TF094947 / Services, FDI & Endogenous Productivity Effects
in the European Neighborhood Policy: the Case of Armenia
KCP Window / Investment Climate & Trade and Integration
Country / Armenia / TTL Name / Ana Margarida Fernandes
Project Period / 05/16/2009 –06/30/2011 / Approving Manager / Aaditya Mattoo
Grant Amount / 90,000.00 / Disbursement / 89,825.68
Research Team (with affiliations and country of origin for non-Bank members)
Ana Margarida Fernandes (TTL); Pedro Rodriguez; David G. Tarr (USA); Jesper Jensen (Denmark); Edward J. Balistreri (USA); James Markusen (USA); Grigol Modebadze (Georgia); Karine Eroyants (Armenia).
Research Objectives
- Summary of Aims, Objectives and Significant Achievements
Objective: Given the inclusion of services in modern FTA agreements negotiated with the EU, the US and in some other agreements, it is important for the economics profession to be able to assess the impact of preferential services commitments. It is also important for the Bank to be able to provide advice to governments on these issues. Given client demand, it is unfortunate that the profession does not have a framework for assessing the ex ante welfare impact of including liberalization of barriers against foreign direct investment in business services in regional agreements. This project would fill that void.
A second objective is to improve our knowledge of services policies and their effects in Georgia and especially Armenia. These would be translated into ad valorem equivalents of the barriers against services providers in various sectors, estimates that are required by the model.
Significant achievements including innovations:The principal achievement of the project is that it succeeded in developing an innovative model to assess regional commitments to multinational investors in services. This model was applied in Armenia and a paper with both modeling and data details, as well as a policy based introduction, was discussed with the Government of Armenia. In addition, at the request of the Ministry of Economy of Armenia, the modeling framework was extended to include the impact of trade facilitation and standards harmonization, and the relative importance of these aspects of a Deep and Comprehensive Free Trade Agreement (DCFTA) with the European Union were assessed. The regulatory barriers in services were assessed and estimated in both Armenia and Georgia and discussed with the government of Armenia. The project team developed an input-output table for Armenia based on preliminary survey data from the Armenian National Statistical Office. This was much appreciated by the National Statistical Office. Training in how to develop and interpret the model was provided under co-financing by the World Bank Institute.
- Nominated Outputs
Jensen, Jesper and David G. Tarr (2012), “Deep trade policy options for Armenia : the importance of trade facilitation, services and standards liberalization,”Economics: the open-access, open assessment e-journal, Journal article No. 2012-1. Published in the special issue “Trade Facilitation, Transport Costs and Logistics: A New Challenge for European Competitiveness.” Available at:
A longer version of the above paper, with all appendices, appeared as:
Jensen, Jesper and David G. Tarr (2011), “Deep trade policy options for Armenia : the importance of services, trade facilitation and standards liberalization,”World Bank Policy and Research Working Paper No. 5662, May. Available at:
Modebadze, Grigol and Karine Eroyants (2010), “Ad Valorem Equivalents to FDI Restrictiveness in Armenia.”File attached.
Modebadze, Grigol (2010), “ Ad Valorem Equivalents to FDI Restrictiveness in Georgia.” File attached.
Jensen, Jesper and David G. Tarr (2011), “Development and Documentation of an input-output table for Armenia,” Appendix G to Jensen and Tarr (2011) cited above.
Eroyants, Karine (2011), “Market Shares by Region in Key Services Sectors in Armenia,” Appendix B in Jensen and Tarr (2011) cited above, pages 74-93..
[w1]
- Major Difficulties[describe any expected or unexpected issues and how the research team dealt with them, and the end result.]
The project was originally proposed for Georgia. Upon more detailed discussions while on mission in Georgia, however, the Government of Georgia indicated that the project was a rather low priority project. On the other hand, the Government of Armenia expressed a strong demand for this project and Armenia and Georgia are in identical negotiating stages regarding a deep and comprehensive free trade agreement with the European Union. That is, the project was strongly demanded by the Government of Armenia and received attention from the powerful Ministry of Economy in Armenia, where the chief trade negotiator is located. The World Bank's Lead Economist for Georgia, Armenia and Azerbaijan recommended that it would be better to implement the project in Armenia. The KCP II project team believed that KCP resources would be much more effectively employed if the project were implemented in Armenia, and the switch of countries was approved by the KCP II Trust Fund manager in the World Bank. The project was successfully implemented in Armenia.
- Surprises[describe any outcomes of the research, beneficial or otherwise, that were unexpected at the outset. Further details should be included in the full report.]
The gains from trade facilitation were found to be considerably larger than the gains from services liberalization or harmonization of standards in the case of Armenia. This allowed the project team to inform the government where the resources would be most effectively spent.
