Document of The World Bank Report No: ICR2698 IMPLEMENTATION COMPLETION AND RESULTS REPORT (IDA-H2930 TF-91620) ON A GRANT IN THE AMOUNT OF SDR 10 MILLION (US$15 MILLION EQUIVALENT) TO THE REPUBLIC OF MALAWI FOR A BUSINESS ENVIRONMENT STRENGTHENING TECHNICAL ASSISTANCE PROJECT (BESTAP) June 25, 2013 Finance and Private Sector Development AFCS3 Africa Region CURRENCY EQUIVALENTS (Exchange Rate Effective December 31, 2012) Currency Unit = Malawian Kwacha (MWK) US$1.00 = MWK 324 FISCAL YEAR January 1 -December 31 ABBREVIATIONS AND ACRONYMS AfDB African Development Bank BESTAP Business Environment Strengthening Technical Assistance Project BRPs Business Residency Permits BUGS Business Growth Scheme CAS Country Assistance Strategy COMESA Common Market for Eastern and Southern Africa DB Doing Business Dfid Department for International Development DO Development Objective DRG Department of Registrar’s General EC European Commission ERR Economic Rate of Return FIAS Foreign Investment Advisory Services FDI Foreign Direct Investment FPD Financial and Private Sector Development FSTAP Financial Sector Technical Assistance Project GAC Governance and Anti-Corruption GDP Gross Domestic Product GNI Gross National Income GoM Government of Malawi HIPC Heavily Indebted Poor Countries ICA Investment Climate Assessment ICR Implementation Completion and Results Report ICT Information and Communication Technology IDA International Development Association IFC International Finance Corporation ISR Implementation Status and Results Report M&E Monitoring and Evaluation MCCCI Malawi Confederation of Chambers of Commerce and Industry MEPC Malawi Export Promotion Council MGDS Malawi Growth and Development Strategy MGDS II Second Malawi Growth and Development Strategy MIPA Malawi Investment Promotion Agency MITC Malawi Investment and Trade Centre MITPSD Ministry of Industry, Trade and Private Sector Development MOIT Ministry of Industry and Trade MoJ Ministry of Justice MRFC Malawi Rural Finance Company MSB Malawi Savings Bank MWK Malawi Kwacha NGO Non-Government Organization NPV Net Present Value OPCS Operations Policy and Country Services PAD Project Appraisal Document PC Privatization Commission PDO Project Development Objective PIU Project Implementation Unit PSD Private Sector Development PPD Public Private Dialogue PPP Public Private Partnership PURP Privatization and Utility Reform Project QAG Quality Assurance Group TEPs Temporary Employment Permits WB World Bank WBG The World Bank Group Vice President: Makhtar Diop Country Director: Kundhavi Kadiresan Sector Manager: Irina Astrakhan Project Team Leader: Brian G. Mtonya ICR Team Leader: Francisco Moraes Leitao Campos MALAWI Business Environment Strengthening Technical Assistance Project (BESTAP) CONTENTS Data Sheet A. Basic Information B. Key Dates C. Ratings Summary D. Sector and Theme Codes E. Bank Staff F. Results Framework Analysis G. Ratings of Project Performance in ISRs H. Restructuring I. Disbursement Graph 1. Project Context, Development Objectives and Design 1 2. Key Factors Affecting Implementation and Outcomes 9 3. Assessment of Outcomes 18 4. Assessment of Risk to Development Outcome 30 5. Assessment of Bank and Borrower Performance 32 6. Lessons Learned 35 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners 37 Annex 1. Project Costs and Financing 39 Annex 2. Outputs by Component 40 Annex 3. Economic and Financial Analysis 58 Annex 4. Bank Lending and Implementation Support/Supervision Processes 67 Annex 5. Beneficiary Survey Results 69 Annex 6. Stakeholder Workshop Report and Results 78 Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR 83 Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders 87 Annex 9. List of Supporting Documents 88 Annex 10 Other relevant information 89 Annex 11 Figures 93 MAP 95 A. Basic Information Business Environment Strengthening Country: Malawi Project Name: Technical Assistance (BESTAP) Project ID: P103773 L/C/TF Number(s): IDA-H2930,TF-91620 ICR Date: 06/28/2013 ICR Type: Core ICR GOVERNMENT OF Lending Instrument: SIL Borrower: MALAWI Original Total XDR 10.00M Disbursed Amount: XDR 10.00M Commitment: Revised Amount: XDR 10.00M Environmental Category: C Implementing Agencies: Ministry Industry Trade and Private Sector Development Cofinanciers and Other External Partners: European Commission B. Key Dates Revised / Actual Process Date Process Original Date Date(s) Concept Review: 01/08/2007 Effectiveness: 10/10/2007 10/10/2007 Appraisal: 04/02/2007 Restructuring(s): 11/19/2010 Approval: 05/24/2007 Mid-term Review: 09/30/2010 10/18/2010 Closing: 12/31/2012 12/31/2012 C. Ratings Summary C.1 Performance Rating by ICR Outcomes: Moderately Satisfactory Risk to Development Outcome: Moderate Bank Performance: Moderately Satisfactory Borrower Performance: Moderately Satisfactory C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings Moderately Quality at Entry: Moderately Satisfactory Government: Unsatisfactory Implementing Quality of Supervision: Moderately Satisfactory Moderately Satisfactory Agency/Agencies: Overall Bank Overall Borrower Moderately Satisfactory Moderately Satisfactory Performance: Performance: i C.3 Quality at Entry and Implementation Performance Indicators Implementation QAG Assessments Indicators Rating Performance (if any) Potential Problem Project Quality at Entry No None at any time (Yes/No): (QEA): Problem Project at any Quality of Yes None time (Yes/No): Supervision (QSA): DO rating before Moderately Closing/Inactive status: Satisfactory* *Based on the ISR submitted on December 20, 2012. D. Sector and Theme Codes Original Actual Sector Code (as % of total Bank financing) Banking 5 0 Central government administration 58 64 General industry and trade sector 25 20 Law and justice 8 16 SME Finance 4 0 Theme Code (as % of total Bank financing) Export development and competitiveness 13 0 Legal institutions for a market economy 13 10 Micro, Small and Medium Enterprise support 25 16 Other Private Sector Development 24 4 Regulation and competition