Document of The World Bank FOR OFFICIAL USE ONLY ON A LOAN TO THE ARGENTINE REPUBLIC ENTERPRISE EXPORT DEVELOPMENT PROJECT.

Transcription

1 Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Document of The World Bank FOR OFFICIAL USE ONLY IMPLEMENTATION COMPLETION REPORT (39600; 3960A; 3960S) ON A LOAN IN THE AMOUNT OF US$ 38.5 MILLION TO THE ARGENTINE REPUBLIC FOR AN ENTERPRISE EXPORT DEVELOPMENT PROJECT June 28, 2000 Finance, Private Sector & Infrastructure Department Country Management Unit 7 Latin America and the Caribbean Region Report No: This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.

2 CURRENCY EQUIVALENTS (Exchange Rate Effective April 2000) Currency Unit = Argentinean Peso 1 Peso = US$ 1.00 US$ 1.00 = I Peso FISCAL YEAR January 1 December 31 ABBREVIATIONS AND ACRONYMS AGN AFIP ANA AOP DGI GDP IASCAV ICR INAL INTI IRAM ISO MERCOSUR MSA MSC MSTQ PREX QAG SAR SIGEN SMEs SENASA SEPYME UAP UCPFE UKAS VAT WTO Government's General Auditors Federal Administration of Public Income National Customs Administration Annual Operating Plan General Tax Directorate Gross Domestic Product Argentine Sanitary and Phytosanitary Institute Implementation Completion Report National Food Institute National Institute of Industrial Technology Argentine Institute of Materials Rationalization International Standards Organization Common Market of the Southern Cone Management Services Agreement Management Service Contract Metrology, Standards, Testing and Quality Export Development Program Quality Assurance Group Staff Appraisal Report National Internal Auditors Small and Medium Enterprises National Service of Animal Health and Agribusiness Quality Secretariat for Small and Medium Enterprises Project Administration Unit Coordinating Unit for Export Promotion Policies United Kingdom Accreditation Services Value Added Tax World Trade Organization Vice President: Country Manager/Director: Sector Manager/Director: Task Team Leader/Task Manager: David De Ferranti Myma Alexander Danny M. Leipziger Mariluz Cortes

3 FOR OFFICIAL USE ONLY CONTENTS Page No. 1. Project Data 1 2. Principal Performance Ratings 1 3. Assessment of Development Objective and Design, and of Quality at Entry 2 4. Achievement of Objective and Outputs 5 5. Major Factors Affecting Implementation and Outcome Sustainability Bank and Borrower Performance Lessons Learned Partner Comments Additional Information 21 Annex 1. Key Performance Indicators/Log Frame Matrix 22 Annex 2. Project Costs and Financing 23 Annex 3. Economic Costs and Benefits 26 Annex 4. Bank Inputs 27 Annex 5. Ratings for Achievement of Objectives/Outputs of Components 28 Annex 6. Ratings of Bank and Borrower Performance 29 Annex 7. List of Supporting Documents 30 Annex 8. Beneficiary Survey Results 31 Annex 9. Stakeholder Workshop Results 34 Annex 10. Borrower's Final Evaluation Summary 36 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not be otherwise disclosed without World Bank authorization.

4

5 Project ID: P Project Name: ENT.EXPORT DV. Team Leader: Mariluz Cortes TL Unit: LCSFR ICR Type: Intensive Learning Model (ILM) of ICR Report Date: June 28, Project Data Name: ENT.EXPORT DV. CountryIDepartment: ARGENTINA Sector/subsector: MT - Trade Policy Reform L/C/TFNumber: 39600; 3960A; 3960s Region: Latin America and Caribbean Region KEY DATES Original Revised/Actual PCD: 11/11/94 Effective: 06/25/96 07/18/96 Appraisal: 05/15/95 MvTR: 06/25/97 11/18/97 Approval: 11/28/95 Closing: 07/31/98 09/30/99 Borrower/lImplementing Agency: REPUBLIC OF ARGENTINA/MINISTRY OF ECONOMY/SECRETARIAT FOR SMALL AND MEDIUM ENTERPRISES (SEPYME) Other Partners: STAFF Current At Appraisal Vice President: David De Ferrranti Shahid Javed Burki Country Manager: Myma Alexander Gobind T. Nankani Sector Manager: Danny M. Leipziger Paul Meo Team Leader at ICR: Mariluz Cortes Mariluz Cortes ICR Primary Author: Mariluz Cortes; Sonia Plaza 2. Principal Performance Ratings (HS=Highly Satisfactory, S=Satisfarctoy, U=Unsatisfactory, HL=Highly Likely, L=Likely, UN=Unlikely, HUN=Highly Unlikely, HU=Highly Unsatisfactory, H=High, SU=Substantial, M=Modest, N=Negligible) Outcome: S Sustainability: L Institutional Development Impact: M Bank Performance: S Borrower Performance: S QAG (if available) ICR Quality at Entry: Project at Risk at Any Time: Yes Sustainabilitv: It will be likely,for the INTI component, Fundacion Exportar andfor the firms that participated in thie PREX program.

6 3. Assessment of Development Obkjective and Design, and of Quality at Entry 3.1 Original Objective: 1. The primary objective of the project was to help accelerate the export response of Argentine enterprises in the contexk of increasing trade liberalization and stabilization. The project was to achieve this objective by: (i) assisting small and medium scale enterprises (SMEs) in Argentina to become internationally competitive; and (ii) helping improve the perfornance of export-related public agencies. To help improve the competitiveness of SMEs, the project would encourage a limited number of companies to invest their own resources in productivity improvements and export developme:at expenditures, through the incentive of a one-time grant to share these costs. The project would help improve the performance of export-related public agencies through simplified export procedures, greater efficiency of quality certification agencies, and more effective export promotion programs. The project was designed as a pilot project to test the effectiveness of this type of interventions and of the project design. A larger project would be contemplated if the assessment of the pilot project resulted positive. 2. The objective was clear and realistic for the country, and in accordance with the Bank's country assistance strategy for Argentina, discussed by the Board on May 4, This strategy aimed at assisting in the consolidation of the comprehensive reforms being implemented by the Government -- liberalization of the ecdnomy, deregulation, and privatization of state enterprises -- by helping to accelerate the export response of the private sector, which was viewed as necessary to maintain employment levels and reduce public opposition to the reforms. The objective was also in line with the development objectives of the Government, which viewed exports as the engine of growth and the private sector at the center of that effort. 3.2 Revised Objective: 3. The project objectives remained unchanged. 3.3 Original Components: 4. A. Enterprise Assistance Component (US$32.7 million): This component included: (i) a cost-sharing grant scheme to cover up to 50% of the costs of export-related consulting services and information required to improve the international competitiveness of SMEs; and (ii) the creation of a D)irectory of Consultant; and Market Information Suppliers, to ensure easy access by Argentine exporters to infornation about the availability of export-related services worldwide. The cost-sharing grants financed e sport-related consulting services, acquisition of market information, trips to fairs, mailing of samples, design and mailing of promotion materials, quality certification, and other services identified in an export plan approved by the program. The grants were for up to US$75,000 per export plan and up to a maximum of US$125,000 per enterprise. 5. This component was well designed. The cost-sharing grant scheme was designed based on the findings of a sample of 339 SN[Es and interviews with 20 export focus groups carried out by the Borrower as part of project preparation. The interviews revealed that SMEs had limited ability to compete internationally because: (i) exporting activities tended to be relatively new ones and secondary for them; (ii) few SNMEs made organizational changes before entering foreign - 2 -

7 markets; (iii) SMEs overestimated their managerial capabilities relative to exports; and (iv) few exporters sought professional advice to improve their export competitiveness. The basic assumption in the design of the program was that access to information about export markets, technology and business practices would help improve the ability of SMEs to make wiser decisions about export activities and improve their export performance in the long run. The expectation was that by helping a relatively large group of SMEs to learn how to export, the program would generate a demonstration effect that would induce other enterprises to seek external consulting services to improve their export performance. 6. Although overall project coordination was exercised by the Borrower's Project Administration Unit (UAP), the cost-sharing grant scheme was to be managed by a private management service contractor (MSC). This was the first time that a publicly financed program in Argentina was managed by a private contractor, selected through international competitive bidding. The MSC was responsible for determining if the interested SME had the potential (technical and financial) to become an exporter; assisting the interested SME in preparing an export plan (that is to identify the consulting services that they required to enter in the export business); and providing the SME with access to information about potential export related service suppliers. The contract required the MSC to open offices in Buenos Aires, Cordoba, Rosario and Mendoza. The UAP's role was to give its no-objection to the cost-sharing grants approved by the MSC, and to audit and assess the performance of the MSC through periodic surveys carried out by third parties. 7. The use of a private MSC to administer the scheme was an innovative feature in the design of the project. This administrative arrangement was considered essential in order to: (i) ensure independence in the approval of the grants; and (ii) provide technical assistance to the SMEs to prepare their export plans. The MSC was also in charge of reimbursing the assisted firms of half of the cost of the consulting services, once these services had been rendered. For this, the MSC had access to a revolving fund that was periodically replenished by the Borrower. This feature was considered important to reassure the assisted firms that they would be reimbursed without bureaucratic problems. 8. The management contract signed with the MSC was a Performance Fee Contract, that differed from the traditional approach in two major ways: (i) the focus was on achieving results or "outputs", as opposed to the traditional approach which is "input" oriented; and (ii) the emphasis was on performance-linked incentives to reward the efforts to deliver higher levels of performance, as opposed to the traditional approach of relaying solely on penalties for providing motivation to achieve the minimum levels of performance. The contractor was paid based on a combination of fixed-and performance-based payments. Payments were divided as follows: (i) 60% as a fixed amount paid monthly; (ii) 33% as an incentive payment paid quarterly based on the fulfillment of quantitative and qualitative targets by the contractor; and (iii) 7% as a final payment at the end of the contract, based on having in effect disbursed 98% of the approved grants. The incentive payment was designed based on a polynomial formula that tried to reach a balance between quantitative indicators, such as the number of export plans approved in each quarter, or the percentage of grants approved that met the expected disbursement schedule; and qualitative indicators such as whether the SME was a first time exporter, if it was located outside Buenos - 3 -

