cations in terms of attraction of foreign direct investment equipment. This type of policy could be delivered through (as other neighboring countries continue to offer fiscal an institutionally empowered export promotion agency benefits). Any decisions in this area should be based in a (EPA), and this section will discuss how the Dominican in- solid technical assessment, and a cost benefit analysis of stitutional structure supporting international trade might special economic zones. help foster both market and product diversification. The first sub-section highlights how the current institutional In the long run, nacting export-promotion policies that architecture in the Dominican Republic seems to present are "horizontal" in nature may help the government some overlaps. The second subsection looks at some inter- avoid the distortions created by vested interests in ex- national experiences in discussing the role export promo- isting discriminatory fiscal policies. Mauritius exempli- tion agencies and other public and private institutions in fies how a country can transition from a dualistic economy promoting market and product diversification. with SEZs to a dynamic "duty-free island" (Rodrik, 1999; Baissac, 201 1).This was achieved by segmenting the labor This section concludes that the Dominican Repub- market,removing tariffs,and harmonizingat l5%corporate lic seems to lack an overarching national strategy for taxes for both SEZ and national regime companies. How- trade competitiveness, export promotion and orienta- ever, it is worth noting that this process in Mauritius took tion, and investment attraction. It could benefit from strong political leadership and a competent bureaucracy having a strengthened EPA. International practices point to manage the pressures from different stakeholders." to the advantages of a unified, strong EPA that addresses significant information shortages and assists companies in C. THE INSTITUTIONAL INFRASTRUCTURE exploring new markets. Second, EPAs improve their effec- SUPPORTING INTERNATIONAL TRADE57 tiveness when they offer bundled services and tailored ap- proaches. Third, the Dominican Republic needs to clarify As suggested in section I.b, the Dominican Republic the role of the different institutions dealing with foreign might be missing export opportunities in emerging trade policy and support, as well as to try to better balance markets; this section discusses the role of public agen- the EPA budget so that payroll is not a burden limiting re- cies in supporting exporters. Deepening regional inte- sources available to support export promotion activities. gration and taking full advantage of preferential market access policies are important to export success; so are 0 56 Baissac (2011) explains that S. Ramgoolam, the first prime min- ister of independent Mauritius, initially had little support from his ranks in labor and business, but he won them over gradually by ar- The Ministry of Industry and Trade (MIT) is the lead- guing that the new policies would ultimately benefit all Mauritians. In tandem with the political art of persuasion, the country significantly ing entity in charge of formulating and implementing strengthened the capacities of the Ministry of Commerce and Indus- try in terms of staff size and expertise, often with the help of interna- trade policies." Legally, its role is to (i) formulate and tional donors. This included, for example, new cells to protect export- ers through project evaluation, monitoring, investment promotion, export marketing, project funding, and insurance. 58 Created in 1966 by the law 290-66, the Ministry of Industry and o 57 This section has been prepared by Rafael van der Borght and Trade is the public entity in charge of formulating and implementing Aleksandra Iwulska industrial, trade, and mining policies. eqimn.hstp6fplcycudb eiee hog execute trade policies aimed at steadily increasing the Originally, CEI-RD was created to consolidate some at- presence of domestic goods and services in international tributions and become the main agency for promoting markets; (ii) to manage the implementation of free trade exports promotion and attracting investment. Created agreements (FTAs) and assist the private sector to make the in 2003,11 CEI-RD absorbed the previously existing exports most of FTAs.The MIT's current organizational structure in- promotion center (CEDOPEX) and added attracting invest- cludes two entities directly dealing with trade, headed by ment to the main objective of "promoting and boosting a vice-minister for international trade and a vice-minister Dominican exports and investment to stimulate the com- for free-zones and special regimes. The Directorate for In- petitive insertion of the country in the global economy." ternational Trade (DICOEX) executes trade policies in the Gathering private and public actors, CEI-RD aims at of- framework of political decisions made by vice-ministers. fering services ranging from participation in export pro- Traditionally, DICOEX has focused mostly on administrat- motion activities (i.e. commercial missions, commercial ing and implementing trade agreements, especially from exhibitions), to detection of new market opportunities or a legal perspective, to ensure full compliance with inter- assistance to implement quality norms national obligations arising from FTAs. It is worth noticing that a new draft law aiming at reestructuring the Ministry So far, CEI-RD has mainly focused its assistance on of Industry and Trade is, as of July 2014, under discussion incentives established by Law 84-9911 and assistance at the Congress. in taking advantage of FTAs and preferential trade schemes. In particular, as the entity in charge of validat- Export promotion activities and, more generally, ing certificates of origin, it helps exporters obtain such trade policies are scattered around different entities. certificates. In the region, other agencies undertake both A myriad of commissions and public entities, not always promoting exports and attracting investment: CEPROBOL circumscribed to the MIT, are directly or indirectly dealing in Bolivia, FIDE in Honduras, JTI in Jamaica, PROMEXICO with trade-related issues. Some institutions have been in Mexico, VCE/DNPE in Panama, REDIEX in Paraguay, and established to manage very specific cross-sectorial is- APEX in Brazil. Some agencies are responsible for only ex- sues-for example, the Regulatory Commission on Unfair port promotion: PROCHILE in Chile, PROCOMER in Costa Competition and Safeguards Measures and commission Rica, EXPORTA in El Salvadorian, and DPC/ME in Guatema- on the Application of Sanitarian and Phyto-Sanitarian Ia. Still others are also have a tourism mandate, such as Pe- Measures. Other institutions have a direct role in defining ru's PROMPERU and Colombia's PROEXPORT. 0 industrial policy-i.e., the National Council of Free Trade Zones (CNZFE). Other ministries are also involved in trade Although CEI-RD offers a comprehensive set of ser- policies; for example, the Ministry of Foreign Affairs is in vices to exporters and investors, its actions are com- charge of looking for new FTA opportunities and leading plemented by CNZFE for companies located in SEZs. the negotiations.The Ministry of Economics, Planning and CNZFE is the official public entity in charge of managing C. Development (MEPyD), with its newly created vice-minis- SEZs. Comprising public and private actors, it has a dou- E 0 ter for competitiveness and the National Council of Com- formuatio and 59 Law 98-03, which created CEI-RD, acknowledged dispersion and petitiveness (CNC), participates too in thefomltnad implementation of trade-related policies. Even when the overlaps in public resources devoted to these tasks and aimed at in- tegrating them in a central body.7 current institutional architecture takes into account the 60 Article 6, Law 98-03. Z 61 For a comprehensive presentation of services offered by CEI-RD, ai need for cross-sectorial coordination to comprehensively see the institution web page: www.cei-rd.gov.do. approach trade-related issues, the multiplicity of actors 62 Law 84-99 on "exports reactivation and promotion" establishes the regime of (i) temporal admission for asset completion, 00i sim- and interlocutors could to some extent hamper the con- plified compensation of customs duties, and (iii) reintegration of tax- es and customs duties. CEI-RD handles the official authorization for 0 sistency of trade policies (Figure 27). these three mechanisms.e ble mission: (i) regulate the Law 8-90 on the establishment This duality in public export promotion agencies is and functioning of SEZs and (ii) design an integrated pol- mirrored in private-sector organizations. While the icy for promoting free zones, attracting new companies, Dominican Association of Exporters (ADOEXPO) gath- and developing existing SEZs.63 Even if legally integrated,64 ers established domestic and SEZ export companies, the relationship between CNZFE and CEI-RD may in prac- ADOZONA only represents companies located in SEZ tice become relatively complicated, given that both insti- or operating free-zone parks. According to World Bank tutions have similar objectives. For example, both CNZFE conversations with various actors, ADOEXPO has over and CEI-RD have been participating in international exhi- the past few years been especially active in support- bitions with the aim of attracting foreign ivestors. At the ing and promoting exporters through exhibitions and same time, CEI-RD is assisting SEZ companies as well as do- commercial missions. Meanwhile, ADOZONA focuses on mestic companies in their export processes, while CNZFE defending the interests of SEZs when engaging in pol- is not involved in export promotion.This dual institutional icy dialogue with other public and private institutions. framework, which seems to present some overlaps, reflects ADOZONA represents SEZ employers when negotiating the dual structure of the economy and the two-tiered ex- SEZ minimum wages." Finally, it is worth noting that port basket highlighted in the Trade Outcomes Analysis. some members of ADOZONA also participate in ADOEX- The MIT has separate vice ministers dealing with exporters PO, which seems to facilitate the discussion of policies in the national regime and in the SEZs, which could to cer- on both sides. Both entities are represented at the board tain extent complicate the coordination between CEI-RD of directors of CEI-RD and CNZFE. and CNZFE. The incentives offered to attract foreign multination- In addition, companies located near the border with als seem to vary widely. The task of attracting FDI is infor- Haiti (the "border zone") benefit from special treat- mally delimited-the CEI-RD focuses in the non-tradable ment, granted