policy 25 70 E. Bank Staff Positions At ICR At Approval Vice President: Makhtar Diop Gobind T. Nankani Country Director: Kundhavi Kadiresan Michael Baxter Sector Manager: Irina Astrakhan Demba Ba Project Team Leader: Brian G. Mtonya Constantine Chikosi ICR Team Leader: Francisco Moraes Leitao Campos ICR Primary Author: Francisco Moraes Leitao Campos ii F. Results Framework Analysis Project Development Objectives (from Project Appraisal Document) The objective of the Project is to support capacity development and investment climate reforms in order to accelerate economic growth. One of the medium term outcomes in Malawi's Country Assistance Strategy (CAS) is an improved climate for private business. This project is one instrument by which the Bank seeks to achieve this outcome. This will be done by reducing the regulatory burden and costs of doing business; improving service delivery to the private sector; access to finance for Small and Medium Enterprises (SMEs); and by strengthening support for technical and business management skills, thereby improving productivity and competitiveness at the firm level. Revised Project Development Objectives (as approved by original approving authority) The objective of the Project is to improve the ease of doing business processes in Malawi in order to increase foreign and domestic private sector investment in the economy. (a) PDO Indicator(s) Original Target Formally Actual Value Values (from Revised Achieved at Indicator Baseline Value approval Target Completion or documents) Values Target Years Indicator 1 : Cost to formally start a business as a % of GNI per capita Value quantitative or 215.7% 70% 90.9% Qualitative) Date achieved 11/30/2006 11/30/2011 11/30/2011 Comments Target largely met. Over 85% of the reduction in cost was achieved. This is using (incl. % the DB 2012. The figure for the DB 2013, which is not in the M&E framework, achievement) is 83.7%, continuing the downward trend. (b) Intermediate Outcome Indicator(s) Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised approval Completion or Target Values documents) Target Years Indicator 1 : Legislative backlog of bills impacting on cost of doing business Value (quantitative 40 0 16 or Qualitative) Date achieved 05/24/2007 12/31/2012 12/31/2012 Comments Target largely met. Twenty four bills / acts / regulations were either enacted or in (incl. % force, or stand before the Cabinet or Parliament. Another 22 are in advanced achievement) stages before submission to Parliament. Indicator 2 : Time to settle commercial disputes Value 96 days in CD_HC 337 180 (quantitative / 432 in DB's iii or Qualitative) Magistrates Court Date achieved 11/30/2006 11/30/2011 11/30/2011 Target met. The Commercial Division of High Court takes an average of 96 days Comments to settle a commercial dispute. The indicator in the PAD indicates that the source (incl. % is the DB, which only follows the Magistrate's Court, not addressed by activities achievement) in BESTAP. Indicator 3 : Automated business registry database at DRG Value Not listed as target at (quantitative entry. The baseline at Yes Yes or Qualitative) restructuring was "No". Date achieved 11/19/2010 12/31/2012 12/31/2012 Comments (incl. % Target met. Database system is installed but it has not been launched. achievement) Indicator 4 : % Business Registration transactions that are electronic Value Not listed as target at (quantitative entry. The baseline at 50% 0% or Qualitative) restructuring was 0%. Date achieved 11/19/2010 12/31/2012 12/31/2012 Comments Target not met. Database system is installed (see Indicator 3), but it is not yet (incl. % operating. achievement) Indicator 5 : Time to formally start a business Value (quantitative 37 10 5 39 or Qualitative) Date achieved 11/30/2006 11/30/2011 11/30/2011 11/30/2011 Comments Target not met. There is a number of regulations that once effective can improve (incl. % this rating in the DB. Trend went downwards due to DB reclassification. achievement) Indicator 6 : % Electronic Transactions in Department of Lands Value Not listed as target at (quantitative entry. The baseline at 50% 100% or Qualitative) restructuring was 0%. Date achieved 11/19/2010 12/31/2012 12/31/2012 Comments Target met. System of automated registration of property operational even in (incl. % regional centers. achievement) Indicator 7 : Time to register property Value (quantitative 118 60 69 or Qualitative) Date achieved 11/30/2006 11/30/2011 11/30/2011 Comments Target largely met. 84 percent of the target met with a steady decline in the time (incl. % to register property. It was below the target in DB2011 but then went back above achievement) it. Indicator 8 : No of achieved reforms on Ease of Doing Business per year Value Not listed as target at 3 per year 2 per year iv (quantitative entry. Two reforms on or Qualitative) DB per year at restructuring. Date achieved 11/19/2010 11/30/2011 11/30/2011 Comments Target partially met. There is a pipeline of reforms with new legislation that the (incl. % MITPSD is hoping to obtain recognition from the DB report, but that has yet to achievement) happen. Indicator 9 : % Business Permits through electronic transactions Value Not listed as target at (quantitative 50% 0% entry. 0% at restructuring or Qualitative) Date achieved 11/19/2010 12/31/2012 12/31/2012 Comments (incl. % Target not met. Intervention was added at restructuring but did not take place. achievement) Indicator 10 : Time taken to process Business permits Value Not listed as target at (quantitative entry. 