The project team also discovered that the preliminary survey work by the National Statistical Office on input-output industrial relationships was very seriously flawed. The National Statistical Office was very grateful for this information and asked for wider dissemination of this fact in the hope that a full survey could be financed.
- Full Completion Report
- Background and motivation
As of July 2007, 380 preferential trade agreements (PTAs) had been notified to the GATT/WTO, and the number continues to grow. With their proliferation, a growing area of Bank business in trade policy is responding to client country requests on how to best structure the existing or potential PTAs for growth and poverty reduction. Given the inclusion of services in modern FTA agreements negotiated with the EU, the US and in some other agreements, the Bank needs to be able to assess the impact of services commitments as part of its advice to governments. Services commitments in regional agreements could lead to substantial productivity improvements, but possible substantial trade diversion costs as well if high cost services providers obtain preferential market access. Crucially preferential versus global liberalization should be assessed as well as a comparison of preferential liberalization with developed versus developing economies. Given client demand, it is unfortunate that the profession does not have a framework for assessing the ex ante welfare impact of including liberalization of barriers against foreign direct investment in business services in regional agreements. This project intended to fill that void and to provide better information on services sector policies in the selected countries.
- Original objectives of the research, along with any changes and their justification; note their consistency with Bank, and KCP objectives
The key objective of the project was to improve service sector policies in developing countries by developing the analytic capacity to assess service sector policies in free trade agreements.
A second objective is to improve our knowledge of services policies and their effects in Georgia. Available data sources and Bank commissioned surveys of the regulatory regimes in key business services sectors were used to assess the services policy environment. These were to be translated into ad valorem equivalents of the barriers against services providers in various sectors, estimates that are required by the model.
As discussed above, the project was implemented in Armenia, rather than Georgia as originally proposed. This was done to more effectively use KCP II resources where they would be most valued and effective.
- Assessment of the extent to which the objectives have been met, with explanations of any shortfall.
The objectives of the research have been met. A new innovative model of services sector reform in a regional context was developed along the lines of the proposal. This was used as a basis of policy discussions with the Government, a policy seminar in the World Bank with the country and regional team and seminars in Armenia. In addition, the model was extended to include trade facilitation and standards harmonization at the request of the Government.
Services sector policies were evaluated and the ad valorem equivalents were estimated in both Georgia and Armenia. In addition, the ownership shares of different regions in Armenia were assessed and quantified. Finally, the project team constructed an input-output model of Armenia that was much appreciated by the National Statistical Office of Armenia.
- Methodology and data
The core objective of this project was to construct a computable general equilibrium (CGE) model for Armenia (an IDA blend country) that is capable of assessing the impacts of modern preferential trading arrangements that include services and FDI chapters. The analytic approach extended work on services and FDI in a small open economy framework that the lead researchers have recently published in the Journal of International Economics, Canadian Journal of Economics, Review of Development Economics, Eastern European Economics and the Review of International Economics (forthcoming), to allow for multiple foreign trading partners with foreign direct investment from each. A model of Armenia was developed where each foreign trading partner would endogenously determine their level of FDI in each of the key business services sectors. The Dixit-Stiglitz variety mechanism has been an essential element of the new growth theory, the new economic geography models, imperfect competition models in international trade going back to Krugman and Ethier, and even the new heterogeneous trade models of Melitz and others. The model in the project contains the Dixit-Stiglitz variety mechanism and thereby contains endogenous productivity effects from additional varieties of business services and imperfectly produced goods. In accordance with empirical evidence, we developed a model in which productivity impacts of FDI would vary with the stage of development of the foreign partner.
It was necessary to estimate the ad valorem equivalents of barriers to foreign direct investment in key business services sectors. This allowed the team to follow the methodology in the earlier work of the key researchers on Russia (developed by several Australian authors). To do so, it was necessary to use survey work on the regulatory regimes of Armenia in business services and base the estimates of the ad valorem equivalents on the scoring of these surveys. Further it was necessary to collect data on shares of ownership by region in the key business services sectors.
It order to construct the model it was necessary to construct an input-output table for Armenia, since none were available from the National Statistical Office of Armenia. The project team used preliminary survey data on the input-output relationships of Armenia provided by the National Statistical Office. Since the team found that the data had rather serious deficiencies, the team supplemented the survey data with data from the GTAP data base on the range of reasonable input-output coefficients. The construction of the Armenian input-output table was used both sources of data.
- Results and outcome: What have we learnt so far?