independently of the destination of sector and the CNZFE usually concentrates on manufac- trade (domestic market or exports). This special regime turing companies. At the same time, the incentives offered is not administrated by the CNZFE but by the Council of to foreign investors may vary widely depending on the Coordination of the Development Border Special Zone, a bargaining power of the multinationals and the Domin- part of the MIT. Under this regime, companies located in ican officials conducting negotiations. For instance, the provinces bordering Haiti are granted full exemption on president directly negotiated the agreement with Barrick internal taxes (including local VAT) and customs taxes. In Gold, while other deals have gone through the Ministry of addition, they are granted a 50% reduction on any tax, Foreign Affairs, CEI-RD, or CNZFE. If the institutional chan- rate, or contribution established or to be established. As nels in this area not be clearly defined, companies could we write this report, the modification of this regime is un- suspect unequal treatment. der consideration due to its apparently modest effective- ness in attracting new firms to the border zone and the In this context of multiple actors and blurry attribu- revenue losses it implies." tions, the Dominican Republic should try to establish o C an overarching national strategy on trade compet- itiveness, export promotion and orientation, and 63 Article 19, Law 8-90. investment attraction. There have already been a se- 64 The executive director of the CEI-RD is on the CNZFE board of di- a) rectors and vice versa. CEI-RD also promotes free zones and regularly refers CNZFE to foreign companies. 66 SEZ minimum wages are established following the same mod- 65 For 2014, revenues losses associated with Law 28-01 are estimat- el as the domestic market-a tripartite commission meeting on a ed at US$24.7 million, according to the Ministry of Finance; see http:// bi-annual basis. They were set at RD$7,222 for 2013, compared with o www.dgii.gov.do/publicaciones/estudios/Documents/GastoTributar- RD$11,292 for big domestic companies and RD$6,880 for small do- U io2014.pdf. mestic companies. Institutional architecture supporting trade policies in the Dominican Republic Ministry of MEPyD Foreign Affairs Vice-Ministry for National Competitiveness oTrdNeoitosMINISTRY OF INDUSTRY AND TRADE Ntoa opttvns Minisry ofCouncil Agriculture Natial Council on Pro-Industria Council on Safeguard CEI-RD CNZFE Council of Coordination measures Export Promotion Agency Promotion of FZs, attraction of the Development Border and Investment Attraction of new companies and Special Zone development of existing ones Adoexpo Adozona Represents export Represents FTZs companies companies 0 MIT at heads the Board of Directors of the entity 0 Participation of the MIT -. Participation in Source:Author's Elaboration. ries of coordination / prioritization efforts, which would to reinforce Dominican competitiveness to mitigate the need to be consolidated and added to the abovemen- impact on local industry of the elimination of tariffs on tioned strategy. manufacturing products under CAFTA-DR.61 In order to 0 fulfil the roadmap foreseen in "en ruta ol 2015, enhanced First, it is worth mentioning that the Ministry of Indus- coordination DICOEX and CEI-RD could be needed. try and Trade's Institutional Strategic Plan for 2013-1767 seems to grant the Vice Ministry for International Trade In addition, President Medina has highlighted three prior- responsibility over policies aimed at achieving the "rise in ities for 2013-16: (i) creating a ventanilla 6nica to ease the C) Dominican exports" and developing "productive sectors registering process for investors (already established); (ii) E 0 able to take advantage of trade agreements."The role of replacing the National Housing Bank (NHB) with a new Na- CEI-RD is barely mentioned, limited mainly to implemen- tional Bank for Export Development to facilitate financing, tation aspects. and (iii) modifying the diplomatic service to add commer- U cial agents abroad.z Second, the MIT launched a program called "en ruta al 2015" in early 2014, with certain sense of urgency. It seeks 68 In the framework of the CAFTA-DR, 97% of the products will en- ter the Dominican Republic free of customs duties by 2015. This in- 67 Available at http://www.seic.gov.do/transparencia/plan-es- cude nearly all industrial products and will be key for the Dominican 0 tratigico/planificaci6n-estratmgica.aspx. industrial sector. The Dominican Republic should look beyond measures al- past decade, studies of EPAs in developing countries have ready under implementation.The formulation of a consen- been more positive. For instance, Alvarez (2004) investi- sus national strategy dealing with trade competitiveness, gated Chilean companies and showed that export-pro- export promotion and orientation, investment attraction, motion programs were positively correlated with export productivity enhancement and innovation, would help performance.0 Similarly, Volpe Martincus (2010) demon- align existing and future strategic plans, improving in- strated a positive and significant impact of EPA activities ter-institutional coordination and, ultimately, reducing the on exports in LAC. fragmentation of activities and achieving a more efficient outcome. How does CEI-RD compare with other EPAs in LAC In addition, a stronger and institutionally empowered EPA could help foster Dominican exports. Research on Like most agencies in the region, CEI-RD is a public the effectiveness of EPAs presents inconclusive findings. institution with a private-public board of directors, a Lederman et al. (2009), for example, shows that EPAs posi- best practice in the developing and developed world. tively affect exports but finds strong diminishing returns in CEI-RD is managed by an executive director nominated by the resources devoted to export promotion. In the Domin- the Dominican president, a common practice in LAC; the ican Republic, authorities admit there is no clear selection exceptions are Bolivia, Ecuador, Honduras, and Jamaica, criterion for strategic sectors when granting companies where executive directors are selected through a public financial or technical support. Efforts are general and do competition, and Costa Rica, where the board of directors not always consider the type of FDI, or the industries more makes the selection. CEI-RD has a board of directors com- likely to be engines for growth and employment. This is posed by a mix of private and public officials,"a common likely to result in a suboptimal use of existing resources. practice in LAC and developed world. In fact, research sug- The next section offers as more in-depth discussion of the gests that exports increase with the share of board seats organization, scope, and resources of CEI-RD, the EPA in held by the private sector (Lederman et al 2009). the Dominican Republic, comparing it with regional peers and trying to improve its efficiency by drawing on the les- CEI-RD has nine representatives abroad and 49 repre- sons of international best practices. sentatives in embassies in charge of commercial issues. CEI-RD has its headquarters in Santo Domingo, regional offices in Santiago de los Caballeros, San Pedro de Maco- ris, and Puerto Plata, and 36 offices distributed around the world.2 CEI-RD does not have as many foreign offices as The mandate of LAC export promotion agencies (EPA) has changed significantly over the past four decades 70 Following Belloc and Maio (2011), several management studies confirmed a positive contribution of export promoting activities on due to global trends; however, research on their effec- company export performance. Authors refer to Gentrurk and Kotabe tiveness(2001), Alvarez (2004), and Lages and Montgomery (2005). tivnes ha ben lmitd nd rodcedmixd ot- 71 Public representation spans from the MIT to the General Cus- comes. Several studies conducted in early 1990s showed toms Directorate. Other public sector members are: Ministry of For- eign Affairs, Ministry of Economy, Planning and Development, Minis- .2 that EPAs had no significant impact on export perfor- try of Finance, Ministry of Agriculture, Proindustria, Consejo Nacional 0 clde Competividlad, and Consejo Nacioal dle Zonas Francas dle Exporta- mance in developing countries (Belloc and Maio, 2011).6 cion.TheprivatesectorisrepresentedbyConsejoNacionaldeEmpre- sas Privadlas (CONEP), CODOPyME, C6mara dle Comerio y Produccion Q Low effectiveness has been partly attributed to uncertain- de Santiago, Cimara de Comercio y Producci6n de Santo Domingo, ty rlatd t diburemet poceure an deays Inthe ADOEXPO, ADOZONA, Asociacion dle Inversion dle Empresas Extran- Sty related to disbursement procedures and delays. In the Qj jeras, and Junta Agroempresarial Dominicana (JAD). CL 72 The 36 offices abroad are: 14 LAC countries (Argentina, Brazil, 0 69 Belloc and Maio (2011) quote several studies; for instance, Ho- Chile, Colombia, Cuba, Ecuador, El Salvador, Guatemala, Haiti, Hon- U gan (1991), Keesing and Singer (1991), and Low (1982). duras, Uruguay, Peru, Mexico, Trinidad and Tobago), three in North such large economies as Brazil, Chile, Mexico, or Peru (more Research suggests that EpAs help most when export- than 50) but is in the upper band of the Central American ers are inexperienced and small. Evidence from Chile countries (from 30 to 50) and above its Caribbean coun- suggests that export-promotion activities were most terparts (less than 15 offices). Nonetheless, the majority of beneficial for companies with lower exports per country Dominican diplomats and representatives abroad devote or exports per product. A study of Argentinian exporters only part of their time to export-promotion activities. For assisted by EXPORTAR implies that only SMEs benefit from this reason, President Medina in 2013 made a priority of export promotion activities--in particular, companies as- modifying the diplomatic service to add the role of the sisted for the first time. Studies from developed countries commercial agent, specialized in supporting foreign trade. seem to confirm that." On the other hand, Lederman et al Research suggests that regional EPAs have a greater impact (2009) find that firm size had no impact on exports, sug- on export flows than national embassies and consulates gesting that any potential targeting scheme should be (Volpe Martinicus, 2010; Belloc and Maio 2011), although based on evaluation of the particular country's sector pri- results are not conclusive (Lederman et al, 2009). orities. Detailed information on the profile of CEI-RD-sup- ported companies could not be found .14 According to international evidence, EPAs make great- er impact when they address significant information EpAs