8 Aires, and if it intended to export outside neighboring countries.. 9. The incentive scheme in the management contract was a powerful tool to achieve the program's targets. However, it was not always easy to interpret and apply. Some of the shortcomings in the design of the incentive scheme can be summarized as follows: * The indicators used to measure the contractor's perfonnance may have not been the most appropriate, because of the difficulties to specify production of outputs in advance (e.g. number and value of approvals and disbursements). Also, given the demand-driven nature of the program, it was difficult to specify in advance the implementation timetable of the activities specified in the export plan, because the timing in the use of the enterprises' resources depended on their perception of opportunities and risks. For that reason, compliance with the timetables specified for the implementation of the export plans tended to be low. * In practice, the incentive scheme may have put excessive pressure on the private operator to meet the contract's quantitative targets, some times in detriment of the quality of the export plans. * The management contract did not take into account unforeseeable events that could limit the disbursement of the approved grants and have an impact on the final incentive payment. Eventually, the program disbursed only 67% of the approved grants (disbursement of US$17.7 million versus grants approved for US$26.6 million). The Borrower and the MSC had very different opinions about the final payment. The Borrower argued that the MSC was not entitled to the final payment because it had not reached the 98% disbursement target, and the MSC argued that lack of compliance was due to factors outside its control and, therefore, it was entitled to the final payment. Both parties agreed to bring the issue to an arbitration panel. At the time of the ICR preparation, the issue had not been solved. 10. B. Institutional Strengthening Component (US$5.8 million): This component had four sub-components: (i) simplification of export procedures; (ii) strengtiening of quality control systems; (iii) strengthening of export promotion policy formulation, and rationalization of export promotion programs; and (iv) strengthening of commercial promotion services. The different sub-components were to be implemernted by the relevant public agency, under the coordination of the UAP. For this, the UAP would sign Letters of Agreement with each beneficiary agency. The UAP had to prepare an Annual Operating Plan (AOP) for each sub-component of the Institutional Strengthening Component in coordination with the respective beneficiary agency. This component was also well designed based on the findings of a review, carried out by the Borrower, of the constraints faced by Argentine exporters due to the inefficiencies of the export-related public agencies. This review revealed that cumbersome and time consuming export and certification procedures, and the difficulties to obtain reimbursement of indirect taxes were major constraint to small exporters. 3.4 Revised Components: 11. The design of the project remained unchanged until project completion. However, at the suggestion of the UAP, there were important modifications and reallocation of funds between components. One modification was the decision not to hire a contractor to compile and operate - 4 -

9 the Directory of Consultants, because the UAP had compiled an extensive Directory using other sources of funds. The unused funds were later reallocated to another sub-component. Another modification was the decision not to go ahead with the Simplification of Export Procedures program, under the argument that the objectives of that program had been largely achieved by the merger of the Customs and Tax Authorities. 3.5 Quality at Entry: 12. The project pre-dates the introduction of the QAG process at the Bank. 4. Achievement of Objective and Outputs 4.1 Outcome/achievement of objective: 13. The objective of helping to increase the competitiveness of a group of SMEs through a cost-sharing grant scheme was substantially achieved. The effectiveness of this scheme was evaluated through two impact assessment surveys carried out during project implementation. Since the surveys were carried out before the assisted firms had had enough time to imnplement the results of the consulting services obtained under the program, they focused mainly on changes in behavior of the assisted enterprises in areas likely to improve export performance in the future. The second impact assessment survey shows that many of the enterprises that participated in the cost-sharing grant scheme had adopted measures that could help them to be more competitive intemationally, such as quality control programs, training programs, introduction of new or improved products and channels of distribution, and cost reducing measures. Compared to a control group of similar enterprises not assisted by the program, the assisted enterprises consistently showed that they were more likely to adopt export-oriented measures and had better export performance. The results of the second impact assessment were encouraging, despite the fact that it was conducted during a general contraction of Argentine exports, particularly to Brazil, the main export rnarket for the assisted SMEs. A more conclusive assessment of the impact of the program on the export performance of the assisted enterprises should be conducted in two or three years after the end of the program, when it is more likely that the enterprises will have had the opportunity to put into practice the advice of the consultants. 14. The objective of improving the performance of export-related public agencies was only modestly achieved. Only two of the four Institutional Strengthening sub-components were implemented: (i) the strengthening of the commercial services provided by Fundacion ExportAr, and (ii) the strengthening of the institutional infrastructure for quality control. Of these two sub-components, only the second, which involved the accreditation of several laboratories of the National Institute of Industrial Technology (INTI) for quality control certification, can be expected to play an important future role in improving the export performance of Argentina's SMEs. None of the other three sub-components were implemented. 4.2 Outputs by components: 4.2 Outputs by components: -5 -

10 A. Enterprise Assistance Component 15. In terns of outputs, the Enterprise Assistance Component should be rated as satisfactory. This component accounted for 70% of the loan funds. 16. The Cost-sharing Grant Scheme (PREX Program): This component was originally designed to assist 900 enterprises in three years. By June 1999, the program had approved grants to 971 firms, with 1090 export projects. In total, US$26.6 million in matching grants was approved, equivalent to export projects for about US$52 million. With respect to the number and amount of matching grants approved, the program fully met its objectives. However, the Brazilian devaluation at the beginning of 1999 affected the export plans of a number of enterprises that had Brazil as their main export rnarket. Many of these enterprises decided not to implement their export plans and canceled their participation in the program (60 enterprises). Other enterprises were able to implement only part of their export plans (about one third of the export plans had reached less than 50% of implementation at the end of the program). As a result, by the end of the program in July 1999, only US$17. 7 million had been disbursed in matching grants, 67% of the approved amount. 17. One of the objectives of the PREX was to extend the program to enterprises outside Buenos Aires. In fact, the formula for the incentive payment to the MSC included a premium for matching grants approved to SMl,s outside Buenos Aires. This objective was largely accomplished, since about 56% of the assisted firms were located outside Great Buenos Aires. About 94% of the total amount of grants were disbursed to enterprises located in Provincia de Buenos Aires, Capital Federal, Cordo ba, Santa Fe and Mendoza. 18. The PREX program succeeded in it objective of assisting small enterprises: 83% of the assisted SMEs had sales of less thai US$5.0 million in 1995, 59% had between 11 and 50 workers, and 31% had never exported before. 19. Although the program was open to enterprises in all areas of activity, the great majority of assisted enterprises were in manufacturing: 68% produced manufactures of industrial origin, 19% produced manufactures of agricultural origin, 5% produced primary products, and 5% produced services. The manufacturing subsectors with greater representation were: food and beverages (18% of the enterprises), machinery and equipment (10% of the enterprises), and autoparts (8% of the enterprises). These are all industrial subsectors in which Argentina has had some export success. 20. The export services approved umder the matching-grant program were consistent with the kind of services identified as important for export development during project preparation. About 2,639 activities were approved as part of the 1,090 export plans. The export services fell into the following categories: 60% for market development and commercial promotion (including fairs); 12% for product development; 12%No for enterprise development (including quality control certification); and 16% for other enterprise development activities. -6 -

11 21. Impact Assessment of the Cost-sharing Grant Scheme. An Impact Assessment Survey was carried out in mid-1999 covering 100 assisted enterprises and a control group of 100 similar enterprises not assisted by the program (see Annex 8). The survey results show the following: * In 1998, two thirds of the assisted enterprises had exported compared with only one third of the non assisted ones * More assisted enterprises had introduced new products since 1995 (77%) than the non assisted ones (64%) * More assisted enterprises had introduced product improvements (85%) than the non assisted ones (72%) * More assisted enterprises had used new distribution channels since 1995 (59%) than the non assisted ones (26%) * About 49% of the assisted enterprises had diversified their export markets * More assisted enterprises had introduced training programs (78%) than the non assisted ones (44%) * More assisted enterprises had reduced their production costs (54%) than the non assisted ones (46%) * More assisted enterprises had adopted international quality control programs (61%) than the non assisted group (36%). Only assisted enterprises had obtained ISO 9000 and ISO 9002 * About 70% of the assisted enterprises indicated that they had achieved the objectives they had when they joined the program 22. Directory of Consultants. The project allocated US$700,000 to hire a contractor to compile and operate a worldwide Directory of Consultants and Market Information Suppliers. Expecting delays in the implementation of this sub-component, the UAP compiled (with resources from other external sources) a temporary directory of consulting services. Eventually, this temporary directory collected information about several hundreds of consulting firms worldwide. During the mid-term review, the Bank and the Project Coordinator concluded that there was no need to hire a contractor to compile and run a directory, and that the funds could be reallocated to other components. The objectives of this sub-component were met without utilizing the allocated funds. Local business and consultants benefited from participating as service providers under the project. One of the most important benefits is that the Directory of Consultants has remained accessible to the SMEs in Argentina, after the completion of the project. 23. There is some evidence that by stimulating demand for business support services the PREX program helped to expand the supply of these services in Argentina. Interviews with local consultants revealed that many of them had benefited from participation as service providers under the scheme. They had been incorporated in the Directory of Consultants managed by the UAP, which was accessible to the assisted SMEs. Anecdotal evidence suggests that consultants specialized in quality control systems and related advisory services benefited greatly from the scheme. One indicator of the program's success was a reported decrease in the fee for obtaining ISO 9000 certification. -7 -