40 days at 20 40 or Qualitative) restructuring Date achieved 11/19/2010 12/31/2012 12/31/2012 Comments (incl. % Target not met. Intervention was added at restructuring but did not take place. achievement) Indicator 11 : Turnover in firms accessing BUGS vs control group Value No rigorous impact (quantitative 0% 20% evaluation was or Qualitative) possible Date achieved 05/24/2007 12/31/2012 12/31/2012 No counterfactual was captured to measure the difference between firms Comments accessing BUGS and control group. Impact assessment suggests positive trends (incl. % but difficult to attribute to the project despite the team's best intentions to achievement) measure impact. G. Ratings of Project Performance in ISRs Actual Date ISR No. DO IP Disbursements Archived (USD millions) 1 12/20/2007 Satisfactory Satisfactory 2.00 2 06/19/2008 Satisfactory Satisfactory 2.00 3 12/29/2008 Satisfactory Satisfactory 3.20 Moderately Moderately 4 04/13/2009 4.48 Unsatisfactory Unsatisfactory 5 12/12/2009 Moderately Satisfactory Moderately Satisfactory 6.53 6 06/30/2010 Moderately Satisfactory Moderately Satisfactory 9.53 7 05/02/2011 Moderately Satisfactory Satisfactory 14.32 8 12/28/2011 Satisfactory Satisfactory 15.44 v 9 06/29/2012 Satisfactory Satisfactory 15.44 10 01/06/2013 Moderately Satisfactory Moderately Satisfactory 15.44 H. Restructuring (if any) ISR Ratings at Amount Board Restructuring Disbursed at Restructuring Reason for Restructuring & Approved Restructuring Date(s) Key Changes Made PDO Change DO IP in USD millions The proposed restructuring included a change to the project development objective (PDO), the dropping of two 11/19/2010 Y MS MS 12.08 subcomponents due to a change in government policy and the reallocation of funds to support two new activities. If PDO and/or Key Outcome Targets were formally revised (approved by the original approving body) enter ratings below: Outcome Ratings Against Original PDO/Targets Moderately Satisfactory Against Formally Revised PDO/Targets Moderately Satisfactory Overall (weighted) rating Moderately Satisfactory I. Disbursement Profile vi 1. Project Context, Development Objectives and Design 1.1 Context at Appraisal Country and Macroeconomic Background 1. At the time of BESTAP preparation, Malawi was among the ten poorest countries in the world. GDP per capita was just US$167 in 2004, or approximately US$640 in terms of purchasing power parity. The macroeconomic environment was experiencing an improvement at the time of project preparation with inflation decreasing from an annual average of 28 percent in the period between 1999 and 2002 to an average of 13 percent in the period between 2003 and 2006 (EIU, 2013). Economic growth was starting to pick up but with high volatility on a year-on-year basis. GDP grew on average 0.3 percent annually during the period of 1999 to 2002 and improved to an annual average of 5.3 percent in the following four years. Still, growth in two of these four years was close to zero. At the time of the project design, given population growth rates, it was estimated that 5 to 6 percent real economic growth would be required to have any impact on poverty. Sector Background 2. Against this backdrop, establishing a less volatile economic growth path was a critical objective for Malawi’s government. The 2006 Investment Climate Assessment (ICA) based on a survey of 300 firms mentions the story of a small private sector struggling to compete in a difficult business environment. The issues raised in the note included macro-instability, high cost of finance, limited access to electricity, low availability of skilled workers, and crime and corruption. 3. BESTAP’s PAD went further in this analysis, seeking to identify the sources of Malawi’s low economic growth: limited investment and limited productivity growth. Net fixed investment at just 8.1 percent of GDP was half the average for Sub-Saharan Africa, while growth in labor productivity was virtually non-existent. Malawi’s trade structure was also unbalanced with the country continuing to maintain a dependence on tobacco exports1. The preparation of BESTAP was also informed by the results of the Doing Business (DB) surveys. Overall, Malawi was ranked 110 out of 175 countries on the 2007 2 Doing Business indicators 3 . According to the PAD, the poor regulatory 1 In 2006, the share of tobacco in exports was 56 percent. There were signs of an encouraging if fragile trend towards export diversification, but tobacco continued to dominate. While tea and sugar were growing in relative importance, this was largely at the expense of minor cash crops such as coffee, pulses and cotton. 2 The DB business from 2007 is published in late 2006. The same is true for any other year (2008 published in late 2007, etc). 3 Although this rank does not seem particularly low in comparison to the GDP level of the country, the government and the team preparing BESTAP thought that there were sub-aspects that were particularly problematic for improving the climate of investment, including issues of “enforcing contracts� (ranked 134 1 environment for business activities was a critical obstacle to attracting the investment needed to generate economic growth. Many of the critical laws that governed private sector activity were out of date and in need of wholesale revision. An excessively bureaucratic process of registering the ownership of land in Malawi slowed the process of transferring private property. This was identified by the government as a significant deterrent to new foreign investment where land was required, and to new credit for local investors where collateral was needed. Estimates from the second Integrated Household Survey (2004-2005) suggest that the informal sector represented 93 percent of the non- farm small scale enterprises in the country. The DB indicators showed high costs and processes of registering enterprises. The PAD argued that these costs of formalization were limiting the growth of a vibrant SME sector. 4. The economy was characterized by weak contract enforcement and limited access to commercial justice. Criminal and political cases tended to absorb most of the capacity and received higher priority in Malawi's judicial system, resulting in a large backlog of commercial cases. The net effect was a general reluctance within the private sector to extend credit or enter into contract arrangements to all but the most trusted of clients. 