An Armenian Deep and Comprehensive Free Trade Agreement with the European Union should produce substantial gains from the deep aspects of the agreement: trade facilitation, services liberalization and standards harmonization; and improvement of the investment climate in the long run.Trade facilitation should provide the greatest benefits, followed by services liberalization. Tariff preferences to the European Union, are not likely to produce benefits due to trade diversion. But not all aspects of a deep agreement with the EU would be beneficial. Experience has shown that adopting all EU sanitary and phyto-sanitary (SPS) standards for domestic production will produce substantial costs that likely exceed the benefits in many of the cases; thus, the report to the Government recommended a case by case approach to adopting SPS standards of the EU. In services, multilateral (rather than preferential) services liberalization should contribute substantially to Armenian real income. Liberal rules of origin on services commitments would be a useful step toward non-discriminatory liberalization. CIS services agreements are not likely to be very helpful.
- Dissemination details (including future plans)
A long version of the paper was produced for the World Bank working paper series that discussed the results and provided documentation of the model and the data set. A shortened version of the paper was submitted to Economics, a professional open access e-journal. It is currently under review, but has been published on their website as a discussion paper while under review.
The results were discussed with the Government while on mission in Yerevan in December 2010. In addition, seminars were held at a think tank in Yerevan along with discussions aimed a capacity building of the think tank members. A seminar was held in the World Bank with the regional team in March 2011.
With the support of the World Bank Institute, a one-week training seminar on the use of this model and related models was held in Moscow in December 2010. Twenty-two well qualified participants were selected. Eighteen were from the CIS, including three from Armenia nominated by the Armenian Ministry of Economy
The Armenian model results will be presented at one of the sessions of the World Bank Institute course entitled “Services Trade Liberalization” to be held at the Joint Vienna Institute during October 31-November 4, 2011. A member of the project team will make the presentation. We expect that this research would be presented at future World Bank Institute seminars on regionalism or services.
Impact—within and outside the Bank
Inside the Bank, the seminar in the World Bank in March 2011 was led by the PREM Director for ECA who focused on how to operationalize and use the results in the policy dialogue with the Government. A member of the project team served as a peer reviewer on related work on Armenia and Georgia (on improving the National Quality Infrastructure) and contributed to that project based on the knowledge learned in the KCP II financed research.
Outside the Bank, the discussions with the Government of Armenia were very fruitful and were generally agreed. Regarding services, it was agreed that there was more to gain from a preferential arrangement with the EU than with the CIS, but that multilateral services liberalization was the best option. The Government also recognized the importance of developing the National Quality Infrastructure, but agreed with the assessment of taking a case by case approach to adopting SPS standards of the EU as part of a Deep and Comprehensive Free Trade Agreement. The National Statistical Office used the results of the research to appeal for funding of a full survey to improve the quality of the input-output coefficients.
The research ideas are new ones and are under review at a professional journal. As of October 25, 2011, the paper has been frequently downloaded. The World Bank working paper has been downloaded from the SSRN website (49 times as of October 25, 2011). In addition, it has been downloaded from the World Bank research website. The on-line journal Economics posted the paper as a discussion paper while it is under review and reported that the discussion paper version has been downloaded 100 times as of October 9, 2011.
Implications for future research
The research has shown the importance of multilateral rather than preferential liberalization of services, and the greater value in regional commitments with technologically advanced countries. It has also demonstrated that it is important to consider the impact of trade facilitation and other mechanisms that lower trade costs, since in countries with low barriers in services, trade facilitation could be the more important aspect of regional trade agreements. It suggests that the model should be extended to a multi-regional trade model that can endogenously capture the terms of trade gains that could be expected in a preferential trade agreement.
- Capacity building (Have you used local consultants or firms? Have you provided training to client country counterparts?Number of developing country researchers and institutions engaged with KCP research, breakdowns of resources used for capacity building out of total disbursements)
The project team used a Georgian think tank (involving one Georgian) and one Armenian local consultant for the work on the services sector policies, estimation of the ad valorem equivalents and the estimation of the share of the services sectors owned by multinationals in the various regions of the model. Training was provided in two parts. First, about $80,000 of WBI resources were used to support a one-week training seminar on the use of this model and related models in Moscow in December 2010. Twenty-two well qualified participants were selected. Eighteen were from the CIS, including three from Armenia nominated by the Armenian Ministry of Economy. In addition, seminars were held at a think tank in Yerevan along with discussions aimed a capacity building of the think tank members. A seminar was held in the World Bank with the regional team in March 2011. The Armenian model results will be presented at one of the sessions of the World Bank Institute course entitled “Services Trade Liberalization” to be held at the Joint Vienna Institute during October 31-November 4, 2011. A member of the project team will make the presentation. Although only about $10,000 of KCP II resources were used for capacity building, KCP II resources were leveraged with WBI resources to provide substantial resources for capacity building in this project.