12 B. Institutional Strengthening Component 24. In terms of outputs, the Institutional Strengthening component is rated moderately satisfactory, because only two of its four sub-components were implemented. 25. Simplification of Export Procedures. At the end of 1996, the UAP argued that the objectives of the Simplification of Export Procedures sub-component had been met by the recent merger of the Customs and Tax Authorities into the AFIP (Federal Administration of Public Income) and other reforms being carried out under the Second Reform of the State Project. Interviews with exporting firms in the context of preparing this ICR suggest that the problems of cumbersome and slow export proceclures, and the difficulties experienced by exporters to get reimbursement of VAT, had not been solved. Although during the second half of 1999 the AFIP gave priority to reimburse the VAT to exporting SMEs, reimbursement are still lagging due to budgetary constraints. 26. Within the simplification of export procedures sub-component, the project had allocated funds to improve the certification procedures of the Argentine Institute of Plant Health (IASCAV) and the National Institute of Animal Health Services (SENASA), for agricultural and food exports. These two agencies merged during project implementation. The UAP decided not to implement this component because of alleged irregularities in the use of international funds within the merged agency. 27. Strengthening of Quality Certification System. The main objective of this sub-component was to help the National Institute of Industrial Technology (INTI) to obtain international accreditation of ten of its research centers (each research centers is composed generally of more than one laboratory) for quality control certification. Once accredited, INTI's centers would be able to provide internationally recognized test results, which could be used by their clients in the international market. Accreditation of INTI's centers would reduce the cost of such certifications, which had otherwise to be obtained in other countries. Implementation of this sub-component started as late as June 1997, due to uncertainty about the availability of counterpart funds. After initial delays, a Letter of Agreement between the UAP and INTI was approved, and a contract was awarded to an international firm to assist INTI to upgrade its laboratories. INTI accomplished many of the objectives agreed during appraisal. A variety of measures to improve INTI's performance and increase private sector participation were undertaken, including the development of an Strategic Plan for INTI for 1998/2000; the design of the National Plan of Quality and Technological Innovation; the certification of INTI according to ISO 9000; the international accreditation by the United Kingdom Accreditation Services (UKAS) of 91 tests in 13 of INTI's centers according to ISO 25 (four in 1999 and 9 in the first half of 2000); and the preparation of a proposal for a new Charter (Organic Law) for INTI. The project helped INTI's laboratories upgrade their physical infrastructure to satisfy the requirements of international accreditation. Another eleven of INTI's research centers are expected to be accredited before end Strengthening of Export Promotion Policy Formulation and Rationalization of Export Promotion programs. The objective of this sub-component was to rationalize the large -8 -

13 number of export promotion agencies and programs operating in Argentina at the federal, and provincial level. This component was to be coordinated by a recently created Coordinating Unit for Export Promotion Agencies (UCPFE), with representatives from the Ministry of Economy and External Relations. Implementation of this component was delayed because of difficulties to agree on the use of the funds for a rationalization plan. In November 1997, the UAP hired a well-known international consultant to help develop terms of reference for the studies contemplated under this sub-component. However, the recommendations of the consultant were not implemented. The transfer of the UAP from the Ministry of Economy to the Secretariat of Small and Medium Enterprises (SEPYME), in early 1998, complicated even more the coordination problems. Eventually, the sub-component was dropped and the funds reassigned to other components of the project. As a result, the objectives of this sub-component were never met. 29. Strengthening of Commercial Promotion Services. The objective of this sub-component was to improve the promotion services provided by Fundacion ExportAr. By the time of the project's mid-term review, the funds allocated for this sub-component had been almost fully utilized in improving the quality of the participation of Argentine enterprises in two international fairs in Fifteen exporters that participated in Fancy Food Fair, that took place in Philadelphia, were able to arrange export sales in an amount of US$565,000. In addition, 198 out of 1138 new contacts were expected to end in new export sales. Twenty six firms that participated in the Saitex Fair, that took place in South Africa, were able to sell US$280,000 worth of products in that fair. Although the objectives of this sub-component were achieved with the participation of Argentine businessmen in fairs and in trade missions, the small amounts allocated to this sub-component has resulted in a modest impact in terms of new sales. However, its impact in terms of learning experience for ExportAr and for the participant exporters is likely to be significant. 4.3 Net Present Value/Economic rate of return: n.a 4.4 Financial rate ofreturn: n.a. 4.5 Institutional development impact: 30. The institutional development impact of the project was modest. The project had an important institutional development impact on INTI, as indicated above. INTI has now the capability to certify Argentine exporters on quality control systems. The project also had some institutional development impact on Fundacion ExportAr, which is now better equipped to organize the assistance of Argentine exporters in international fairs. However, the project had no institutional development impact on other export-related Government agencies, because of the failure to implement the sub-components for the simplification of export procedures and of export promotion policy rationalization. The institutional development impact of the project on the UAP is likely to be modest, due to the various transfers of this unit among Government agencies, which involved three different project coordinators and many changes of key staff. -9-

14 5. Major Factors Affecting Implementation and Outcome 5.1 Factors outside the control of government or implementing agency: 31. Project implementation was negatively affected by a deterioration in Argentina's economic conditions as a result of the international financial crisis that began to impact Argentina's economy during the third quarter of Argentina's GDP growth during 1998 was 3.9% compared to 8.1% in The downward trend accelerated in 1999, which ended with a negative GDP growth of -3.0% for the whole year. The economic recession affected particularly hard the manufacturing sector, which accounts for half of the value of goods produced. Manufacturing output declined by 7.1%, during the first semester of 1999, compared to the first semester of Brazil's devaluation in January 1999 had a major impact on Argentina's exports, because Brazil is Argentina's main trading pawtner, absorbing one third of Argentina's total exports and 50% of Argentina's manufactured exports (and as much of 85% of automobile exports). Argentina's manufactured exports of agricultural origin, which had increased by 8% in 1997, contracted by 4% in 1998 and by 6% in The contraction in exports of industrial manufactures was even more dramatic. After growing by 29% in 1997, they only grew by 3% in 1998, and contracted by 20% in Total exports to Mercosur in 1999 contracted at twice the rate (-25%) of total exports (-12%). 32. By the time the crisis hit Argentina, a total of US$26.6 million in cost-sharing grant had been approved under the PREX, but only US$10.5 mnillion had been disbursed. The difference was due to the time lag between approval of the grants and the reimbursement to the enterprises of half of the cost of the consulting services agreed under the export program. Many enterprises that were in the process of implementing their export plans wanted to export to Brazil. When the prospects of exporting to Brazil disappeared, many of them decided not to implement their export plans, to avoid investing resources needed for more immediate needs. By the time the PREX program stopped disbursing in July 1999, a total of 60 previously approved cost-sharing grants had been canceled without any implementation, and 32% of them had reached completion rate of less than 50%. 33. It is difficult to ascertain to wahat extent these results were due to the crisis, or whether some cancellations would have occuwred anyway for other reasons more directly linked to the management of the program, or to the enterprises involved. One factor that delayed the implementation of the PREX, and may explain some cancellations, was the inability of many of the assisted firms to comply with the implementation timetable of their export projects. From the beginning, the program underestimated the time that took the assisted firms to implement the export plans agreed under the program, particularly in the case of more complicated plans such as those involving obtaining an ISO 9000 certification, which took an average of one year. Although the project closing date was extended to allow for these delays, when this component of the loan was closed at the end of July 1999, a nujmber of firms were still in the implementation stage. 5.2 Factors generally subject to government control: 34. Management changes in the UAP. Project implementation was negatively affected by frequent changes in the institutional set up of the UAP and in the position of project coordinator