5. Limited access to finance was also a leading constraint to private sector investment, as per the ICA. Still, there were positive signs in Malawi’s macro-economy. The Reserve Bank of Malawi’s base rate was reduced from as high as 45 percent in 2004 to 20 percent in November 2006. Although not tackled by the project, the PAD and other preparatory documents suggest that other issues were also important at that time, including: (a) inadequate infrastructure services and (b) low technical skills4. Rationale for Bank assistance 6. The Government of Malawi (GoM) had at the time finished a national development strategy, the Malawi Growth and Development Strategy (MGDS), which recognized the private sector as the “engine of economic growth� and as an important factor in reducing poverty. Malawi’s attainment of irrevocable debt relief under the Highly Indebted Poor Countries (HIPC) initiative contributed to the return to macro- economic stability. Malawi’s macro management regained the confidence of the international community, demonstrated by the resumption of budget support by bilateral donors, an IMF Poverty Reduction and Growth Facility (PRGF), and through debt relief. 7. A stable macroeconomic environment is a pre-requisite for sustainable private sector development (PSD). Thus, Malawi’s gains in macro stability meant the in DB2007) and “trading across borders� (ranked 153). Furthermore, there were sub-indices where a number of countries were in the process of doing reforms, and hence the relative position would likely fall if nothing was done in that area. That is the case the for instance for “starting a business�, where the ranking was 89 in 2007, but was trending downwards (108 in 2008 and 122 in 2009). 4 This would partly be tackled by the project through the matching grant scheme if there was demand for these training activities. 2 government was increasingly looking to a “second generation� of reforms to stimulate private sector investment. BESTAP was part of that process. 8. The Ministry of Industry, Trade and Private Sector Development (MITPSD) program aimed to address most of the ICA / DB constraints through a combination of different projects and programs. This involved the support of local and international partners including donors: EU and UNDP were doing separate work on trade policy, and USAID on agribusiness and SMEs’ access to finance. The WB had previously supported reform in the utilities sectors and in improving the business environment (PURP5). In that vein, BESTAP was the continuation of the specialization of the WB on the investment climate area. 9. BESTAP was seen as an integral part of the Bank’s CAS, which was in line with the MGDS. BESTAP was particularly related to the second strategic area of the CAS: putting in place a foundation for longer term economic growth through improved infrastructure and investment climate. 1.2 Original Project Development Objectives (PDO) and Key Indicators (as approved) 10. As stated in the PAD, “the objective of the Project is to support capacity development and investment climate reforms in order to accelerate economic growth. One of the medium term outcomes in Malawi’s Country Assistance Strategy (CAS) is an improved climate for private business. This project is one instrument by which the Bank seeks to achieve this outcome. This will be done by reducing the regulatory burden and costs of doing business; improving service delivery to the private sector; access to finance for Small and Medium Enterprises (SMEs); and by strengthening support for technical and business management skills, thereby improving productivity and competitiveness at the firm level�. 11. The project focused on investment climate reform and institutional capacity building. The theory of change identified by the project team is summarized in Annex 11 (adapted from the PAD) – tackling business environment as a means of promoting investment, which could lead to economic growth and job creation. The area within the frame was the target of the project. 12. The M&E Indicators were defined with the clear objective of seeking to improve the Doing Business rating. Project Outcome Indicators at approval were (see Data Sheet for targets): (i) Cost to formally start a business (ii) Turnover in firms accessing the matching grant scheme (BUGS) versus control group 5 PURP – Privatization and Utilities Reform Project 3 13. The two PDO indicators were complemented by the following seven Intermediate Outcome Indicators: Improving economic governance, property rights institutions and business facilitation (i) Legislative backlog of laws impacting cost of doing business (ii) Time to settle commercial disputes (iii)Time to formally start a business (iv) Time to register real property Strengthening private sector development policy and services (v) Time to obtain business licenses at the Malawi Investment and Trade Centre (MITC), using the license for construction in the DB Promoting access to finance and enhancing productivity of SMEs (vi) Number of SME start-ups accessing the established SME fund (vii) Turnover in firms accessing the matching grant scheme (BUGS) versus control group6 1.3 Revised PDO (as approved by original approving authority) and Key Indicators, and reasons/justification 14. The main Level 1 restructuring included a change to the PDO, the dropping of two subcomponents and the reallocation of funds to support two new activities. According to the restructuring paper, the PDO and the outcome indicators needed to be modified to be consistent with the type and level of interventions supported by the project. The formulation of the PDO as approved by the Board was seen at restructuring to be high level and not directly attributable to the operation. 15. The revised PDO was: “The objective of the Project is to improve the ease of doing business processes in Malawi in order to increase foreign and domestic private sector investment in the economy.