can also target their actions by the type of good shortages, and they are mostly successful in increasing exported. Volpe Martincus (2010) finds that EPAs abroad the number of destination countries. Support for these and diplomatic foreign missions have larger impact along conclusions comes from Argentina, Costa Rica, Chile, Co- extensive margins in heterogeneous goods" (75 percent of lombia, Peru and Uruguay (Volpe Martincus 2010). In fact, sectors for heterogeneous goods with positive impact) and the higher the trade barriers, the more effective EPAs are the greatest impact was observed in textiles, yarns, fabrics, (Lederman et al 2009). Evidence from firm-level data in Pe- and related products. At the same time, foreign diplomat- ru-a country that like the Dominican Republic has diver- ic missions have a positive effect on promoting homoge- sified product offerings faster than destination markets- neous goods and commodities-among others, textile suggests that companies supported by Peru's PROMPEX fibers, scrap metal, and non-monetary gold. Along these had an export growth rate 17 percent higher than un- lines,CEI-RDcouldtrytolocateexport-promotionofficesin supported firms. Among supported firms, the number of markets that import heterogeneous Dominican products destination countries was 8 percent higher, while number (medical instruments, clothing), while diplomatic missions 0 of products rose 10 percent. Similarly, evidence from Uru- would continue to be the main representation in markets guayan firm-level data suggests that companies that used importing homogeneous goods (metal scrap, gold). URUGUAY XXI had 10 percent more destination countries, although supported companies continued having prob- althughsuporte copanes cntiuedhavig pob- 73 Francis and Collins-Dodd (2004) studied 183 Canadian high-tech lems entering OECD markets. In Argentina and Colombia, SMEs and concluded that companies that export sporadically and actively benefit more from export assistance than established export- a the activities of EXPORTAR were also positively correlat- ers.Similarly,BellocandMaio(2011)aswellasVolpeMartincus(2010) E confirm that export-assistance programs were successful with com- 0 ed with the number of products exported. In Costa Rica, p export assistance had an impact only on companies that from Australian Trade Commission (2002) indicates that 74 percent of companies that used export assistance succeeded, whereas the 1 were already exporting heterogeneous goods and helped success rate stood at 16 percent for non-users. Mexico's experience U suggests that companies find EPAs most relevant when they are in mainly by increasing the number of destination countries. the initial stages of export activities (Belloc and Maio 2011). CL 74 According to data from CEI-RD officials, the institution support- ai ____________________________________________ ed 1,461 companies in 2012 and 1,261 in 2013, but no information America (US, Canada, Puerto Rico), 12 in Europe (Portugal, Switzer- wasavailableonthesizeornatureofthefirms. land, Austria, Holland, Czech Republic, Germany, Belgium, Spain, 75 Homogeneous goods denote goods for which prices are set on France, the UK, Ireland, Sweden), one in Africa (Egypt), four in Asia an organized exchange--e.g. commodities. Heterogeneous goods (South Korea, Taiwan, Malaysia, India ) and two in the Middle East have no reference price and the exchange of information is organized 0 (United Arab Emirates, Israel). by agents. S EPA budgets per capita in LAC (in US$) EPA budgets in LAC (in millions of US$) 2.5 120 2.0 100 1.5 80 60 1.0 0.5 20 Source:~ p 0atnu 21)adClR.Suc:VleMrics(00 n E-D ~ =80 Suc:Volpe Martincus (20 10) and CElPRD. Source: Volpe Martincus (20 10) and CEI-RD. CEI-RD has offered products lines similar to other firm-level data, Volpe Martincus (2010) finds that a strat- agencies in the region, but it has recently begun con- egy combining counseling, assistance with trade agenda, sidering more innovative and tailored approaches. and trade fairs and shows has greater effects on potential The services can be grouped in the following categories: exporters than employingjust one ortwo of the elements. (i) country image building (promotion of the Dominican Similarly, Kotabe (2001) finds that export assistance itself export offering, receiving 3.7% of total budget); (ii) export improved companies' profitability but failed to increase support services (one-stop shop for exporters, technical sales volume, suggesting that export promotion needs to assistance, specialized consulting services, ensuring the be complemented with other supportive activitieS.71 Trade correct application of norms relating to export and invest- shows seem to be relevant for developed markets but ment promotion, and validation of certificates of origin, not so much for developing oneS.77 Since EPAs are more receiving 19.3% of total budget); (iii) marketing (trade fairs efficient when trade barriers are high, tailored market in- and missions, network of representatives abroad); (iv) mar- telligence can benefit exporters. For instance, Mexico's ex- ket intelligence (product profiles and markets, legal and perience suggests that companies find EPAs most relevant economic information). In recent months, CEI-RD has also when providing information on foreign markets and finan- been developing a more pro-active approach in the agri- cial assistance to SMEs (Belloc and Maio 2011). Seringhaus cultural sector. For example, it directly contacts agricultur- and Botschen (1991) confirm that helping companies plan al exporters whose containers were consistently rejected their international engagement in a tailored way yields because they do not meet SPS requirements and offers better results. In a similar fashion, Kotabe (2001) finds that them tailored support. Some LAC countries also offer ad- programs succeed if they address specific needs and help ditional services, such as formation of consortia of export- companies overcome particular obstacles. 0 ing SMEs in similar sector (Argentina) or assistance with product placements on international markets (Colombia) (Volpe Martincus 2010). 76 Kotabe works on a sample of US Midwestern states. For more, c see Belloc and Maio (2011). a) 77 Evidence from the US shows that govern ment-sponsored trade Intenatona reearh sggess tat PAsareabl to shows had a positive and significant effect on exports, while trade SInternational research suggests that EPAs are able to a) missions did not (Belloc and Maio 2011). Alvarez (2004) investigates CL improve their effectiveness when they offer bundled Chilean exporters and finds that trade shows and missions do not 0 shave a significant impact on the probability to export but exporter stailored approaches. Based on Colombian committees do. EPA budgets per employee (in US$) 100000 80000 60000 -2 40000 20000 Peru Chile Costa Rica Ecuador Jamaica Guatemala Argentina El Salvador Panama Honduras Uruguay Paraguay Bolivia DR Source: Volpe Martincus (2010) and (El-RD. CEI-RD has a budget slightly below the regional av- RU partly finances its tourism programs with tax receipts erage, but the payroll burden leaves it with scarce on airline tickets. EPAs in developed countries have larger resources compared to other EPAs in LAC. The CEI-RD budgets and reply on copayments from companies. Al- budget of US$2.4 million was below the regional median though research on how additional EPA resources impact and ranked ninth among 16 countries sampled (Figure export performance is not conclusive, Lederman et al 25). When population is taken into account, the budget (2009) studied 88 EPAs worldwide and found a significant per capita of CEI-RD of US$0.23 is below the regional me- positive impact of EPAs'overall budget on promotion ac- dian of US$0.4 (Figure 24). Finally, CEI-RD seems to have tivities on exports." The authors estimate that a 10% in- a significant payroll burden. Its ratio of EPA budget per crease in an EPA's budget for promoting exports can lead its employee is the lowest among sampled countries. In to a 0.6% increase of exports. However, increasing the re- 2013, the institution devoted 53.4% of its budget to pay- sources of EPAs needs to be done with care because there roll and employed 285 people, higher than the regional are diminishing marginal returns. In addition, extra funds median (97) and similar to Brazil (214) and Colombia (281). tend to have a larger impact on heterogeneous goods 0 Remuneration of CEI-RD employees is fixed and, although than on homogeneous goods. some EPAs such as PROEXPORT in Colombia offer a per- formance-based bonus, research has not found conclusive Finally, a program aimed at fostering a strong EPA, evidence that bonuses have an impact on employee per- with any public intervention, must be accompa- formance. It is worth noting that we are comparing bud- nied by a monitoring and evaluation program for C. gets with the caveat that attributions and legal mandates export-promotion activities. The CEI-RD compiles an E 0 differ across agencies. For instance, Colombia's PROEX- annual accounting of its activities undertaken and the PORT only performs export-promotion duties, while CEl- RD is responsible for export and investment promotion. CEI-RD's budget is entirely funded by the government, ledtoaadionlU$1infrgnses(locndMo20) making it similar to other LAC countries, with excep- TeU td id htaU$ nraei auatrn rmto tion of Mexico and Jamaica. Some EPAs in LAC are also 7Ert promotin is unerstod asolhervics offerd as- fom aretingts Asi ed advice For instnce, Dnis Trade council bdCgepts ndiel thton adonalentspm comnsies Al CEI-D' udgt isentielyfundd bythegovenme thed og ansearchion US$21 aintorignsalEPresoces Miompact) it smilr t oter AC ounties wihecep ThpUsitud findpact f UPs$1 ineaslbet inmncuipromotionc makingcouldtead o addional. US$4e ators emaatiexprts (10%in cesen 204 foundbudgt proroo tingen xpots adn impactd funedfrm peifc txe; orintace,Peu'POME- o ac0.6%ninc'reae iof exporti.oer,icasnthr- number of companies accompanied; however, it would ent and separate entities dealing with trade-related is- be desirable to go further and establish a well-defined sues depending on the legal regime of the company or monitoring and evaluation program. Nevertheless, it is the nature of the issue. In the case of export-promotion worth acknowledging that EPAs in LAC and the devel- activities, for example, at least three institutions are cur- oped world rarely measure the effectiveness of their pro- rently assuming this role, with overlaps. The Ministry of grams and focus on input indicators, such as number of Industry and Trade's DICOEX is in charge of a program to firms using each service or the number of responses to help producers take full advantage of existing free