15 From 1996 to 1998, the UAP was within the Secretariat of Commerce and Investment, which was later transformed into the Secretariat of Commerce, Industry, and Mining. Both Secretariats were within the Ministry of Economy. This first change resulted in the replacement of the first project coordinator and large part of the UAP staff, producing delays in project implementation. In February 1998, the UAP was transferred to the newly created SEPYME, under the Ministry of the Presidency. This transfer, which involved another change of project coordinator and of UAP staff, also affected project implementation. 35. Changes in other implementing agencies. The implementation of the Institutional Strengthening Component was negatively affected by changes in the implementing agencies and lack of coordination and rivalry between Government agencies. The implementation of the sub-component for simplification of export procedures was never implemented, partly because of the elimination of the Sub-Secretary of Economic Deregulation (which was in charge of the implementation of this sub-component), and lack of Borrower's commitment to further deregulation. Implementation of the sub-component for the rationalization of export promotion policies had to be agreed between SEPYME and the UCPFE (which remained under the Secretariat of Commerce, Industry and Mining). The dialogue between both agencies was affected by rivalry generated during the transfer of the PREX program to SEPYME. Eventually, through Bank intervention, both agencies agreed to reduce the scope of the program, but lack of counterpart funds precluded any use of the Bank funds allocated to this sub-component. 36. Lack of counterpart funds and restrictions in the use of project funds. Lack of counterpart funds affected the project since early in its implementation. During the initial stages, implementation of the Institutional Strengthening Component was delayed because the UAP was reluctant to sign Letters of Agreement with the agencies in charge of implementing the different sub-components, until the Government provided the agreed counterpart funds. 37. During the last quarter of 1998 and the first quarter of 1999, the Government, faced with difficult budgetary constraints, cut counterpart funds and restricted the use of loan funds to many programs financed with international loans. Even if Bank funds were available, SEPYME was unable to pay contracts already executed or under execution, because they had to be financed 70% with Bank funds and 30% with counterpart funds. This included, among others, payment of work already executed under contracts with the MSC, the auditing firm, the firm in charge of conducting the impact assessment, and the firm assisting INTI in the certification of its laboratories. Since the contract with the MSC stipulated fines in case of. payment delays, the lack of counterpart funds generated additional costs for the Borrower in project management. 5.3 Factors generally subject to implementing agency control: 38. Each change in Project Coordinating Unit generated serious delays in the implementation of both project components. The first project coordinator had been involved in the design of the project and was convinced of the importance of giving the administration of the PREX to a private contractor. Shortly after the signing of the management contract, but before the project started operating in full, a second project coordinator took over. The second project coordinator was initially ambivalent to the concept of using a private management contractor to operate an important government program. It took time and efforts by the Bank to establish a good working

16 relationship between the new UAP and the MSC. Nevertheless, the work of the MSC was frequently delayed by second guessing by the UAP of the decisions taken by the MSC about which enterprises to accept in the program and the nature of the export plans. The initial ambivalence of the new project coordinator also affected the launching of the project by the Govenmment, which was delayed by about three months. It is apparent that the new authorities wanted to make sure that the project had a good reception among the enterprises before committing itself to supporting it. 39. When the project was transferred from the Secretariat of Commerce and Industry to SEPYME, in February 1998, the agency was very new and lacked the staff and the technical infrastructure to manage the project effectively. Things improved with the hiring of a new Project Coordinator in May 1998, but the effectiveness of the project coordinator was limited by having to respond to many layers of decision making within SEPYME. The controversy that had surrounded the transfer of the project from one agency to the other contributed to an atrnosphere of confrontation between SEPYME and the MSC. This confrontation initially centered on the insistence of SEPYME to review all the invoices presented by the assisted SMEs before replenishing the revolving fund used by the MSC to reimburse the SMEs of expenditures already made under the agreed export plans. This insistence produced delays in disbursing the grants, which severely affected the reputation of the project. The confrontation between SEPYME and the MSC intensified due to frequent delays (some times of several months) in payment of the MSC's fees. Although some of the delays can be attributed to lack of counterpart funds, there were instances in which the payment to the MSC was not made although the funds were available. The management of the relationship between the UAP and the MSC tended to occupy the attention of the UAP, leaving little room for dealing with the Institutional Strengthening Component. 5.4 Costs andfinancing: 40. The project's original closing date was extended from 07/31/98 to 09/30/99, due to the following: (i) the initial delay of about three months in the Government's official launching of the PREX program (see par. 38); and (ii) the longer than expected time that it took between the approval and the disbursement of the grants. The grants were disbursed to the assisted enterprises once they had received and paid for the consulting services agreed under the program. This process took longer than expected due to lack of experience of many of the assisted SMEs in dealing with consultants. Project cost was less than originally estimated for two reasons: (i) the cancellation of many approved grants (US$8.9 million) following Brazil's devaluation in January 1999 (see par. 33); and (ii) the failure of the UAP to implement several sub-components of the Institutional Strengthening Component. 6. Sustainability 6.1 Rationale for sustainability rating: 41. Enterprise Assistance Component. Project sustainability is likely to be strong for this component. Sustainability should be seen in terms of the likelihood that the assisted enterprises will be able to improve their export performance, if they have the opportunity to implement what -12 -

17 they have learned from the services and information obtained under the project. Sustainability, however, will be affected by the economic context in which these enterprises operate, in particular the availability of credit to implement the changes required to improve export performance. Although SMEs are particularly vulnerable to external factors, sustainability of this component is reinforced by the fact that the program selected enterprises that were relatively close to being able to export and had the financial ability to do so. Also, the assisted enterprises are predominantly in areas (food processing, machinery and equipment, autoparts) in which Argentina has some comparative advantages. It is also likely that the local consultants, who were established to provide export-related services to the enterprises under the program, will continue to provide their services to these and other enterprises, since firms are now more aware of the benefits of the use of consultants to improve export performance. 42. Institutional Strengthening Component. Project sustainability is likely to be strong for the sub component dedicated to strengthen Argentina's quality certification system. The United Kingdom Accreditation Services (UKAS) has accredited 13 of INTI's laboratories, and nine more are in the process of accreditation. With this accreditation, INTI can provide internationally recognized laboratory results, which could be used by their clients in the international markets. INTI appears well-positioned to continue serving the needs of the country's firms with a well-trained technical staff and administration, and state-of-the-art laboratory equipment. It has also made substantial progress in increasing quality awareness in the country. The experience gained by Fundacion ExportAr under the project may help this agency to better help other firms to participate in international fairs. 6.2 Transition arrangement to regular operations: 43. The authorities are in the process of implementing another cost-sharing grant scheme to support restructuring of SMEs. This operation, with IDB financing, has a similar design as the PREX, with two management service contractors. Although this project is not exclusively oriented towards developing export capabilities, it includes some aspects common to the PREX such as ISO 9000 certification. 7. Bank and Borrower Performance Bank 7. 1 Lending: 44. Bank performance in the identification, preparation, and appraisal of the project was highly satisfactory. The Bank responded in a timely manner to the Government's request to prepare an operation in support of a rapid export response by Argentine enterprises. Project preparation was a collaborative effort between the Bank and the Borrower. During project preparation, the project team was able to respond satisfactorily to a number of concerns with respect to the cost-sharing grant scheme, raised during the project concept discussions. These concerns were whether: (i) Argentina followed a liberal trade policy and the scheme was not a substitute for trade liberalization; and (ii) the cost-sharing grant scheme did not violate WTO principles. To respond to these concerns, a study was conducted that showed that Argentina had adopted a liberal trade regime and had phased out some remaining export subsidies. The task team also requested the opinion of an expert on WTO regulations who determined that the

18 cost-sharing grant scheme did not violate the WTO principles, because the grants supported the acquisition of knowledge and information by the assisted enterprises, and were not directly linked to exports. Also during project preparation, the Bank involved well known international experts in cost-sharing grants programs anid Metrology, Standards, Testing and Quality (MSTQ) infrastructure, in order to expose the Argentine government participants to international experience in these areas. At Bank suggestion, several studies were undertaken by the Argentine counterpart during project preparation. The findings of these studies were incorporated into the design of the project. 45. The Argentine authorities had initially asked for a much larger loan (US$150 million) to help improve the competitiveness of ]ocal exporters. The Bank was justified in its insistence on starting with a small pilot project. The agreement with the authorities was that preparation of a second project would be considerec if the evaluation of the matching grants program was positive. The Bank was also justified in its insistence that the matching grants program had to be managed by a private company under a management service contract. 46. The Bank was effective in working with the Borrower in project design. The most complex part of the design was that of the management service contract. At the time, there was not much experience in the Bank, or elsewhere, with the use of this type of contracts to manage a Government program. The Bank included in the preparation team two experts (a Bank staff and a consultant) in this area, who worked with the Argentine authorities in the design of the payment formula of the management service contract. The design of the incentive scheme in the contract took several months of discussions between the Bank staff and the Government team. The Bank staff thought that the incentive formula proposed by the Government team was too complex and tried to convince the Government team to simplify it. But, the Government team insisted in maintaining their design, and the Bank staff could not rely on previous experience to convince the Government team to accept a simpler formula. The Bank concerns proved right during project implementation (see section 3). In hndsight, the Bank should have insisted in simplifying the incentive formula before approving the loan. 7.2 Supervision: 47. Bank performance during supervision was satisfactory. Bank staff was very effective in reacting to the disruption to project imriplementation generated by the frequent changes in project coordinator. When the initial project coordinator was replaced, the Bank staff met with the new coordinator to discuss the rationale for the use of a MSC to manage the matching grants scheme and to clarify the different roles of the project unit and the MSC. The Bank staff exercised appropriate flexibility in agreeing to modify some aspects of the incentive scheme of the management contract and the Operating Manual, to reflect the realities of project implementation. The main agreed change to the incentive formula of the management service contract involved the interpretation of lags in the disbursement schedule of the cost-sharing grants. One of the criteria to determine the variable payment to the MSC was the degree of compliance with the schedule of grant disbursements agreed with the participating SME, which depended on the schedule of implementation of the consulting services contracted under the program. During project implementation it became evident that SMEs tended to overestimate their capacity to implement these programs. The UAP, the MSC and the Bank agreed to extend the schedule of disbursement - 14-