� 16. The core objective remains the same: as per the original CAS outcome in the original PDO: improving climate for private business. The revised PDO has two main changes: (i) drops the direct mention of capacity development although those activities are not dropped from the project core focus; and (ii) rewords the change that may happen with the ease of doing business: instead of economic growth, the revised PDO highlights investment, and in particular brings-in the distinction between foreign and local investment. The theory of change is graphed in Annex 11. 17. Investment and growth are affected by a number of factors; thus, it is difficult to attribute changes in these outcomes to the project. In practice, excluding growth and 6 Notice that this last intermediate indicator was at the same time a PDO indicator. This was likely not on purpose and this was corrected at restructuring, from which it was kept solely as an intermediate indicator. 4 investment from the picture, the objective of the project was to improve the business environment, as measured by specific processes7 of facilitating investment. In that respect, the two PDOs were not very different and sought to achieve the same end goal8. The original PDO also mentions capacity building, but that goal was encompassed in the overarching objectives at both design and restructuring phases. 18. The PDO indicators were changed at restructuring: from the two initial PDO indicators, the team kept one, the cost to formally start a business as a percent of income per capita. The indicator on the turnover of the firms of the matching grant program that appeared as both outcome and intermediate indicator at approval, was only kept as intermediate at restructuring. 19. The restructuring paper indicates that the government used it to amend some of the M&E indicators and make them more clear, relevant, adequate, and measurable. Intermediate Outcome Indicators included adding (i) “Percentage of electronic transactions�; and (ii) “Average processing time for public services (in hours)�, and (iii) “Costs to user for public services (in US$)� for each of the following sub-components: business registration, land registration, and strengthening PSD policy and services. This allowed the project team to have more immediate measures such as cost and proportion of electronic transactions to work towards. 20. The intermediate outcomes at restructuring were (those introduced at restructuring are underlined): Improving economic governance, property rights institutions and business facilitation (i) Legislative backlog of laws impacting cost of doing business (ii) Time to settle commercial disputes (iii)Business registration - % of electronic transactions (iv) Business registration – average processing time (hours)9 (v) Business registration – costs to user10 (vi) Land registry - % of electronic transactions (vii) Land registry – average processing time (hours)11 7 In fact, the Amendment No 3 to the Financing Agreement, reflecting the restructuring changes, defines the revised project description as simply “The objective of the Project is to improve the ease of business processes in Malawi�. 8 Doing Business processes, elements, aspects and reforms should be interpreted as similar for this purpose. Although some of these may be seen as more outputs that outcomes, the ease of doing business indicators are measured by indicators like time, cost, and other features of the processes, not by welfare measures. Although different people will have different suggestions on the right word to use, we use here processes because this was recommended by OPCS for the PDO of a project going to Board in FY13 (MSME Development Project – P128443) 9 This is the same as the “Time to formally start a business�. It was just reworded and converted in hours. The DB reports in days. 10 No target was set for all the costs to users of these services. Basically, it was set that the prices would remain constant from restructuring to the end of the project. Nonetheless, ISRs kept mostly reporting on the main indicators set at the time of approval, as if restructuring had not introduced these new ones. 5 (viii) Land registry – costs to user Strengthening private sector development policy and services (ix) Number of achieved reforms on ease of Doing Business (x) Immigration’s business permits - % of electronic transactions (xi) Immigration’s business permits – average processing time12 (hours) (xii) Immigration’s business permits – costs to user Promoting access to finance and enhancing productivity of SMEs13 (xiii) Turnover in firms accessing the matching grant scheme (BUGS) versus control group 1.4 Main Beneficiaries 21. The PDO does not specify the main target group of the project, but it can be inferred to be private investors – both local and international. This includes potential investors that had not been operating in Malawi, but once conditions were to be improved could be attracted to the country, as well as existing investors. Through facilitation of investment and the access to finance component, the project could also benefit firms’ employees. In addition, given the considerable investment in capacity building, the main beneficiaries also included the government agencies and their officials that are trained or given better working conditions to do their job. 1.5 Original Components (as approved) 22. The project supported four components: a) Component One: Strengthening private property rights institutions and business facilitation; b) Component Two: Strengthening private sector development support institutions and services; c) Component Three: Promoting access to finance and productivity enhancement d) Component Four: Capacity building and implementation support Component One: Strengthening Private Property Rights Institutions and Business Facilitation (Total: US$4.3 million) 23. This component supported activities that aimed to improve the regulatory environment in which the private sector operates, to strengthen the institutions that 11 This is the same as the previous “Time to register real property�. It was just reworded and converted in hours. The DB reports in days. 12 This is not a complete new M&E indicator. At approval, the licenses being tracked were the (construction) permits monitored in the DB. This was an objective for the subcomponent 2.1 on MITC. At restructuring, it was changed to permits provided by the Department of Immigration to international investors for staying in Malawi – Business Residence Permits (BRP) and Temporary Employment Permit (TEP) – but time taking to provide these licenses was also monitored. 