trade requests for support. Output indicators mostly consist of agreements, something also done by the CNZFE. CEI-RD satisfaction surveys that have usually very low response is trying to increase the competitiveness of Dominican rates and may be inflated because companies have an exporters (generally outside SEZs but also inside) by of- incentive to rate EPAs higher, hoping for more funds in fering tailored support, at the same time that promotes the future. It is also worth noting that communication the attraction of investment outside SEZs. Meanwhile, and reporting is an important aspect of the export-pro- CNZFE focuses its effors in attracting multinationals to motion activities. For instance, interviews with manag- SEZs. This structure implies the need for a high degree ers of Australian small businesses revealed that low ef- of coordination to avoid overlaps and may result in sig- ficiency of export-promotion programs can stem from nificant transaction costs that, in the end, hamper policy lack of awareness about such programs. design and implementation. Some countries in the region try to do more to monitor Other institutional arrangements exist throughout and evaluate EPA results. Chile's EPA, supported by the the region and have proven to be efficient in certain IDB, introduced a Balance Scorecard to map how each de- cases. An empowered EPA-both in terms of legal man- partment is fulfilling its objective. Volpe Martincus (2010) date and funds for export promotion-might be a rele- propose two statistical methods to improve the quality of vant option in the Dominican Republic. The CEI-RD could measuring EPAs'impact on exporters: (i) difference-in-dif- fulfill this role, but public support for exporters must be ference, which assesses over time the differences between thoughtfully designed because experiences in develop- companies that used the service and ones that did not; ing countries have been uneven. Drawing from a review and (ii) matching, which pairs a company that received of regional and international best practices, authorities support from EPA with a similar company that has been might consider the following principles when designing not assisted, then compares the outcomes, public programs to promote exports. First, EPAs make a greater impact when they address significant informa- To summarize, the Dominican Republic faces further tion shortages and are mostly successful in increasing liberalization of key sectors of the economy in 2015 the number of destination countries. In fact, the higher in the context of DR-CAFTA implementation and the trade barriers, the more effective EPAs. Second, EPAs WTO compliance, increasing the need to enhance improve their effectiveness when they offer bundled ser- trade competitiveness. With a year to go, the coun- vices and tailored approaches. This can include support- 0 try seems to lack an overarching national strategy ing the formation of consortia of exporting SMEs in simi- for trade competitiveness, export promotion and lar sectors, partly financing their management during the orientation, and investment attraction, and support first years or helping SMEs place their products on inter- to productivity and innovation. More precisely, the national markets, including searching for foreign buyers. two-tier export basket in terms of sophistication and Actually, CEI-RD is moving towards a more tailored and C quality highlighted in the first chapter is mirrored in the pro-active approach with its recent support brought to 0 U country's institutional structure, which includes differ- agricultural exporters to help comply with SPS norms. Having a strengthened (and unified) EPA for promot- ing exports and attracting investment could result in a much more effective use of public resources devoted to these activities in the Dominican Republic. The EPA would work both inside and outside SEZs, with well-de- fined programs and targeting specific sectors and/or des- tination markets. To converge into this model EPA, it will be necessary to clarify the role of the various institutions dealing with export promotion and better balance the EPA budget so payroll does not represent a burden that limits available resources for export-promotion activities. More immediately, the relevant actors in the Domin- ican Republic would need to agree on a strategy for foreign trade competitiveness. To pave the way towards that eventual better empowered EPA, as it has been dis- cussed in subsection c. 1, Institutional Architecture in the Dominican Republic: A Dual Structure, relevant public and private sector bodies shout start aligning strategic plans, establishing standards and procedures for attracting in- vestment (avoiding arbitrary benefits and unequal treat- ment), and coordinating export promotion efforts, as well as other resources (such as information or access to financ- ing). The idea would be to make a set of well-coordinated horizontal policies and tools available for actors both in- side and outside free trade zones. It would be equally im- portant to explicitly support in that strategy the definition of public policies supporting productivity improvements and innovation. Related to this, there could be certain space for the specification of vertical policies focused in key sectors, probably those that are more likely to gener- ate positive externalities for the rest of the economy (tech- nology transfer, learning by doing, forward and backward CL X a1) linkages, etc.). 0 0 0 O LJ 85 Abduliaev, U. and E. Marcello. 2013. "Growth and Em- Baldwin, R. and D. Tagliloni. 2006. "Gravity for Dummies and ployment in the Dominican Republic: Options for a Job- Dummies for Gravity Equations" NBER Working Paper No. Rich Growth." International Monetary Fund Working Paper, 12516. Cambridge MA: National Bureau of Economic Research. WP/1 3/40. Banco Mundial. 2012. "Haiti, Dominican Republic, More ADOZONA, 2012. "Impacto Econ6mico y Social dle las than just the Sum of Two Parts. A research about bilater- Zonas Francas: Una Visi6n dle 360 Grados." La Asodiaci6n al Economy Relations" Country Economic Memoramdum. Dominicana de Zonas Francas. http://www.adozona.org. Washington DC: The World Bank. http://www.worldbank. Accessed in June 2014. org/content/dam/Worldbank/document/HAITI-RD.pdf Alginger, K. 2001. "Measuring the Intensity of Quality Com- BnoMnil 04 Cetn utr facutblt petition in I ndlustries." Austrian Economic Quarterly, WI FO, 6 i h s fpbi eore"Pbi xedtr aae (2):73-101 (May). mn n iaca conaiiyRve.Wsigo C Ali, M. Y. 2006. "Barriers to Export and Export Promotion Programs: Insights from SME Managers." Queensland Uni- BelcM.anM.DMio201"Sryofteieaue versity of Technology, Mimeo. Accessed in: http://anzmac. o ucsflsrtge n rcie o xotpoo org/conference_archive/2006/documents/All_Yunus.pdf to ydvlpn onre. nentoa rwhCne Alvarez, R. E. 2004."Sources of export success in small- and medium-sized enterprises: the impact of public programs." BenrA.R.MsaiJ-.ey,adDTglo.204Ex International Business Review 13 (3): 383-400.potrDnmc,FmSieadGwh,ndPtalYrEf Anderson J. E. and E. van Wincoop. 2003. "Gravity with EcnmcRsah,I. Gravitas: A Solution to the Border Puzzle." American Eco- nomic Review, 93 (1): 170-92.Benr,AB.anJ.BJesn204"Wyomfisex Baissac, C. 2011. "Planned Obsolescence? Export Process- reuoEcnmcRsah,I. ing Zones and Structural Reform in Mauritius." In Special 6 Economic Zones: Progress, Emerging Challenges and Future BrhuA n .Elne.21."h rd ntVle Directions, eds. T. Farole, and G. Akincl. Washington, DC:dabseEcnmeItrtial,CPIeerhcne, Thel World Bank (U9oin-Pblcaio. Besedeg,T., and T. J. 2006."Product differentiation and du- CNMSF 2005. "Operating bylaws of the national commit- ration of US import trade"JournaloflnternationalEconom- tee for the Implementation of sanitary and phytosanitary ics 70 (2): 339-58. measures of the Dominican Republic.' Comit6 Nacional de las Medidlas Sanitarias y Fitosanitarias. Accessed in June Brenton, P., C. Saborowski, and E. Von Uexkull. 2010. 30, 2014. http://www.cnmsf.gob.do/Portals/0/docs/Rega- "What Explains the Low Survival Rate of Developing mento%200perativo%20CNMSF.pdf Country Export Flows?" The World Bank Economic Review 24 (3): 474-99. De Battisti, A. B., J. MacGregor, & A. Graffham. 2009. Stan- dard bearers: Horticultural exports and private standards in Buono, I., and H. Fadinger. 2012. "The Micro Dynamics of Africa. lIED. Exporting - Evidence from French Firms." Temi di discussi- one (Economic working papers) 880, Bank of Italy, Economic Eaton, J., M. Eslava, M. Kugler, and J. Tybout. 2008. "Export Research and International Relations Area. Dynamics in Colombia: Firm-Level Evidence;' in The Orga- nization of Firms in a Global Economy, ed. by E. Helpman, Burgaud, J-M., and T. Farole. 2011. "When Trade Preferenc- D. Mann, and T. Verdier. University of Chicago Press. ISBN es and Tax Breaks Are No Longer Enough: The Challenge 9780674030817. of Adjustment in the Dominican Republic's Free Zones!' In Special Economic Zones: Progress, Emerging Challenges, and Farole, T. 2013. "Raising taxes on EPZ firms and complying Future Directions, eds. Farole and Akinci. Washington DC: with WTO.'World Bank, mimeo. World Bank. Farole, T., and G. Akinci, eds. 2011. Special Economic Zones: Buzby, J., L. Unnevehr, and D. Roberts. 2008. "Food Safety Progress, Emerging Challenges and Future Directions. Wash- and Imports: An Analysis of FDA Food-Related Import Re- ington DC World Bank. fusal Reports" Economic Information Bulletin 58626, Unit- ed States Department of Agriculture, Economic Research Feenstra, R. C. 2004. Advanced International Trade: Theory Service. and Evidence. Princeton, NJ: Princeton University Press. Cebeci, T., A. M. Fernandes, C. Freund, and M. D. Pierola. Francis, June and Collins-Dodd, Colleen, Impact of Export 2012. "Exporter Dynamics Database;' Policy research work- Promotion Programs on Firm Competencies, Strategies ing paper 6229. Washington DC: The World Bank. and Performance: The Case of Canadian High-Technology Smes (2004). International Marketing Review, Vol. 21, Issue 0 -a CEI-RD. 2009. "Estudio sobre el comercio fronterizo entre 4/5, p. 474-495 2004. Available at SSRN: http://ssrn.com/ Rep6~blica Dominicana y Ia Rep6~blica dle Haiti: Propuesta abstract= 1503893 CL a1) Para Ia Formalizaci6n y Humanizaci6n del Mercado Bina- 0 cional. Centro dle Exportaci6n e Inversi6n de Ia Rep6blica Gaulier, G., G. Santoni, D. Taglioni, and S. Zignago. 2013." n Dominicana. the wake of the global crisis: evidence from a new quar- terly database of export competitiveness Policy Research Central Bank- Banco Central dle Ia Rep6blica Dominicana. WorkingPaperSeries6733,Washington DC:The World Bank. 2014. "Censo Econ6mico del Sector ne Zonas Francas2 De- partamento International, Enero 2014. Accessed in June Gereffi, G., J. Humphrey and T. Sturgeon. 