19 in specific cases in order not to penalize the MSC. This flexibility was in line with the "pilof' nature of the project. The changes were documented in several "Actas" signed by the two parties and approved by the Bank. The Bank used effectively the mid-term review, in November 1997, to recommend changes to improve project implementation, including actions to: address the paucity of disbursements, reduce the excessive information that the UAP required from the MSC, and to strengthen the monitoring of the SMEs' implementation of the approved export plans. Despite the Bank's efforts the monitoring of the implementation of the export plans was not fully accomplished. 48. When the project was transferred to a third project manager (SEPYME), Bank staff organized a supervision mission to help in the transition. Subsequently, Bank staff worked intensively to solve problems between the MSC and SEPYME's authorities concerning the disbursement of the approved matching grants. Many of these problems were generated by a discrepancy between the new UAP and the MSC with respect to the documentation that the MSC had to present to justify disbursements. After numerous discussions, the parties agreed on the minimum documentation required to justify disbursements; and that the UAP would relay on ex-post reviews of that documentation, to avoid affecting the timely reimbursement of the grants to the participating enterprises. The Bank staff also intervened to solve difficulties between the UAP and INTI that arose in the course of INTI's implementation of its institutional strengthening program. Much of the Bank's mediating efforts were in terms of daily telephone conversations with the parties involved. These efforts are not adequately reflected on the supervisory forms 590 and project status reports. The Bank staff was diligent, although less successful, in its efforts to accelerate implementation of other sub-components of the Institutional Strengthening Component. In hindsight, Bank staff should have explored alternative institutional arrangements to implement the other sub-components, particularly the one aimed at simplifying export procedures. 49. The Bank staff responded effectively to the crisis produced by the failure of the Government, in early 1999, to allow disbursements of loan funds and to allocate counterpart funds to the project. By mid-march 1999, SEPYME owed about US$1.5 million in matching grants to the assisted enterprises; about US$1.1 million to the MSC; and other lesser amounts to the auditors and to the consulting firm assisting in the accreditation of INTI's laboratories. In March 1999, the Bank sent a letter to the Secretary of SEPYME, with copy to the Minister of Finance, explaining that these funds were needed to pay for services already rendered, and that lack of disbursements penalized SMEs that had already paid for the consulting services agreed under the project. The Bank warned the Government that this situation violated the Loan Agreement, and that unless the problems were solved, the Bank could suspend disbursements. The Government responded to the Bank letter by first, allowing the use of the loan funds to reimburse the assisted firms, and later on, by assigning counterpart funds. As a result, the project was upgraded to satisfactory. 7.3 Overall Bank performance: 50. Bank performance in preparation and supervision of the project was satisfactory. There was a high degree of continuity in the Bank staff. The original task manager stayed until the end of the project. The project benefited from frequent supervision missions (at least two per year)

20 involving Bank staff, and the use of high level consultants to evaluate progress in project implementation. The supervision missions provided relevant advice and followed up on the previous recommendations and agreed actions. It was also found that the recommendations were proper and timely. The supervision fcrms were found to be realistic with strong attention to the development objectives. Performance ratings of the project were realistic. The project was rated unsatisfactory almost at the end of the project, due to the Government's failure to allocate counterpart funds. The project rating was upgraded to satisfactory in the last project status report after the Government provided the required counterpart funds. Overall, implementation problems and progress were well identified and fiollow up actions were taken. Borrower 7.4 Preparation: 51. Borrower's performnance during project preparation was highly satisfactory. The Argentine authorities were heavily involved in project preparation. The potential demand for export-oriented consulting services was carefully assessed during the appraisal process. A project preparation team, operating under the Deputy Minister of Economy, visited several countries, including Ireland, Great Britain, South Korea, and Japan to review different export support programs. The preparation team conducted several studies as part of project preparation, including an enterprise survey (339 fmns) and interviews with focus groups (20) to estimate the potential demand for export related consulting services. A study was also undertaken of the steps involved in concluding an export operation, to assess the main regulatory and administrative barriers to foreign trade operations. Also, a review was carried out of the country's MSTQ infrastructure. The preparation team undertook a systematic analysis of the key agencies involved in export procedures and certification (public and private) to identify their institutional capacity gaps utilizing a methodology proposed by the Bank. The results of these studies were the basis for the design of the project's institutional strengthening component. The Argentine team produced a comprehensive background report that was integrated into the appraisal effort. 7.5 Government implementation performance. 52. The Government's performance during project implementation was mixed. The authorities changed the project management three t:imes for political and not for technical reasons. The first transfer occurred shortly after the signing of the management service contract. Because of the transfer, the launching of the matching grants scheme (PREX) by the Government was delayed by three months. This delay obliged the MSC to step in and do its own information campaign to generate interest in the project. Later on, the transfer of the project management from the Ministry of Economy to SEPYME created strong resentments between SEPYME and other implementing agencies, with negative effects on project implementation. 53. The Government failed to provicle the needed counterpart funds on a timely fashion at the beginning and at the end of project implementation. Lack of counterpart funds at the beginning of the project delayed the official launching of the PREX and the implementation of the Institutional Strengthening Component. During the first quarter of 1999, the Government, faced with tight budgetary constraints, restricted the use of the Bank loan and failed to provide counterpart funds. Lack of funds greatly affected the project at a time when the bulk of the assisted enterprises were

21 requesting reimbursement of the matching grants. This situation generated a flood of complains from the participating SMEs. The MSC, unable to reimburse the SMEs, blamed SEPYME. This soured the relationship between the MSC and SEPYME. Lack of payment to the MSC complicated even further the strained relations between SEPYME and the MSC. 7.6 Implementing Agency: 54. Performance of the Project Coordinating Unit (UAP). The UAP's performance during project implementation was mixed. The main problem with implementation of the PREX program was that each of the two project coordinators that replaced the original one (who had been involved in project design), had difficulties in delegating the administration of the PREX to the MSC (which had been selected under the initial project coordinator). Their reluctance to delegate to the MSC led them to keep a large staff in the UAP that tended to duplicate the work of the MSC, increasing the cost of managing this component. 55. The difficulties with the second project coordinator centered on the approval of the matching grants. The UAP initially questioned all the decisions of the MSC in this respect, greatly delaying the approval process. The situation improved after the Bank staff intervened to clarify the respective roles of the UAP and the MSC and a revised Manual of Policies and Procedures was produced. SEPYME, the third project coordinator, took over the project when most of the PREX funds had been committed and the reimbursement to the firms was gathering momentum. Initially, SEPYME reviewed all the documentation presented by the MSC and suspended reimbursement of the MSC's revolving fund if it found that any document was missing. This scrutiny delayed the reimbursement to the assisted enterprises. The Bank argued that the ex-ante review process was unnecessary because the contract's incentive scheme allowed the UAP to suspend payments if an ex-post evaluation revealed that the export plans did not meet the agreed criteria, and because the revolving fund was covered by a bank guarantee that could be called if the external auditors found any wrong use of the funds. Eventually, SEPYME agreed to do its review ex-post, to avoid delays in reimbursing the assisted enterprises. The strained relationship between SEPYME and the MSC led to a delay (until practically the last day) in extending the Management Service Contract, which was necessary because of the extension of the loan's closing date. This delay generated a high degree of uncertainty for the private operator, who could not take measures in advance to keep operating the program with adequate staff resources. 56. The UAP's performance with respect to the implementation of the institutional strengthening component was mixed. The working relations between SEPYME and the beneficiary agencies, particularly the UCPFE, were strained, which made it difficult to agree on the work programs of several sub-components. Also, the preoccupation of the UAP with managing the relationship with the MSC, left it little time or energy to deal with the other project component. The UAP's performance with respect to the INTI sub-component was satisfactory, although INTI's management complained that the unit lacked experience in project management and procurement knowledge to be able to manage the bids and procurement of specialized equipment that its laboratories required. SEPYME never presented the Annual Project Implementation Plan, despite the Bank's repeated reminders. 57. Political considerations, particularly the expectation of a new administration taking over