13 At restructuring, the indicator “Number of SME start-ups accessing the established SME fund� was dropped following the removal of the respective sub-component. 6 protect private property rights, and to allow speedy and low cost business facilitation. It included three sub-components: (i) streamlining the business regulatory environment, including addressing the large legislative backlog of economic laws and introducing key new legislation; (ii) providing support for strengthening and expanding the capacity of the newly established Commercial Division of the High Court to relieve the large backlog of commercial cases in Malawi’s court system and to strengthen contract enforcement; (iii) enabling the business and land registries to improve their effectiveness in registering businesses and facilitating the registration and securitization of land-based assets. Component Two: Strengthening Private Sector Development Support Institutions and Services (Total: US$4.6. million) 24. This component focused on building capacity of institutions that provide services to the private sector in order to improve the quality and volume of services delivered, and to strengthen institutions that provide policy direction on PSD. The proposed activities included (i) supporting the institutional capacity of the newly established Malawi Investment and Trade Centre, including setting up the legal and regulatory framework for a one-stop business licensing, investment and exporting promotion centre; (ii) establishing an institutional framework for Public Private Partnerships; (iii) building capacity in the Department of Private Sector Development of the MITPSD, including hiring of a small team dedicated to working on DB reforms; (iv) establishment of a public-private dialogue (PPD) secretariat in the Malawi Confederation of Chambers of Commerce and Industry (MCCCI) by funding two full time positions. Component Three: Promoting Access to Finance and Productivity Enhancement (Total: US$4.6 million) 25. The objective of this component was to support the growth and development of micro and small enterprises, into a “missing middle�, given few enterprises were located in between the many micro and informal sector businesses, and the larger multinationals and conglomerates. This component included (i) the establishment of a sustainable SME Investment Fund to increase access to finance for small-scale enterprises; (ii) the introduction of a matching grant scheme aimed at supporting business development services; and (iii) merger of two state-owned banks - the Malawi Rural Finance Company (MRFC) and the Malawi Savings Bank (MSB) to improve the service delivery of rural finance. Component Four: Capacity Building and Implementation Support (Total: US$2.9 million) 26. This component aimed at supporting the establishment of a Project Implementation Unit (PIU). This included supporting the staffing of the PIU and provision of operational costs and goods necessary for project implementation support. The PIU for the Privatization Commission (PC), which had been managing the WB- funded Privatization and Utility Reform Project (PURP), were assuming the PIU functions while the MITPSD built its capacity. The objective was to review MITPSD’s capacity after two years of implementation before deciding whether the project management functions could be mainstreamed into the MITPSD. 7 1.6 Revised Components 27. The four headline components remained unchanged throughout the duration of the project. There was one Level 1 restructuring process, where two subcomponents under Component 3 were discontinued, namely, the (i) planned establishment of a sustainable SME Investment Fund and (ii) the merger of Malawi Rural Finance Company (MRFC) with the Malawi Savings Bank (MSB). 28. In their place, the project reallocated the funds from the discontinued sub- components to two new activities: (i) upgrading the Department of Immigration’s capacity to deal with Business Residence Permits (BRP) and Temporary Employment Permits (TEP); and (ii) review and update of the Companies Act. The restructuring of the components was the result of changes in government policy. Both of the new activities (Immigration and Companies Act) were classified in the restructuring paper and in the Amendment to the Financing Agreement14 under Component 115, but under the M&E Framework of the restructuring paper the interventions related to the Immigration Department appear under Component 216. The component where the activities belong is important because the project received co-financing from the European Commission (EC), and this was initially thought to cover all components (as per the PAD), but was later – before the Level 1 restructuring17 - found out to be covering only Component 1. Although these new interventions were supposed to have been financed by the two dropped activities in Component 3 (hence through IDA funds), the over-commitment of other activities led to a practical problem. When the Immigration Department submitted Terms of Reference for its activities, there were insufficient IDA funds available. Although the improving of processes to obtain BRPs and TEPs was part of restructuring in late 2010, it was never implemented before the end of the project in 2012. 