2005. "The gover- 30, 2014. http://www.cnzfe.gob.do/index.php/cnzfe/pub- nance of global value chains' Review of International Politi- 0 licaciones/censo-economico-del-sector-zona-franca cal Economy, 12 (1): 78-104. Gorg H., M. Henry, and E. Strobl. 2008."Grant Support and Jouanjean, M-A., and A. Le Vernoy. 2010. "Trade and SPS Exporting Activity:' Review of Economics and Statistics 90 Regulations: The Importance of Being Earnest?" 84th An- (1):168-74. nual Conference, March 29-31, 2010, Edinburgh, Scotland 92000, Agricultural Economics Society. Hamlin, Margaret. 2011. "Similar Policies, Different Out- comes: US Policies toward Haiti and the Dominican Repub- Jouanjean, M-A., J-C Maur, and B. Shepherd. 2012. "Repu- lic." Dietrich College Honors Theses, Paper 124. Accessed in tation Matters: spillover effects in the Enforcement of US June 30, 2014. http://repository.cmu.edu/hsshonors/1 24. SPS measures." Policy Research Working Paper Series 5935. Washington DC: The World Bank. Hauser Ch., and A. Werner. 2010. "Fostering International Entrepreneurship: Are SMEs targeted adequately by offi- Kaplinsky, R. 1993."Export Processing Zones in the Domin- cial foreign trade promotion schemes?" SSRN Working Pa- ican Republic: Transforming manufactures into commodi- per, 1120824. ties," World Development 21 (11): 1851-65. Hausmann, R. and J. Hwang, and D. Rodrik. 2006. "What Koopman R., Z. Wang, and S-J Wei. 2008. "How Much of you export matters"CEPR Discussion Papers 5444. Chinese Exports is Really Made in China? Assessing Do- mestic Value-Added When Processing Trade is Pervasive." Hausmann, R., C.A. Hidalgo, J. Jim6nez, R. Lawrence, E. NBER Working Paper 14109 Levey-Yeyati, C. Sabel, and D. Schydlowsky. 2012. "Con- struyendo un mejor futuro para la Republica Dominicana: Lall, S. 2000."TheTechnological structure and performance Herramientas para el desarrollo" CID Working Papers. Har- of developing country manufactured exports, 1985-98' vard University, July. Oxford development studies 28 (3), 337-69. Helpman, E., M. Melitz and Y. Rubinstein. 2008. "Estimating Lawless, M., 2009. "Firm Export Dynamics and the Geogra- Trade Flows: Trading Partners and Trading Volumes." The phy of Trade;' Journal of International Economics, Elsevier Quarterly Journal of Economics 113 (2): 441-87. 77 (2): 245-54 (Abril). IDB. 2012. Public-Private Partnerships to enhance SPS Lederman, D. and W. F. Maloney. 2012. Does What you Ex- capacity: What can we learn from this collaborative ap- port Matter? In Search of Empirical Guidance for Industrial proach? Joint document of the Standards and Trade De- Policies. Washington DC:The World Bank. velopment Facility and the Inter-American Development Bank. Lederman, D., M. Olarreaga, and L. Payton. 2009. "Export Promotion Agencies Revisited." The World Bank Policy Re- J. F. Hornbeck. 2009. "The Haitian Economy and the HOPE search Working Paper 5125. Washington DC: World Bank. Act." CRS Report for Congress, Congressional Research Ser- vice, 7-5700, RL34687, available at http://fpc.state.gov/ Manova K., and Z. Zhang. 2012. "Export Prices across documents/organization/145132.pdf. Accessed in April Firms and Destinations" Quarterly Journal of Economics 30, 2014. 127: 379-436. Ln U Jouanjean, M-A. 2012. "Standards, reputation, and trade: Manzano, Osmel; Christine Richaud; Rudy Loo- evidence from US horticultural import refusals"' The World Kung Aguero; and Miguel Eduardo Sanchez Mar- Trade Review 11(3): 438-61. tin. 2013. Rep0blica Dominicana - notas de politi- ca: versi6n para el dia del didlogo. Washington DC: ed for export to the European Union, European Commis- World Bank Group. http://documents.worldbank. sion Health and Consumers Directorate-General (SANCO) org/curated/en/201 3/03/19436485/rep%C3%BAbli- 2010-8588 - MR FINAL. ca-dominicana-notas-de-politica-versi%C3%B3n-pa- ra-el-d%C3%ADa-del-di%C3%A1 logo. S6nchez-Ancochea, D. 2006. "Development Trajectories and New Comparative Advantages: Costa Rica and the Ministerio de Industria y Comercio. 2013. Plan EstratOgi- Dominican Republic under Globalization". World Develop- co Institucional, 2013-2017. Santo Domingo, R.D. http:// ment, Elsevier, 34(6):996-1015 (June). www.seic.gov.do/transparencia/plan-estrat%C3%A9gico/ planificaci%C3%B3n-estrat%C3%A9gica.aspx. Accessed in SSnchez-Ancochea, D. 2012. "A Fast Herd and a Slow Tor- April 30, 2014. toise? The Challenge of Upgrading in the Dominican Re- public.' Studies in Comparative International Development Molina A.C., M. Bussolo, and L. lacovone. 2010. "The DR- 47: 208-30. CAFTA and the extensive margin: a firm-level analysis"Pol- icy Research Working Paper Series 5340. Washington DC: Schrank, A. 2005. "Entrepreneurship, Export Diversifica- The World Bank. tion, and Economic Reform: The Birth of a Developmental Community in the Dominican Republic.' Comparative Poli- Murray, D. and P. Hoppin. 1992."Recurring Contradictions tics 38 (1): 43-62. in Agrarian Development: Pesticides Problems in Caribbe- an Basin Nontraditional Agriculture.' World Development Senderowitsch, Roby ed.; Tsikata, Yvonne M ed. 2010. 20 (4): 597-608. From the international financial crisis towards inclu- sive growth in the Dominican Republic. Washington DC: Murray, D. L. 1994. Cultivating crisis: The human cost of pes- World Bank. http://documents.worldbank.org/curatedl ticides in Latin America. University of Texas Press. en/201 0/01/17193928/international-financial-crisis-to- wards-inclusive-growth-dominican-republic Reis, J.G., and Farole, Th. 2012. Trade competitiveness diag- nostic toolkit. Washington DC: The World Bank. SIRA-BID. 2011. "Diseho y formulaci6n de lineamientos para dlesarrollar un plan nacional de transporte de car- Rodrik, D. 1999. The New Global Economy and Developing ga en Ia Rep6blica Dominicana.' Ministerio de Economfa, Countries: Making Openness Work. London: Overseas De- Planificaci6n y Desarrollo, Unidad Ejecutora SIRA-BID ATN. velopment Council. Santo Domingo, December 2011. Rufin, C., D. Zucchini, R. Senderowitsch, and M.E. 56n- Studwell,J.2013.HowAsia Works:Success andFailure in the X chez. 2014."The Dominican Republic: Moving from Exit to World's Most Dynamic Region. NewYork: Grove Press. 0 Voice-Shifting Incentives in the Power Sector." In Prob- lem-Driven Political Economy Analysis: The World Bank's Ex- Subervie J., and 1. Vagneron. 2013. A drop of water in the perience, eds. V. Fritz, B. Levy, and R. Ort. Washington DC: Indian Ocean?The impact of Global Gap certification on ly- The World Bank, pp. 91-117. chee farmers in Madagascar. World Development, 50:57-73. SANCO. 2010. Final report of a mission carried out in the Do- UNICEF, MEPyD, and CONANI. 2012. "An6lisis de Situaci6n minican Republic from 08 to 23 November 2010 in order to de Ia Infancia y Ia Adolescencia en Ia Rep6blica Domini- 0 evaluate controls of pesticides in food of plant origin intend- cana Report by the Children's Rights & Emergency Relief sion ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~8 HelhadCnuesDretrt-eea9SNO Organization (UNICEF), Ministerio de Economia, Planifi- World Trade Organization. 2006. Exploring the Links be- caci6n y Desarrollo (MEPyD) and Consejo Nacional para la tween Subsidies, Trade and WTO. Geneva: World Trade Nifiez y Adolescencia (CONANI), available at http://www. Report. unicef.org/republicadominicana/Analisis Situacion- Ninez_corregido_agol3_web(1).pdf. World Trade Organization. 2008. Trade in a globalizing World. Geneva: World Trade Report. UNIDO. 2012. "Dominican Republic Trade Standards Compliance Footprint"' http://www.unido.org/fileadmin/ user_media_upgrade/What_we_do/Topics/Quality_and compliance/TSC_Footprint_Dominican_Republic.pdf. Ac- cessed in April 30, 2014. USAID. 2005. "Guia Logistica de la Republica Domini- cana" http://www.competitividad.org.do/wp-content/ uploads/2009/01/guia-logistica-de-la-republica-domini- cana-usaid-2005.pdf. Accessed in September 2005. USAID. 2008. "Impacto del Transporte y la Logistica en el Comercio Internacional de la Repiblica Dominicana. Hor- talizas Orientales e Invernaderos:" http://www.competi- tividad.org.do/wp-content/uploads/2009/01/guia-logisti- ca-de-la-republica-dominicana-usaid-2005.pdf. Accessed in April 30, 2014. Volpe Martincus, Ch. 2010. Odyssey in International Mar- kets: An Assessment of the Effectiveness of Export Promotion in Latin America and the Caribbean. Washington DC: In- ter-American Development Bank. Wade, R. 1990. "Governing the Market: Economic Theory and the Role of Government in East Asian Industrializa- tion." Princeton University Press. WagI, 5. 2005. "International Trade in Textiles and Cloth- ing and Development Policy Options." Policy paper for the UNDP Regional Center in Colombo, January 2005. Waters, J. 2013. "Achieving World Trade Organization Com- pliance for Export Processing Zones while Maintaining U Economic Competitiveness for Developing Countries.' Duke Law Journal 63:481. AN EE ANNEX I: YearAII products No Oil Products 2007 83 88 2008 87 95 2009 106 106 2010 113 111 2011 104 104 2012 105 98 Source: Author's calculations using DGA data and COM TRADE. ANNEX 2: Export growth is computed using the so-called "mid-point The G measure is monotonically related to the conven- growth rate" (Davis and Haltiwanger, 1992; as applied to tional growth rate measure (g,), and it represents a good firm-level exports by Bricongne et al., 2011 and Reyes and approximation of the latter for small growth rates. Both Taglioni, 2012). A main advantage of this methodology growth measures are linked by the following identity: is that it allows estimating growth rates associated with newly created or destroyed flows. Other methodologies 2grt that use normal growth rates are unable to do so; for this (2 - grt) reason, they rely on probabilistic methodologies to quan- tify the impact of determinants of exports at the extensive At the aggregate level, the index well approximates the margin. standard measures of growth rate. Mid-point growth rates are computed as follows. For a firm In the Dominican Republic, non-SEZ exporters show i exporting a value x to a country c of product k at year t, high dynamism in terms of the entry and exit of firms, the mid-point growth rate is defined as: products, and markets. To provide more precise and detailed information on the patterns emerging from the gickta= Xickt - Xick(t-1) decomposition of export growth for firms outside SEZs, (xickt + Xick(t-1)) Table 14 reports actual contribution to overall year-to- year export growth in 201 0-12 and each margin of trade. This growth rate measure is symmetric around zero, and Within each margin, we further report the composition it lies in the closed interval [-2,2] with exits (entries) corre- of entries and exits. The data have been computed us- sponding