22 shortly after project completion, made it difficult for SEPYME to reach a compromise with the MSC on the final 7% payment. Both parties agreed to send the issue to arbitration. The final payment was linked to the percentage of matching grants disbursed at the end of the contract. The MSC did not reach the required 98% disbursement level due to a combination of problems: (i) its failure to properly supervise the assisted SMEs and cancel the matching grants of those that were not using them effectively; (ii) the emphasis given by the MSC to meet the quantitative targets (number of firms and amount of grants approved) of the incentive formnula, at the expense of quality considerations, which resulted in a number of export plans not being finally implemented; (iii) the decision of maniy enterprises to cancel their program in the aftermath of Brazil's devaluation; and (iv) the fai]ure of the Borrower to provide funds to reimburse the assisted enterprises on a timely fashion. Arbitration procedures had not been finalized at the time of preparation of this ICR. 58. Performance of the Beneficiary Agencies. INTI and Fundacion ExportAr were the only beneficiary agencies fully committed to implementing their programs under the Institutional Strengthening Component. INTI's president took a very active interest in ensuring the implementation of the program of accreditation of its laboratories for quality control certification. The perforrnance of this agency was highly satisfactory. The performance of Fundacion ExportAr was also satisfactory. 59. Performance of the Managemnent Service Contractor (MSC). The MSC was very effective in promoting the program threugh trade associations and in generating confidence on the program among potential users. It met its quarterly targets in terms of number of grants and total amounts approved. Judging from the results of the two client satisfaction surveys carried out by consulting firms, the MSC was successfll in providing its services on a transparent manner and in assisting the firms in preparing their export plans. The MSC was less successful in supervising the implementation of the export plans by the SMEs. The need to meet quarterly targets (to qualify for the maximum incentive payment) put strong pressure on the MSC to emphasize quantity over quality. The staff of the MSC had greater incentives to bring in new firms than to supervise and follow up the implementation of the export plans by the assisted firms. The MSC should have been more diligent in canceling the agreements with the enterprises that were unable to implement their export plans within a reasonable amount of time. During the ICR mission, some firms reported that the MSC had only visited the firm at the beginning of the program. The delays in receiving payment from the UAP for services rendered and the uncertainties about the extension of the management contract also affected the performance of the MSC, because it had to reduce its staff during the most active period of the program. The MSC did not devote sufficient resources to administrative aspects of the project. For example, the managerial information system was very rudimentary and an interconnected computer system among the offices was never put in place. The external auditor noted the lack of adequate documentation of the MSC's activities. This deficiency made the app:.oval process by the UAP more difficult. 7.7 Overall Borrower performance: 60. Overall, Borrower's performance was satisfactory with reservations. The Borrower performed well in the preparation phase, but not as well during project implementation. The continuous difficulties between the Borrower and the MSC seem to indicate that the Argentine - 18-

23 authorities were not ready to operate a public funded program through a private sector operator. 8. Lessons Learned 61. Several lessons can be obtained from this project: * Use of matching grants to improve competitiveness. The PREX program generated a very positive response among SMEs. The evidence from the impact assessment shows that participating SMEs had started to make changes (such as the introduction of new or improved products, new channels of distribution, quality control programs) that will put them in better position to become exporters. However, the full impact of a matching grant program and its cost effectiveness has to be evaluated providing sufficient time for the assisted enterprises to be able to implement the recommendations of the consulting services co-financed by the program. * Link between matching grants and access to credit One lesson from the PREX is that the impact of the matching grant program can be hindered by lack of access to credit by the SMEs to finance the consulting contracts involved in an export plan, and to implement the changes recommended by the consultants. The PREX tried to reduce the financial burden on the SMEs by allowing partial grant payments when the implementation of the export plans required long time, such as in ISO certification. One way to reduce the financial burden could be to divide the export plans into stages or modules, tu reimburse the enterprises more frequently. This approach would also allow for more flexibility to introduce changes in the export plans, based on the implementation experience Also, a program of this type should contact the country's financial institutions to explore if SMEs participating in such a program could have greater access to export credit. * Use of a management service contract to manage a Government progranl The management of a publicly financed program by a private operator through a management service contract was necessary to ensure independence in the approval of the grants and to provide technical assistance to the beneficiary SMEs. However, in the context of Argentina, it was difficult for a Government agency to delegate the administration of the PREX to a private operator and limit its role to that of ex-post evaluator of the performance of the MSC. This is not just a problem of attitude on the part of Govermnent officials. Part of the difficulties are related to the way in which the performance of these agencies is evaluated in the Government. For example, in the PREX program the UAP had to give its no objection to the grants approved by the MSC and to approve the replenishment of the revolving fund used by the MSC to disburse the approved grants. The UAP felt that the rules of the Government auditing agencies: AGN, SIGEP, obliged it to analyze ex-ante all the documentation involved. On the other hand, it was difficult for the private operator to understand the amount of infornation that a public sector agency requires prior to taking any decision involving the use of public resources. Any future use of this type of management arrangement should ensure that the contract is very clear on the delimitation of the respective roles of the Government and the private operator to avoid imprecision that could lead to disputes, and that the information requirements are clearly specified and approved by the official auditors. The official auditors should be involved in the design of the loan, to ensure that they agree with the

24 principle of ex-post evaluation. * Use of an incentive scheme in the management service contract. The experience with the PREX confrms that the use of a performance-linked incentive scheme in the management contract is a powerful tool to ensure that the MSC meets the contract's performance targets, even if changes in the government management unit changed the approaches and expectations about the project. However, the PREX experience also shows that if the performance-linked incentive scheme is difficult to interpret and puts too much emphasis on quantitative targets, the quality of the outcome may suffer. An incentive scheme in this type of contracts should be simple to interpret, have fewer variables, and have realistic targets. The incentive scheme should contemplate the possibility that the performance targets may be affected by external factors beyond the control of the MSC. In this context, it may be better not to have a final payment based on the percentage of the approved grants actually disbursed. Also, the work of the MSC should be evaluated taking into account not only the reimbursed amounts but also the quality of the activities exectuted under the export plans. Since it is very difficult to establish quality parameters for the diverse type of activities involved, an external panel should periodically review a sample of export plans and assess their quality. Finally, the management contract should have provisions to ensure the use of high quality staff by the MSC throughout the life of the contract. a Technical assistance programs involving multiple agencies. The experience of this project shows that a technical assistance program that involves several implementation agencies, can be undermined by lack of funding and inter-institutional rivalries, even if the program was well designed. A technical assistance program should minimize the number of implementing agencies involved. * Need for continuity in project management. The experience with this project shows that good project design and strong Borrower's commitment at the initial stage of a project are necessary but not sufficient conditions to ensure good project implementation. Good project implementation requires continuity in project management. At the least, the continuity of the technical staff involved in project implementation should be preserved. * Need to ensure timely availability of projectfunding. The experience, with this project is that the uncertainty about availability of funding (either of Government counterpart funds or of loan funds) as a result of the Government's budgetary difficulties can seriously disrupt project implementation. 9. Partner Comments (a) Borrower/implementing agency: 62. Annex 10 is a summary translation of the Final Report prepared by SEPYME. The original in Spanish is available in the Project Files. (b) Cofinanciers: (c) Other partners (NGOs/private sector):

25 10. Additional Information Not applicable

26 Annex 1. Key Performance Indicators/Log Frame Matrix Outcome I Impact Indicators: lndicatormatrix Projected in last P_R_ ActualLatest Estimate Output Indicators: tndicatorlmatrix Cost-sharing grants approved Simplification of export procedurest Accreditation of INTl's laboratones Plan to rationalize export promotion programs. Strengthening of ExportAr commercial promotion services. Projected in last PSR AcWallLatest Estimate Cost-sharing grants approved to at least 615 By the end of July 1998, cost-sharing grants SMEs by the end of July 1998 (the 8th had been approved to 968 SMEs quarter of implementation) Simplification plans in place for: custom No simplification plans were developed procedures, certification procedures and tax reimbursement Accredit3tion of 8 of INTl's laboratories Four of INTl's laboratories were accredied in 1999; another 9 are expected to be accredited before end 2000 Rationalization plan in place by end 1997 No rationalization plan was developed Strengthening ExportAr's participation in two Participation of ExportAr in two international intemational fairs fairs was strengthened End of project

27 Annex 2. Project Costs and Financing Annex 2. Project Costs and Financing Project Cost by Appraisal Estimate Actual/Latest stimate Percentage of Component US$ Million -US$ Million Appraisal Program Administration on UAP % Advertising/Promotion % Enterprise Assistance Program: - Cost Sharing Grant % - MSA % -Directory of % Consultants Institutional Strengthening - Strengthening of UAP % Policy - Formulation UCPFE % - Fundacion Export Ar % -Simplificacion of % Procedures - ANA/DGI % - INTI % - SENASA % - IASCAV % - INAL % - Quality Control - INTIURAM % Base Cost % Price Contingencies % I TOTAL % Project Cost By Component -23 -

28 Category Amount ot the Total Loan Allocated Disbursements Cost-Sharing Grants $27,000, $17,809, Consultant's Services $3,300, $1,553, Management Services Contractors Part 1 $5,000,000.DO $5,435, Part 2 (Directory of Consultants) $700, $0.00 Goods $11,800, $155, Training $700, $0.00 Special Account $286, $38,500,000.'00 $25,239, Project Costs by Prc,curement Arrangements (1JS$ million equivalent) Expenditure Procurement Method Procurement method Categories Appraisal Estimate Actual/Latest Estimate Consulting ICB NCB Other NBF Total ICB NCB Other NB Total Services Cost-sharing MSA (5.0) (5.0) (5.72) (5.72) F Grants (27.0' (27.0) (17.81) (17.81) Directory MSA (0.7) (0.7) Technical Assistance (5.2) (5.2) (1.55) (1.55) Training 0.2) 0.2 (0.1) (0.1) Goods (0.5) (0.5) (0.16) (0.16) Recurrent Adm. Costs Total Costs (5.0) (33.5) (38.5) (25.24) (25.24)

29 Project Financing Appraisal Estimate (US$M) Actual/Latest Estimate(US$M) Local Foreign Total Local Foreign Total Source Costs Costs Costs Costs Government Beneficiaries Bank TOTAL