1.7 Other significant changes 29. Given the level 1 restructuring, it was agreed with the government to proactively address additional shortcomings. This led to: (i) deleting reference of the merger between the Malawi Investment Promotion Agency and the Malawi Export Promotion Council; (ii) amending M&E results; and (iii) allocating the unallocated funds in the project (US$2,005,500). These contingency funds set aside at approval were allocated for modernization/automation of business and land registries; staff costs in the PIU, including extending contracts beyond the initial two years of the project for the full operation; and operating expenses regarding office space. 30. In addition to the level 1 restructuring, there were two level 2 restructuring in 2011 and 2012 respectively. The first of these aimed to reallocate proceeds among the 14 April 28, 2011. The work with the Immigration Department appears jointly with improving processes for Business and Land Registration, which was part of Component 1. 15 Page 21 and pages 24/25 of the Restructuring Paper (September, 2010). 16 Pages 17 and 18 of the Restructuring Paper (September, 2010). 17 Page 10 of the Restructuring Paper (September, 2010). Also, 2008’s Administration Agreement. 8 relevant categories (goods, works, services, operating costs) following overruns in allocated disbursements in part due to the misunderstanding of where the EC funds could be used for. The second Level 2 restructuring in 2012 aimed at extending the closing date of the EC-funded Trust Fund (TF) from December 31, 2011 to June 30, 2012, as well as the Administration Agreement of the TF from June 30, 2012 to December 31, 2012. The reason for this extension was to allow for full disbursement of the contracts signed until December 31, 2011 that were still outstanding, as well as align project’s dates with the EC’s grace-period for disbursement18. 2. Key Factors Affecting Implementation and Outcomes 2.1 Project Preparation, Design and Quality at Entry Soundness of background analysis at design and government’s commitment 31. There is strong evidence in the project documents at design stage to demonstrate that it was informed by detailed background analysis. The project was prepared drawing upon the analytical work carried out by the World Bank Group and the GoM, including the 2004 DTIS, the 2006 ICA, the annual Doing Business surveys, the 2006 Malawi Country Assistance Evaluation (CAE), the 2006 Governance and Anticorruption Report (GAC), and the Portfolio Improvement Plan (PIP) prepared by the Country Management Unit jointly with the Quality Assurance Group (QAG). The PAD also mentions lessons incorporated from the experiences of other donor projects supporting the sector and from the experiences of MCCCI and National Action Group (NAG) in facilitating public- private dialogue in Malawi. 32. The lessons incorporated at project design included seeking to use a balanced implementation arrangement. That meant establishing a PIU instead of implementing directly through the Ministry. This aimed to achieve results and effective coordination while building project management capacity in the Ministry19. 33. At project design, the team discussed different alternative interventions outside the business environment topic. First, the team considered focusing the project on specific sectors such as agri-business, tourism or manufacturing, but this approach was rejected because of the apparent limited success of this type of projects in Malawi and other countries. Focusing on one area of intervention within business environment, and 18 The EC procedures allow for disbursement 6-month after the Completion date, while the Bank’s procedures have a grace period of 4 months. 19 The Bank and the MITPSD discussed at times during the project transferring PIU functions to the Ministry. The government stated they did not have the capacity to take functions of the PIU as officials who were earmarked and trained to take on project functions had been transferred to other Departments. Since the end of BESTAP, the MITPSD has started a project funded by the Africa Development Bank (AfDB) and is coordinating its implementation fully internally, benefitting from training provided through BESTAP. Team members of the MITPSD have informally referred to the difficulties in managing the new AfDB project at the same time as managing other activities of the respective Department. This speaks to the concern the Bank and the government had during BESTAP’s implementation. 9 making interventions more sequential, was also considered. However, it was decided that a combined set of interventions within business environment would generate greater impact. Given the limited knowledge about what would work best and the limited scope for very intensive reforms, this approach was likely appropriate. 34. The rationale for Bank’s intervention was its comparative advantage in PSD versus other donors, especially in business environment issues. This was widely recognized by both the government and other donors. With an in-country FPD team with a strong technical and diagnostic background, the World Bank Group was seen as the natural partner for this work. At the time, the IFC was not very active in Malawi and its expertise was only brought-in later, facilitated by Bank engagement through BESTAP. The EC agreed at design to co-finance BESTAP with over 20 percent co-financing. Assessment of the project design 35. The PDO set at design of supporting capacity development and investment climate reforms was aligned with the assessment made by the team of the needs of the country, reflected in the CAS medium-term outcomes (“improved climate for private business�). However, it was not the best approach to have explicitly linked it to “accelerate economic growth�, even if only mentioned as a second stage20, because of an issue of attribution. The two can go together with new investment generated by the reforms, but the spirit of the project’s interventions – and as measured by the M&E indicators21 – was to facilitate processes for new investors, not claim that economic growth in Malawi would be attributed to the project. 