to the left (right) endpoint. Similarly, the weight ing the so-called "mid-point growth rate" and correcting 0 attributed to each flow gickt is given by the relative share for the bias induced by the partial-year effect in the entry of the flow in total exports (the exports of the whole pop- of new export relationships. The table reads as follows: In ulation of a country's firms): each column, we report the contribution of each margin (in percentages) to the year-to-year annual growth rate. Xickt + Xick(tv) For example, looking at annual growth between 2009 and C. Sickt =detaid2010 (first data column), the first row reports that changes ~0 tination) contributed a net 75.8% to total export growth. u Finally, the year-on-year growth rate of the total export This number is the difference between increments of total value is given by summing each individual flow g,ck export growth within export relationships (169.8%) and a weighted by sckf: reductions within those relationships (94.1%). Subsequent C rows replicate this analysis for the different dimension of =Gt * the extensive margin. The first dimension is the contribu- Fork exmpe lokngainulcrwhkewet20 n 0. i k tion of firm churning. The decline in export growth due Intensive Margin 75.8 29.9 6.7 Intensive Positive 169.8 160.5 208.1 Intensive Negative -94.1 -130.6 -201.4 Extensive Margin 24.2 70.1 93.3 Net Firms 13.4 17.3 44.0 Firm Entry 50.2 25.2 68.5 Firm Exit -36.8 -7.8 -24.5 Net Country 2.5 38.2 2.7 Country Entry 15.7 56.2 35.8 Country Exit -13.3 -18.0 -33.0 Net Products 8.4 14.6 46.6 Product Entry 24.9 70.7 261.7 Product Exit -16.5 -56.1 -215.1 Note: This table decomposes annual export growth into changes in established export relationship (intensive margin) and changes in new export relationships (extensive margin). The extensive margin is divided into three mutually exclusive categories: (i) export growth due to new firms entering (exiting) exporting markets; (ii) export growth due to established exporters entering (exiting) into (from) new market; and (iii) export growth due to established exporters adding (dropping) a product into an existing market. Growth rates are computed from data in US dollars. Source: Authors'calculations based on DGA data. to the exit of firms from exporting represented 36.8% of gives the overall extensive margin.79 In the case, we find it overall export value and the gain due to entry was equal accounts for 24.2% of total growth in export value."o to 50.2%-so the net contribution of firm churning was 13.4%. The decision of existing exporters to enter new foreign markets (country entry) or to exit those markets (country exit) contributed 2.6% to total export growth. In this case, entry into new markets brought 15.7% export growth, partly offset by exits that accounted for a loss of 13.3% of export growth. Next, we look at the decision of 13.3 ofexpot gowt. Net, e lok a th decsio of 79 This decomposition involves a hierarchy along the different di- existing exporters to expand their product range in mar- mensions of the extensive margin. That is to say, country entry and exit is contingent on a firm already being an exporter, and product kets they already serve (product entry) or to reduce their entry and exit is contingent on a firm already being an exporter serv- offerings (product exit). The net contribution of product ing the same market. In this way, we prevent double counting ot new a) export relationships. We keep this pecking order throughout the re- a) churning is 8.4% of export growth.The sum of the net con- port. 80 Note that the contributions along the intensive and extensive tribution of firm, country, and product extensive margin margins sum to 100. exti9otneto4afr led en n xotr n rdc ANNEX 3: SF N PRO UCTS F 32. Quality ladders for Top 3 manufactured products MEDICAL INSTRUMENTS HS:901890: quality ladder, Av. 2005-2006 HS:901890: quality ladder, Av. 2009-2010 2 2 1.5 1.5 .5 .5 0 * * _ 0 0 20 40 60 80 0 20 40 60 80 Rank Rank ELECTRICAL SWITCHES HS:853620: quality ladder, Av. 2005-2006 HS:853620: quality ladder, Av. 2009-2010 1.5 1.5 .5 .5 I 0 0 0 * 0 * 0 10 20 30 40 0 10 20 30 40 50 Rank Rank 0 ARTICLES OF JEWELRY HS: 711319: quality ladder, Av. 2005-2006 HS:711319: quality ladder, Av. 2009-2010 2 2 2a 1.. 1.5 E 1.5 o see U .5 .5 0 **00 0 0 20 40 60 0 20 40 60 Rank Rank 0 Source: Authors'calculations using CEP1l data. 0 ANNEX 4: N CSS TO THE US FOR AFTA OUT CAFTA country PoutYear of entry CAFTA country PoutYear of entry of new product of new product The Dominican -- -- Costa Rica Brassica spp. 2001 Republic Guatemala Rhubarb 1998 Papaya 2001 Papaya 2001 Rambutan 2003 Rambutan 2003 Pepper 2006 Lotus root 2003 Tomato 2006 Tomato 2006 Nicaragua Green bean 1997 Pepper 2006 Mung bean 1997 El Salvador Bean, garden 1995 Faba bean 1997 Eggplant 1997 Radicchio 1997 Brassica spp. 1998 Eggplant 1997 Papaya 2001 Parsley 1998 Lotus root 2003 Brassica spp. 1998 Rambutan 2003 Papaya 2001 Loroco 2003 Yam bean 2003 Yam bean 2003 Yard-long bean 2003 Parsley 2003 Rambutan 2003 Pepper 2006 Lotus root 2003 Tomato 2006 Loroco 2003 Cichorium spp. 2006 Tomato 2006 Pepper 2006 Note: This table lists all products that have been granted access to the US from CAFTA -DR countries over the lost 20 years. Source: Authors collection using the US Fruits and Vegetables Import Requirements (FAVIR) database (http://www.aphis.usda.gov/favir/). ANNEX 5:TTO P International trade research finds that exporters start small We use a theory-grounded gravity model to evaluate the and grow very rapidly in their first year in foreign markets Dominican Republic's pair-wise export relationships with (for example, Eaton, Eslava, Kugler and Tybout (2008) for their trading partners. The gravity model has been exten- Colombian exporters, Lawless (2009) for Irish exporters, sively used in international trade due to its intuitive em- Buono and Fadinger (2012) for French exporters, Molina, pirical and theoretical appeal. Anderson and van Wincoop Bussolo and lacovone (2010) for the Dominican Republic, (2003), Feenstra (2004), and Baldwin and Taglioni (2006), and Cebeci, Fernandes, Freund, and Pierola (2012) for a among others, present exhaustive reviews of research on cross-country analysis).This"stylized"fact is established by the gravity equation as applied to international trade. Our collapsing all export transaction at the firm level in a given specification of the gravity model followsthe micro-found- calendar year. Bernard, Massari, Reyes, and Taglioni (2014) ed model of Helpman, Melitz, and Rubinsten (2008). Spe- show that these findings are largely explained by the fact cifically, we regress 2006-12 bilateral exports among 213 that exporters enter the market throughout the year and countries on the following bilateral characteristics: dis- only part of the first year sales are recorded in the calen- tance, contiguity, common language, colony, common dar year of entry. Two otherwise identical firms that enter colonial power, as well as log of GDP, and log of GDP per the same market in different months, one in January and capita. The model incorporates three main components: one in December, will report dramatically different annu- First, a measure of remoteness is computed by summing al sales for the first calendar year of operations. This par- distances weighted by the share of GDP of the destination tial-year effect in annual data leads to downwardly biased in world GDP. This is to take note of the fact that relative observations of the level of activity upon entry and the fol- distances matter greatly, alongside absolute distances. lowing year. Using Peruvian export data, the authors find Second, we control for zero trade Rows with the use of the that the partial-year bias is substantially high: the average Heckman sample selection correction method. When ob- 0 level of first-year export of new exporters is understated servations with non-existent bilateral trade are dropped, by 65%, and the average growth rate between the first as OLS does, our dependent variable is not really measur- and second year of exporting is overstated by 112%. As a ing bilateral trade but one contingent on a relationship consequence, the use of annual data largely understates existing. Therefore, an important variable left out of the the contribution of the extensive margin in annual export model is the probability of being included in the sample; C. growth. We correct for partial-year effects by collapsing i.e., having a non-zero trade flow. To the extent that the E 0 total export value at the firm-product-destination level by probability of selection is correlated with GDP or distance, year of operation of the firm. Therefore, a firm that enters this has the potential to bias OLS estimates. Third, we con- a market in December would receive the value of exports trol for firm heterogeneity without using firm-level data by between December and November of the following year recognizing the fact that the features of marginal export- as its first-year export value. ers can be inferred from the export destinations reached. With these steps, the gravity results are better grounded on modern trade theory. 0 ed odl f elpan Mliz, ndRuinten(208. pe six-digit level). When product quality is not taken into ac- count, EXPY overestimates the importance of sophisticat- ed products from low-income countries. Xu (2007) shows Calculating export sophistication (EXPY) is a two-stage that once products at the HS six-digit level are further process. The first stage involves measuring the income divided by relative unit values, the structure of China's ex- level associated with each product in the world (PRODY). ports is consistent with its level of development. This has The PRODY of a particular product is the GDP per capita of led authors like Lederman and Maloney (2012) to point out the typical country that exports it, calculated by weighting that how a country produces an export matters more than the GDP per capita of all countries exporting the good.The what it produces. Seemingly high-tech products like com- weight given to each country is based on revealed com- puters can be produced in low-tech ways and vice-versa. parative advantage.The PRODY for a single product is cal- culated by weighting the GDP per capita of all countries The fragmentation of production poses another problem. exporting that product. Therefore, a product that typically While the final export of a sophisticated product might be makes up a large percentage of a poor country's export from a low-income country, its contribution might have basket will have stronger weights towards poor countries' just been in the final assembly of high-value intermediate GDP per capita.This will be less the case for a product that inputs made elsewhere. One should