30 Annex 3: Economic Costs and Benefits N.A

31 Annex 4. Bank Inputs (a) Missions: Stage of Project Cycle No. of Persons and Specialty Perfonnance Rating (e.g. 2 Economists, I FMS, etc.) Implementation Development Month/Year Count Specialtv Progress Objective Identification/Preparation I I I I I Task Manager Senior Consultant (export promotion) Trade Financing Specialist Principal Operations Officer Institutional Capability Gaps Specialist Appraisal/Negotiation I Task Manager 2 Senior Consultants (export promotion, MSTQ) I Management Contracts Specialist I Trade Economist Supervision 5/ Task Manager S S 11/ Task Manager S S 4/ Task Manager S S 11/ Task Manager S S I Consultant 3/ Task Manager S S 8/ Task Manager S S 2/ Task Manager U S ICR 10/ Consultant S S (b) Staff Stage of Project Cycle Actual/Latest Estimate No. Staff weeks US$ (.000) Identification/Preparation Appraisal/Negotiation Supervision ICR Total Note: The ICR includes a cost of US$15,186 for the workshop with stakeholders

32 Annex 5. Ratings for Achievement of Objectives/Outputs of Components (H=High, SU=Substantial, M=Modest, N=Negligible, NA=Not Applicable) Rating E Macro policies O H OSUOM O N * NA Sector Policies O H OSUOM * N O NA z Physical O H OSUOM O N * NA Z Financial O H OSUOM O N * NA X Institutional Development 0 H O SUO M 0 N 0 NA Environmental O H OSUOM O N * NA Social Poverty Reduction O H OSUOM O N * NA Gender OH OSUOM ON *NA Z Other (Please specify) O H OSUOM O N * NA F Private sector development 0 H 0 SU O M 0 N 0 NA E Public sector management 0 H 0 SU * M 0 N 0 NA MOther (Please specify) O H OSUOM O N * NA

33 Annex 6. Ratings of Bank and Borrower Performance (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HU=Highly Unsatisfactory) 6.1 Bank performance Rating M Lending OHS Os OU OHU I Supervision OHS OS OU OHU? Overall OHS OS OU O HU 62 Borrower performance Rating Z Preparation OHS OS O U O HU Z Government implementation performance O HS * S 0 U 0 HU Z Implementation agency performance OHS OS Q U O HU Z Overall OHS OS O U O HU -29 -

34 Annex 7. List of Supporting Documents * Quarterly Progress Reports prepared by the UAP * First Impact Assessment Survey Report * Client Satisfaction Survey Report, January 1999 * Second Impact Assessment Survey Repon- * Audit Reports (several) * Borrower's Final Evaluation Report prepared by SEPYME, December

35 Annex 8. Beneficiary Survey Results Impact Assessment of the Cost-sharing Grant Scheme. The Staff Appraisal Report (SAR) stipulated that the effectiveness of the cost-sharing grant scheme would be subject to three reviews by independent parties. A first assessment had to be carried out for the mid-term review, which was to take place 12 months after effectiveness or after half of the cost-sharing grant funds had been disbursed. A second assessment had to be carried out 18 months after effectiveness or when two-thirds of the funds had been disbursed, and a third assessment was to be carried out three years after loan effectiveness. Only two impact assessments were carried out. The first assessment was completed in October 1997, to have the results ready for the mid-term review. The second assessment was completed in July 1999, three years after effectiveness. The first assessment covered only 50 firms (18% of the total assisted firms by June 1997). The objective of the assessment was mainly to evaluate the management arrangement of the PREX program. The assessment concluded that the management arrangement was effective and that the program was progressing satisfactorily. The results showed that the assisted firms considered that the program was managed by the MSC in a very transparent and effective manner. The survey also asked questions about the impact of the program on the assisted enterprises. About 30% of the firms reported that the program induced them to participate in export activities sooner. About 40% of the firms said that they would not have undertaken the investments required to export without the program. However, after only twelve months of operation, this survey was too early to provide an assessment of the impact of the program on export performance. The conclusion that the PREX program was well managed was supported by another survey of client satisfaction that covered 201 assisted firms. The average score given by the assisted firms to 21 questions related to the MSC's performance ranged between "excellent" and "good" (the resulting average score was 1.65 based on possible answers to each question ranging from excellent "1" to bad "5". The highest score was for the question asking if the enterprise thought that it had received an honest and transparent treatment. The second survey, completed in mid-1999, was designed to assess the impact of the PREX on the assisted enterprises. The focus was on the effect of the program in changing the enterprises' approach to export activities as a result of the information and export consulting services received under the program. Although it was thought that it was too early to expect an impact of the program on the export performance of the assisted enterprises, the survey also tried to identify if they had increased their exports as a result of their participation in the program. The survey, designed and conducted by a consulting firm, covered a sample of 200 firns. One hundred firms were selected from a group of assisted firms that had already received at least 25% of the matching grant amount approved for them. A second group of one hundred firms that had not participated in the program was selected as a control group. It was composed of firms similar in activity and size to the firms of the first group. Firms from both groups were asked questions regarding exports and export markets covering the period from 1996 to They were also asked to describe changes in products, production processes and organizational changes aimed at improving export performance during that period. The main results of the second impact assessment are the following:

36 * Impact on exports. About 69% of the SMEs in the sample of assisted firms had exported at some stage, compared with 42% of the non assisted finns. Nearly half of the PREX firms that had exported, did so before joining the program. In both groups of firms, the number of firms that had exported increased every year between 1995 and In each year the percentage of firms that had exported was larger in the PREX sample than in the control group. In 1998, about two third of the firms in the PREX group had exported compared to only one third of firms in the control group. Exports increased for the assisted firms from 26.8% to 32.4% of sales (for the group). * Impact on the introduction of new or improved products. The survey looked at the effect of the program in helping firms to introduce new products and to improve the quality of their products, both for the local and export markets. More firms in the PREX group had introduced new products into the markets since 1995 (77%) than in the control group (64%). Also, more PREX firms (85%) had introduced significant improvements in their products than in the control group (72%). * Impact on the use of new channels of distribution. More firms in the PREX group had used new international channels of distribution since 1995 (59%) than those in the control group (26%). About 49% of the firms in the PREX group reported that they had diversified their export markets since Impact on the use of training programs. PREX firms were more likely to have adopted training programs for their staff in 1998 (78%) than those in the control group (44%). In both groups the number of enterprises with training programns had increased between 1995 and * Impact on cost-reduction. About half of the firms in both groups had adopted cost-reduction programs since 1995, but the number of firms that adopted these programs was somewhat higher in the PREX group (54%) than in the control group (46%). * Impact on the use of quality control prograsn. About 80% of the firms in both groups had adopted quality control programs since 1995, although a higher percentage of the assisted firms (61%) had adopted international quality control programs compared with the control group (36%). The assisted firms were the only ones that had obtained ISO 9000 and ISO 9002 certification. Firms that received assistance on quality upgrading introduced several changes such as information technology for quality control, changes in the plant lay-out and training programs for their staff. Immediate results were a decline in production defects by an average of 20% and a reduction in operating costs by an average of 10% among the assisted enterprises. * Fulfillment of thefirms' objectives. About 70% of the PREX firmns indicated that they had achieved the objectives they had for participating in the program: 13% indicated that all the objectives had been achieved and 58% indicated that they had been partially achieved. Among the reasons given for not having filfilled their objectives under the program, the most important were: lack of working capital (27%); crisis in international markets (20%); lack of

37 experience in international markets (18%); and barriers in importing countries (8%). * Qualitative evaluation of the program. The consulting firm conducted in-depth interviews with groups of assisted enterprises to get their opinion about the program and its impact on their activities. Most of the enterprises indicated that the program had been implemented with great transparency and that the advice that they had received was good. Many thought that the program had helped them to leam how to handle consultants, to understand what to request from them, how to write terms of reference, and how to evaluate assignments to get useful services. The ICR mission interviewed ten beneficiary firms and found that their views of the program were generally positive. Most of the entrepreneurs interviewed indicated that the program had demonstrated to them the benefits of using independent consulting services to improve export performance

38 Annex 9. Stakeholder Workshop Results During ICR preparation, two stakeholder workshops were held to discuss the results of the project and the lessons learned from it. The first was withi a sample of assisted enterprises and the executing agency (SEPYME), and the second was with SEPY'ME and the other executing agencies (INTI and Fundacion ExportAr), to discuss the draft ICR. The results of the second workshop are incorporated throughout the text of the ICR. This Annex summarizes the main results of the workshop with the assisted enterprises. Objective. The objective of the workshcp was to conduct a beneficiary impact assessment in a participatory manner. Methodology. A sample of assisted enterprises was invited to participate in the workshop. The sample was selected to include enterprises that had reached different levels of implementation of their export plans, from 10% or less to 90% or more, and that were located both in Buenos Aires and in the provinces. About 35 enterprises were contacted; 25 confirmed their participation; and 12 attended the workshop, with a total of 16 entrepreneurs participating in it. The workshop started with a brief introduction in which the Bank staff and SEPYME's management welcomed the participants. This was followed by a slide presentation by the Bank staff to clarify some details of project design and operation. Subsequently, the participants were divided into two discussion groups and asked to answer the following questions: (i) how the program had benefited their enterprises; (ii) if their objectives for entering the program had been achieved; (iii) if they had left the program, what were the reasons for doing so; and (iv) if they had to design a similar program which aspects of the design would they change. Results. The discussion was very animated and the participants provided many interesting and useful observations. The general consensus was that despite some bureaucratic problems during implementation, the PREX had been the most useful program ever implemented in Argentina for small and medium exporters and that they strongly supported ae follow up operation. The participants also welcomed the opportunity to participate in the workshop and indicated that the workshop demonstrated that the Bank and the SEPYMIE authorities were genuinely interested in listening to them and in improving the design of any future similar project. The main benefits of the PREX program perceived by the participants were: * The PREX program started at a time in which many enterprises faced the need to export, but were uncertain about how to do it The Frogram gave them confidence in their ability to compete internationally * The program helped several of the enterprises to improve their export performance by: (i) helping them to improve the quality of their products; (iii) helping them to get certification for quality control (ISO 9000 and one of them ISO 9002); and (ii i) exposing them to potential buyers (through participation in fairs) * The program helped several of the enterprises to improve the management of the enterprise (to become better organized, plan ahead) * The program helped most of the enterprises to improve their competitiveness and image in the local market * The program taught many enterprises how to work with consultants Most of the participants indicated that their main objectives for joining the program had been fully

39 achieved. These were: * increase sales * obtain quality control certification * widen the market, including exports Some of the limitations of the program were: * The PREX was too rigid in that it did not allow for changes in the export plans once these had been approved * There was limited follow up of the export plans by the MSC * There were cases of late reimbursement to the enterprises of half of the consulting costs (the grants) and cases in which no reimbursements were made due to the cancellation of the program * There was limited information about the program. Many enterprises found about the program by chance * There was limited awareness about the time limitation for implementing the export plans Some of the recommendations were: * A program of this type should have flexibility to change the export plans in light of the changing circumstances that the enterprises face. One way to achieve flexibility could be to approve a framework export plan and implement it in stages, subject to periodic reviews to change the subsequent stages if necessary v A program of this type should also include some retroactive financing to cover activities implemented by the enterprise during the period of review and approval of the export plans * The account executives of the MSC should have a limited number of enterprises under their supervision to allow for more interaction with the enterprise during the implementation stage * The program should be complemented with general information about the requirements imposed by importing countries * The Directory of Consultants, currently on SEPYMES' Web Page should include more information about the consultant's areas of expertise and fees

40 Annex 10. Borrower's Final Evaluation Report Summary I. INTRODUCTION In accordance with World Bank guidelines, and given the conclusion of the PREX, the Government of Argentina, through its Secretariat for Small and Medium-Enterprise (SEPYME), has prepared the attached Final Report about the Program's implementation and operations, its cost-benefits and the evaluation of the achievements of the Project's objectives. It is worth highlighting that three different Government agencies were in charge of the implementation of the Program. Since its design and launching in 1996 until February 1998, it was under the Commerce and Investments Secretariat which would later become the Secretariat of Industry, Commerce and Mining, both under the Ministry of Economy and Public Works and Services. Since then, and until its closing on July 31, 1999, its implementing agency was the Small- and Medium-Scale Enterprise Secretariat (SEPYME) under the Nation's Presidency. The PREX was transferred to SEPYME on February 23, 1998, to concentrate in one place all the programs of assistance to small and medium enterprises (SMEs). I. ENTERPRISE ASSISTANCE PROGRAM COMPONENT This component's activities were initiated in August 1996, through the Management Contract between the Secretariat of Industry, Commerce and Mining and the Trade Development Institute of Ireland (TDI) responsible for managing the PREX Program. The expected duration of the Program was 8 months, namely until July Two extensions of the administrative period were agreed upon, with the purpose of completing the disbursement of all the funds allocated to cost-sharing grants: Non Reimbursable Funds (ANR). The effective closing of the PREX Program was on 07/31/99. The PREX's objective was to support a minimum of 900 enterprises, by disbursing US$27 million in ANRs during a 30-month period, with an average of US$30,000 in ANR per export plan. Once the implementation phase was conmpleted, the results show that 968 enterprises received assistance, through the approval of 1060 export projects, which jointly represent a total of US$17,785, in ANRs disbursed to the enterprises. (The report presents tables and graphs with data on ANRs approved and disbursed by size of enterprise, geographical location, sector of activity, and type of consulting services pirovided.). Evaluation of the Component Contractually speaking, the management contractor (CSG) was responsible for implementing the Enterprises Assistance Program. Three indicators have been taken into consideration to evaluate the implementation of this component: * Executed ANRs: are the amounts of ANRs effectively disbursed to the enterprises by the CSG, representing a total of US$17,785, It is important to mention that in order to analyze this indicator, it should be considered that not all projects were implemented in their

41 totality, and that different percentages of implementation exist among them. * ANRs with 0% implementation: are the amounts of ANRs which initially were committed for projects approved for a total of US$1,072,452.30, corresponding to enterprises that did not carry out any of the predicted activities. * Canceled ANRs: is the amount of ANRs that the CSG reported to the UAP as canceled totaling US$1,172, These last two indicators show the weaknesses of the CSG regarding the support it should have given to the enterprises to enable them to redefine their export plans and actively continue implementing their activities in the Program during a macroeconomic crisis. (The report presents several tables, only the conclusions are presented here.). Executed ANRs. When the Program ended, 32% of the projects had implementation percentages lower than 50% and only 38% of the projects were more than 80% implemented. On the other hand, only 175 projects (16%) were more than 95% implemented. From these data it can be inferred the little monitoring and follow-up of the projects undertaken by account executives of the CSG. ANRs with 0% Implementation. There were 60 projects not implemented, representing an ANR amount equal to US$1,072,452. Canceled ANRs. The total original amount of ANRs committed was US$1,172,363, for 60 export projects, approved by UAP. This commitment was later canceled by the CSG. UAP-CSG Relation The CSG, assumed the responsibility for managing the Program's service platform for the enterprises and for assigning the co-financing funds, guaranteeing a wide geographic coverage in the country in the provision of support services to SMEs. The evaluation presented above shows the weaknesses in the program's management. Any future PREX should redefine the elements that rule the UAP-CSG relationship. Performance of the CSG Assistance to the enterprises started in September 1996, when the CSG opened its first office in Buenos Aires. During program implementation there were three offices of services to SMEs located in Cordoba, Rosario and Mendoza. In this regard, it is worth noting that the implementation average of export plans of beneficiary enterprises of the Program was 62% in the Rosario office, 60% in Cordoba and 37% in Mendoza. As contractually established, the CSG constituted a bank guarantee (to cover the account from which it disbursed the ANRs to the assisted enterprises) with an expiration date which was the same as the closing date of the management contract (7/31/99). Therefore, disbursements for ANR payments were authorized only until 06/04/99, to allow for the administrative time necessary to implement this guarantee in case the External Audit would determine so

42 In spite of the extension of the Program's closing date until 07/31/99, the established targets with regard to disbursement of ANR were not met. This has affected negatively the possibility that the CSG would have access to the so called "Final Payment" contemplated as part of its variable remuneration. The CSG was not able to disburse the minimum of 80% of the total funds allocated to ANR, as established in the Mlanagement Contract (Section 4.2, Appendix GC, Special Terms of the Contract) thus it is understood that it is not entitled to this "Final Payment". The controversy generated around whether the payment corresponds or not, has resulted in TDI's request to initiate arbitration procedures, currently in progress. Consequently, a revision of the UAP-CSG relations should contemplate a redefinition of concepts such as "committed" and "implemented" with respect to ANR disbursements, as well as redesigning the way in which the remuneration of the CSG is determined. With respect to this last concept, it is understood that the complexity of the formula established by the Management Contract for the determination of fees was the origin of certain maladjustment presented and, on the other hand, its application has really not contributed to providing incentives for the CSG to reach the program's objectives, consequently to the disbursement of all the funds assigned to the enterprises. Cost-Effectiveness of the Services (The report presents a table with amounts of ANRs disbursed by quarter compared to the payments made to the CSG.). If one brings into consideration the total amount of fees which have been paid to CSG, it can be concluded that the cost of managing each export plan is approximately US$6,000. It is worth mentioning that the total amount in ANRs disbursed for 394 projects is lower or equal to the management cost. Those 394 projects correspond to: 60 canceled projects, 60 projects with 0% irmplementation, 16 rejected projects and 258 implemented projects with ANRs equal or less than US$6,000. Another important variable to take into consideration to assess the CSG's follow-up of the enterprises, related to the implementation of their export plans, is the delays that they incurred in requesting the first disbursement after signing the letter of agreement. According to what is established in the Policies and Procedures Manual, the period permitted to make the request could not exceed four months even though this period was surpassed in 42% of the cases. Only 59% of the Projects requested the first disburseinent within the established deadline. Given that during the implementation period, the CSG did not strictly control the maximum time allowed for enterprises to make the first disbursement, the average time for project implementation was extended for longer than had originally been considered adequate. This variable should also be subject to revision. Analysis of the Implementation Curve of the ANRs The established goals of the PREX Program were: to support, at least, 800 enterprises (although the contract with the CSG specified 800, it is worth mentioning that SAR had predicted a total of 900 enterprises), and to disburse US$27 million during 30 months, with an average of US$60,