not, therefore, lose makes up a small percentage of a poor country's exports sight of the entire value chain and explore which stage but is a significant component of many rich countries'ex- of production creates and captures the greatest value. If port baskets. computers are deemed not to be sophisticated because the final assembled package is exported from a low-in- The second stage is to measure the income associated come country, the innards could be highly skill-intensive, with a country's export basket as a whole (EXPY).The EXPY possibly imported from richer countries. Koopman et al. is calculated by weighting these PRODY of each product (2008) estimate the foreign content in China's exports to by the share that each good contributes to total exports. be about 50% overall and 80% percent in sophisticated For example, if butter makes up 15% of a country's ex- products like electronic devices. In the well-known ex- ports, its PRODY will be given a weight of 0.15. Countries ample of the iPod, an overwhelming share of the final as- whose export baskets are made up of "rich-country goods" sembled value of an iPod exported from China is captured will have a higher EXPY, while export baskets made up of by the creators of intellectual property, not in the form of "poor-country goods"will have a lower EXPY. wages earned by the assemblers. More recently, Hausmann, Hidalgo et al (2011) have im- PRODY, W2 Y. and EXP1. = PRODY proved on this measure by moving away from the reliance of PRODY and EXPY on the income levels of countries. haUnder the new approach-the Product Complexity In- The concepts of PRODY and EXPY are, however, not free of dwex-complexity is not a function of countries' incomes; criticism. The PRODY of some products are counter-intu- instead, it is calculated through an iterative process based itively high, suggesting sophistication in products mere- on the countries' network of relationships and the prod- ly because rich countries produce them. Bacon and ham, ucts they export. Specifically, under this new approach the for example, have a higher PRODY than internal combus- complexity of a country's export basked is a function of tion engines. Further, the quality of products varies-cars two concepts: (i) the diversity of products it produces (i.e. from Country X may not be the same quality as cars from the number of distinct products that it makes); and (ii) the C: C: CountryY (even if they all have an identical code at the HS ubiquity of those products (i.e. how many other countries posbyipre9fo ihrcutie.Komne l make that product). It is generally observed that a county influenced by advertising or reputation, rather than intrin- that can produce a complex product that few other coun- sic characteristics of the goods traded-and some argue tries are able to produce will also produce a wide range of that this should not be considered quality differentiation. non-complex products; the opposite is rare. Thus, a prod- Third, price differences and dynamics could be attributable uct made by only a few countries that also produce a wide to market structure, i.e. heterogeneity in market power. range of other products is relatively complex. In contrast, a product that is produced by most countries, including With these caveats in mind, we propose a method to ex- those that produce few other products, is less complex. ploit the information from unit values. It relies on the Ex- port Unit Values database from the French Centre for Re- search and Studies on the World Economy (CEPII) (Berthou and Emlinger, 2011) to characterize the relative unit values of agricultural imports in the US. As in Schott (2004), unit Goods in the same product category vary widely in quality, values were calculated simply as the quotient of general a fact proxied by unit values (the ratio of values to quanti- imports values and quantities. Within any product (six-dig- ties shipped, or nominal sales divided by quantity). A key it HS codes) for any given year, we then have a distribution finding of research on within-product specialization is the of unit values of imports from various source countries. large differences in unit values across countries, even with- For each good i and exporting country c, in time year t, we in narrowly defined products shipped to the same market generate a measure of relative quality R as: within the same year. Moreover, these differences are per- sistent over time. Ric 90 A key challenge for international economics in recent years has been establishing the reasons for this variation Where u denotes de unit value of the good and u90de- in unit values across countries and across firms serving notes the value at the 90th percentile of the unit value the same destination. One dominant assumption is that distribution across countries for that product. Ri,, denotes unit-value differences may be due to vertical differentia- the relative quality of the country's export of that good, tion in the form of quality differences. According to this i.e., quality relative to other countries exporting the same reasoning, if two firms serve the same destination with good. 0 the same narrowly defined product and they manage to do so by selling their products at substantially different price levels, it suggests a substantial quality difference in what these two firms sell. If the goods were equal in quali- ty, the reasoning goes, competition should eliminate price C. differences. However, there are several possible alterna- E 0 tive explanations. First, it is possible that bilateral trade data-even at the finest level of disaggregation allowed by international trade classifications-are too coarse to Y capture the fact that different countries and different firms are really specializing in different goods. They only appear to be the same product because international trade statis- tics use commodity classification that are too aggregate. Second, price differences could reflect quality perceptions 0 ANNEX 6: S\ ULTS F L I L An analysis of micro data using World Bank Enterprise electricity connections are similar to other parts of the Surveys reveals other barriers for Dominican exporters. world at an average of 40 days-except for the Caribbean This report has thoroughly assessed three bottlenecks for (18 days). However, power outages average 16 a month, export competitiveness: agro exporters' difficulty in com- much more frequent than in other regions, with the ex- plying with SPS standards, a two-tier export basket with ception of the Middle East and SAR (the developing world limited linkages between SEZs and the domestic economy, average is 10). In addition, Dominican companies reported and fragmented institutional support efforts. In this section, on average of 2.1 water insufficiencies a month, compared we briefly look at other potential constraints that emerged to 0.6 in Latin America. Exporters also consider electricity in meetings with stakeholders but could not be included as more of an obstacle to daily operations than in other in this assessment. Some were beyond the scope of the parts of the world, except for Central America. A range of study (electricity sector and labor market factors factors), problems plague the electricity sector-from high techni- others were already covered in a number of studies (trans- cal losses in distribution to insufficient payment recovery, port and logistics). To briefly review these and other issues customer fraud, and high variance in the price of energy of the business environment, we have used World Bank-IFC purchases from private generators. Several studies discuss Enterprise Surveys to compare the obstacles faced by Do- in more detail some of the current challenges in the elec- minican exporters with those in other parts of the world. tricity sector (for instance, Manzano et al, 2013; Rufin et al, Results are presented in tables at the end of this annex. 2014; World Bank, 2014). Dominican exporting companies are younger but larg- The Dominican telecommunications infrastructure is er than those of its neighbors. Measured by the number generally perceived to be as good as most other parts of workers, Dominican exporters tend to be larger than the of the world-and, in fact, better than the rest of the country's non-exporters and exporting peers from other Caribbean. In turn, the transportation infrastructure is regions-with the exception of South Asia (SAR). The Do- more of an obstacle in the Dominican Republic than in minican Republic joins the East Asia and Pacific region in other Caribbean islands-but no more than in the overall having exporters with a higher percentage of foreign own- Central America and Caribbean region. It is nonetheless ership than the Caribbean, Central America, South America worth noting that Dominican gasoline prices are higher (plus Mexico), and Europe and Central Asia.This is likely the than in any other CAFTA-DR members; the pump price result of a large presence of foreign companies in SEZs. At is at around $1.58 per liter, significantly higher than the the same time, average Dominican exporters are younger, $0.97 per liter in the US."1 This study has not focused in withsevenfeweryearsinoperationthanCaribbeanpexport- transportation and logistics because other donors, such ers, six fewer years than Central American exporters, and as the Inter-American Development Bank (among others, 10 fewer years than South American exporters. However, SIRA-BID, 2011) and USAID (2005; 2008, have already pro- Dominican export firms' managers have average years of duced several assessments on the topic. experience that is similar to other countries in the region. On the institutional side, Dominican exporters seem to Access to reliable electricity seems to be the most pro- mainlyfacetax administration and informality hurdles. X f nounced infrastructure obstacle for Dominican export- C: C: iers. According to surveyed exporters, wait times to obtain 81 http://data.wordbank.org/indicator/EPMSGAS.CD. Tax rates and tax administration are perceived as more of to 117th in 2014. Doing Business dimensions in which the an obstacle in the Dominican Republic than in other parts Dominican Republic has more room for improvement are: of the world. Taxes seem to be especially burdensome for resolving insolvency (159th out of 189 countries), start- service exporters, and less so for exporters of manufac- ing a business (144th), dealing with construction permits tured products. When compared to non-exporters, Do- (121st), registering property (115th), paying taxes (106th), minican exporting companies see competition from the and protecting investors (98th). The World Bank is current- informal sector as more of a constraint. Other institutional ly supporting the Dominican Republic through technical factors, such as access to finance and business licensing assistance in some of these dimensions. and permits, are not perceived by Dominican exporters as more of an obstacle than in other regions, although other Dominican companies regard these issues as significant constraints. Dominican exporters perceive the court sys- tem as fairer than their counterparts in Central America, but Dominican exporters lag other Caribbean countries on this metric. The limited availability of highly skilled human capital may be a constraint in the future. Dominican exporting companies have the region's lowest percentage of workers with high school diplomas (with the exception of MENA) and the lowest percentage of skilled workers-on aver- age, 39% in the Dominican Republic, compared to 60% in Central America, 63% in the Caribbean, and 60% in South America. Dominican exporters perceive the lack of ade- quately educated workforce as more of an obstacle than their counterparts in Central America, East Asia and Pacif- ic, and Europe and Central Asia. This is striking since, as concluded in the Trade Outcomes Analysis (part I of this 0 report), the Dominican Republic exports products that are relatively more sophisticated than other countries in Cen- tral America.The lack of a skilled pool of workers may ham- per the chances of Dominican exporters to further climb the value added ladder. C. 0 Finally, lack of improvement of the Dominican Do- ing Business ratings could result in the country and its SEZs being perceived as less attractive by foreign investors.82 According to the World Bank's latest Doing Business indicators, the Dominican Republic's relative rank among 189 economies has dropped from 11 2nd in 2013 82 http://www.doingbusiness.org/data/exploreeconomies/domin- 0 0 ica n-republic. ** * * ** * * c= c= CD CD 6O CD C = D c>C n C n C C cx >r r CD 00 C)CDC * * * * * ** * * * * * * * * * * * * * * * O= m La O in r"" O L r"-- G5 N O """ O O O "-- O i M O C * * * * * * * * * * * * * ** * * * * * * *- i * i * * * * ** * * * * C * * - Cý C0 <= Cý C CO C* Cý Cý Cý Cý C ' <= Cý ' CO Cý - y r',1 Cý0 ýC = * * * * * * * * * * * * * * * * * * * * * * * * C "' C> CD C """ C> C """ CD C CD C> CD C) C> C> C> CD C> C> CD CD C> C> CD CD ID C> CD C) C> C> C> CD C> C> CD C> CD * * ** *** * * * * * * * * * * * * ** * * * * * * * * * * * * * * * * * * m o o o o o o o o, o *o O : I o l o 6 o N o - o D o o*o oo *< mo o Oý oD o6 oo 1o o o * o C>o c|* *'- * * * * * * * * * * * * * y . * * ** * * * * * * * * * * * * * * * * * * ** * * * * * * * * * * * * * * * * * c: o= m c: ýo c> o o: o o o oD o o o o) o) c> 0, ,: o > a, C) o = o o oD o o o c> o o D o o o D * * * * * * * ** * * * * * * * * * * * * * * * * * ~~c Ný '0 <= .' cý '0cý' - ~ . 0 ~ O M O O I N O O I- N '" .." CO N L0 L0 ' NO rO . -. _ - N O O i 0' M N O Ot O rO rT C O n O O CO O~ N M O' O '0 O 0' N 0' 0 r. o o o on o o" o o o o o o r, o' o o o o o o o o o o0o o - o n oO o C5 -- o>C- O N 0" C "0 N c- n - - N c> - oo - c Oý cý O= I O= I1. C> cD O On ON O1 c' L> cq" ,:t O CO 1""" O O O O "" Lrý O ci' Oý O= Laý N O IZI Oý O Oý O '0N C O 0' - 0'"O- O c:> - N r- C - A* L- c O 'N r-n <= c- < r- N N C- LD D - cD I 0 O CO en e .. -C -O NCO0' CO \O O- '0 O N ~ e N C O I.0 O- NO N - " CO - c D O L r- - - N c) N N - . à . . . . I tr . . "' I . :- . . r- N e . n n r-. c-- . - cO % , . . - -C- '0- V CO L N r- O N- C - r- r- r - - mO en N N N ' - r- eN N - r- rNr~N ~ 0 n I.- r:~ N r.~ - O N - O O Oen-C r- Mne c cI t ",. ' e- r CO rn " O -O c" -,'" <2' O N> 'n C O ' L ' | D c" rN , c: c> c OD c N D I c>-I - e r" D .2 - - N '0 ro en- N- - cD- N - N D -> N D- c-> - r- L- D. O - '0c0 o o m0m C- - 'n en - -c- en Io e r CO Z- '0O CO m 0 o o'0 CO I e - - - O O - NN N~ e- r- O CO en- C- CO N~ O - O O N- O N'-0- - O e - . O - CO I CO M M Oe C en M L <| . - N ý L . O . CO O N c i . N . - In L . . - . N c N O . '0 . '0 O . M - O . - . '- en"' ,o O O - rN - enN .- o r-o 4 m - r- - e- r- N c O' '- N r- L- r-, '- m--N I e- n T CO I' - ~ - CO 0 e s-C ' m0 mT-. L Oa- '0 ' CO N O O enN Ni 0- N-t NN N r - COO e-0 ' - - - N NO e - - r.- N - I .- r- O I lO..- O O 0LaI4-i cq Lr s- Cý ri 1^- ry 04m 5 I•© -C n.2 e - Rd '3 1 ' © Lei r, ©O N l- L- c,- D O N r-, . . .- .- .- .OM C O s" r- 4 L e- e O N C-t-N e- r-4 r- L- I D c-D- c- N ý l O Nn ,J c,ý si - C odL i - r - O r- - - ýo r- rl e r-4 N N - n - C- ) 0 æ m a N O A L n M !"s"O r- o- m o,- OD OR r - O O n O ' - c) Oý r- '4 01 Os r-4 - ~ ~ c O - O e- m - r- r n NR 1 lo oo ý- o-R - - N s r-, LI", - e - t r - A - e I O N ý 4 o 0 ý m ~ c > -,t r- E In - - A - - r- r, C) r- .0 00 D Dr, L q cý m r4ý - m ýýL - * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * m O 00 , N O > m N m La <-, 14- Om - I, <- CO Mt N .- CO*N * r, r, I r-I < - D- , - f N O O tN Lt- c DN O - O 4- , 5N cDL c gN COnN Im >O rm r-- - p g L- c> Cý CO rl c> oc o ~ <7 m ' c7 r__ 9 - . Cý rý .tý 9 CO -ý Ln 1 o r-.ý c:> C O c9 c >C r- m o4*o o o o o o o o o o oo o c, oD c> C) C) c> C> cD C> oC)o * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * a CO - NO O I C O © f M t O N O r -N r- NcC c c o c>- F c 5 m " i M O ' CO O CO ä> - O CoO O CO L f - D D ä D- O m o > ) : o o o o c> o oo o o O o o o o o oo o o o o o o D C) * * * * * * * * * * * * ** * * * * * * * * * * * * * * * ** * * * * * * * * * * * * * * * * * * CD M D'" C c 4 O e- N 4 c > CO Ch N C>' rI tN rI m C lo O ' '2 CO CO o- I D O o C O C - - O N M O N m O N - L m t M CO rlf t q '" CO N 8 0 \O CO O CO 'f O g o z o Cn c> o : o n o o n o o r- o o g oo o cm ö, c o, oo m o -g r- o2 o o a p o o lo C> * **** * * c * c>* * * * * * * * * * * * * * * * * * * * t - M L-> N O ' r- CO cp N 5 Omr CO e- N c oo I5 - CO e- a, r r m 'O r - flD O ID , O I 5:' O t r-I 3 c- ko N 5:> Lt c CO 5 C2 8 C5 N CO - g g r N 5 N O F * c g r - ** * * * * * * * ** * * * * * * * * * * * * * * * * * * * * * * * * * * * * O O O I Ol O e- si'I i N r- CO Lrt CO tt Mt f fl' 'f O O 'f e- C' r- tvsfl O ItOt" O O - N r- O CO N O e- O 'f OlN O O i ©f O O 'f CO At M> r- si CO O O N O . 9 .- .t 9 .- 9 . .N - - -ý - tvi .f . 19 r-i N 9t 9v 'fl 9 tt r lo Ln r r, N -c L - rN r, - O CO t M n - fl> O CO C 4 O -- CO flN t n M 00 CD CO CD O I "- r- Afl CO Lt |- O Lt tt M Lt O O CO CO -""" N fl ' " I r """ - "" - O CO Mt -"""" - r""" I \ O Mt N O cD cD r- o ý o o on oo _ cD on m on g o o o r, o -c o o o o m o o oon C. " CO L CO .- fl CO C t CO N - D CO (D t( CO c> CO i" N CO cD, COD N D N CO cD N L> Lt- CO C> > DN " CN O CO C- N O LD -Ct 00 ~ ' V)~ 00 ID M i o N T C I C O M C> OD M ID C3N - D r- e-C O I - - Nm t Lt-n rN m 0o L - r- l N N m N N 0- 1 -2" - N N r'- r N - m j- j- - CO If O,N- - . - - O - 2: fl N O CO r NS og m Q 9 O N CO 'fl t " - O f' M- La Of N T Nf O Ofl t c- r- m c> r- O O O O C O r- - r- D oD c> - - . D O - C) r- tn -c" Ln o mO g -" Ln ON Co Co N- O 9 0 O 9 Ln -O " C --1 Ln -M C- N VO M C- D M N- C"3i O N CR O Ln M C> Ln cD Q N N r- O N' s InN Nt m- r N N- '- r- e- r-7 c- NC - N ' 1- r-l - - - r-- '-l - ID - - Q, CD OI M M CON r- r- m M co L O N L ct N o N c cO CO r- Lt ' CO CO N O N C> N r N- --- N -4 fl r- - O E ,- - cO r- c r- Ln fl t cy m CO Lc), N M " o IN O L -O\O - e -i al' r- D O m N 1 N o m c ý o rm N , c, g pN r- r O m L - - - O N - Mt IDlN - n- o - M CO - - - - N O C> M- - N - - - - I - - D Iý Ln m m 0i" O I Ln mO N1 r- LcD - i CO fl' ; ,-' " L r a 0 © 'n 1" °g c °oq à ? Ln 1" 9 11 6 M - - : N- N- N- c; N, e:i N M - t e a L r- ,- 5- -i L- ý- r- L- L" -- -i - 5- r-4 r - Ln i- - c> ýD n N O14 CD L D OC O) L L oo tN . CO fl t 'O - - o O Lt fl E E- E E " r_ c > fl et a _ # < # o o o .. > :z: a a o o oflo # # fl c,. 4fl . 4f ~~~~c a, ajt t Nt 0 v0 tv~~c2 g u; =f. . t fl ~ l~~ c2.. s> E > a, E _t enE. - - - ~ < Wl 72 ZA A E E ä 1= 2 2 92 c> 0ý c -ý J>r ) C)C) D C 01 c, ä <2 cD D e Non- Exporters Non- Exporters Exporters Exporters Exporters Exporters (74) vs Exporters Services Manuf. Services vs (286) ttest(p) (30) (44) Manuf. Size of a company (1-3, 3=large) 1.86 2.49 0.000*** 2.33 2.59 0.114 % foreign owned 8.46 30.06 0.000*** 24.79 33.60 0.409 Manager experience (years) 21.11 19.47 0.274 20.07 19.05 0.714 Quality certificates 0.15 0.32 0.000*** 0.27 0.36 0.389 Years of operations 20.49 19.08 0.448 15.67 21.47 0.138 Days to receive an electrical conn. 34.43 40.77 0.830 11.67 65.71 0.039** Power outages/month in last year 28.96 16.40 0.110 16.90 16.04 0.882 Days to obtain a water connection 21.57 35.86 0.281 12.33 53.50 0.191 Days to get a telephone connection 10.10 8.05 0.492 5.75 9.46 0.329 Obs: electricity 2.34 2.36 0.886 2.30 2.41 0.750 Obs: telecommunications 1.35 1.34 0.961 1.53 1.20 0.283 % of national sales to total sales 100.00 49.31 0.000*** 56.54 44.70 0.226 % of indirect exports to sales 0.00 17.60 0.000*** 34.61 6.77 0.000*** % of direct exports to sales 0.00 33.10 0.000*** 8.86 48.52 0.000*** Avg days to clear customs in 2007 . 60.03 - 5.71 72.29 0.629 % of inputs of foreign origin 46.19 57.65 0.129 . 57.65 Obs: transportation 1.37 1.64 0.089* 1.73 1.58 0.616 Obs: customs and trade regulations? 1.01 1.49 0.003*** 1.32 1.59 0.365 Obs: the informal sector competitors 2.02 1.65 0.039** 1.67 1.64 0.929 Obs: access to land 0.94 0.75 0.197 0.83 0.70 0.591 Obs: crime, theft and disorder 1.92 1.64 0.089* 1.70 1.59 0.709 The court system is fair 1.85 2.01 0.171 1.97 2.05 0.735 % mang. time to deal with 11.62 11.63 0.992 14.22 10.05 0.069* regulations? % annual sales as informal payments 0.37 0.25 0.706 0.57 0.00 0.082* Number of inspections 6.28 2.93 0.514 3.21 2.64 0.628 Days to obtain an import license 66.26 22.91 0.169 23.00 22.86 0.993 Days to obtain an operating license 93.03 38.64 0.162 48.40 30.50 0.268 Obs :tax rates 2.37 1.93 0.006** 2.52 1.55 0.002*** Obs: tax administrations 1.99 1.59 0.011** 2.10 1.25 0.004*** Obs: business licensing and permits 1.22 1.14 0.600 1.17 1.11 0.822 Obs: political instability 1.87 1.54 0.055* 1.47 1.59 0.693 Obs: corruption 2.51 2.15 0.038** 2.30 2.05 0.462 OBs: access to finance 1.54 1.42 0.468 1.30 1.50 0.499 Obs: labour regulations 1.54 1.63 0.578 1.60 1.65 0.870 Obs: inadequately educated wokforce 1.97 2.10 0.447 2.23 2.00 0.395 % skilled workers 0.59 0.39 0.001*** . 0.39 - Source: World Bank staff calculations based on the World Bank Enterprise Survey. C) 00 l- ~ C) 10 1- 00 C) UIN 00 r14 Lnl 1* z, - O CC Ln j c n 00 0 0 C). C) 4 C) 0 N r,. l- 00 C>CD %1 M i 666 66 viC ýC )c 4 n kc 00 00 <0 r' C> al *- *) *l *, *' r1 n l 3 c CO rN Ce r C e - O - CD C) C D c; c0c C) 0 0 0 0 C) O - O C CD 0N z D 0 0 L( 0 0 0 n ~8 -N r e r- rod 0 - o r- c - ,o i r . - 83 N N r] 0 r0 M - : od m 6 - 0i- 0 C) L- Ln al rn , o ao o al r,]. o' .- - '. Nc i.i E t~< à * 0 Ln mr- o a un e. r--. - o .- r4 r--. eN - C> 4 4- 00 O Ln ,. r-.. e. Ci r, r., C r14 r, 0 ,1 i r c,L c> -0 ti , 'J - _' o ev 0 * S -Ln r ti4 C> U- O -- - o00 od 6> 0 0' 04- i 00 -4 ko rn r0.i 0j Ln- ' .0C 00 0n 0 C> S × c f,; i r 0> C> r' 0 0> 0