36. The PDO could have been set initially as: “The PDO is to improve Malawi’s investment climate, as measured by selected processes, and support the development of Micro, Small and Medium Enterprises (MSMEs). For the purposes of this project, investment climate reform comprises inter alia dealing with commercial disputes, improving the legislative framework, facilitating PPPs and PPD, and improving processes of firm registration, land registration, and business permits.� 37. A second area of discussion is the importance of having focused on business environment reforms instead of other areas of PSD. Given the private sector in Malawi was constrained by a large number of factors identified in preparation documents (access to finance, electricity and other infrastructure, skills development, investment climate, 20 The PDO objectives of the project are “improving the capacity development and investment climate reforms�. The PDO only mentions “accelerating economic growth� when saying that these objectives are “in order to� achieve that overarching goal. 21 The team monitored changes in the performance of businesses as part of its PDO indicators, but only for those that were supported directly by the project through the matching grant scheme. 10 lack of innovation, etc) it is debatable whether it was appropriate to concentrate the project22 on improving the business environment23, especially because it was not known whether that was the most binding constraint. At this stage, it is difficult to tell whether or not it was the right approach, but, as discussed above, the project team considered seriously other alternatives and concluded that it would be more productive – mostly due to the high risks of the alternative interventions - to focus on business environment. One needs to acknowledge the team’s consistency in focusing on one area of comparative advantage and seeking to contribute to solve that problem24. 38. Within the investment climate area of work, the project aimed to improve processes related to property rights but did not tackle the bigger and more critical issue (naturally more difficult): the land titling and the provision of property rights25. However, the decision to take a broad approach and address sub-areas of investment climate where the Bank could be effective and where results could be achieved in a relatively short period of time was sensible. 39. Component 2 could have been integrated in Component 1. By integrating these capacity building programs under the initiatives of Component 1, the project design would have been simpler allowing for easier coordination. With the benefit of hindsight, this would have avoided as well problems with co-financing restrictions. The co- financing from the EC was later26 restricted to Component 1 due to the thematic focus of the source of funds (governance), which did not allow some of the counterparts (in Component 2) to benefit from those funds. 40. This separation of the first two Components involuntary masked another important issue at project design: the project combined a large number of counterparts, located in the two major cities, Lilongwe and Blantyre. The list of organizations involved included: (i) Ministry of Industry, Trade and Private Sector Development (Lilongwe) (ii) Ministry of Justice (Lilongwe) (iii) Commercial Division of the High Court (Blantyre / Lilongwe) (iv) Department of Registrar’s General (Blantyre) (v) Department of Lands - Registry (Lilongwe) (vi) Malawi Investment and Trade Centre (Lilongwe) (vii) Ministry of Finance, PPPs (Blantyre / Lilongwe) 22 This was the only FPD project in Malawi. 23 The project actually incorporated a component on direct support to SME development, but only one of the three sub-components was implemented. The strategy was clearly to target investment climate reforms. 24 In many countries in Africa, this type of work is led by the IFC Advisory Services. It is not clear beyond lack of a full-time person in Malawi why the IFC was not working in this country as much as in others in Africa. The BESTAP facilitated during implementation phase IFC’s involvement in the project. 25 Other projects were operating in the lands space – IRLAD – but mostly focused on improving farmers’ productivity. 26 In essence, it was always restricted to Component 1 but the team at design thought it would be available for all components, as per the PAD. 11 (viii) Malawian Confederation of Chambers of Commerce and Industry (Blantyre) (ix) Matching grant program, BUGS (Blantyre27) (x) Reserve Bank of Malawi28 (Lilongwe) 41. Coordination of a high number of organizations with uneven levels of capacity was a risk that was mitigated well by the PIU’s through effective management. 42. Differences in capacity amongst beneficiary agencies were important in terms of knowledge of how to benefit from the project and work under WB procedures29. This became an issue as the use of the funds became to some extent a horse race within components30. Backed by the Bank’s institutional support31 - not linked per se to the team - of keeping the focus on disbursement 32 , organizations sought to implement their activities within a short period of time. When contracts were more expensive than budgeted in the procurement plan, it became difficult to guarantee that the missing activities would have sufficient funds to complete the work. This led to shortage of funds for the lagging implementers in the process. 43. The location of beneficiaries in two cities is related to a divide of the Malawian government between Lilongwe and Blantyre rather than to project design. With more than one counterpart, there would always be a high probability of working in the two cities, but the problem was that the high number of involved organizations exacerbated the number of trips between the two cities and to other neutral locations, which led to high costs of operations, coordination, and training. Assessment of risks 44. The main risks identified at design included (i) political risk 33 , (ii) macroeconomic risk, (iii) staff turnover in government, and (iv) governance / financial management risk. These risks were well considered at design and mitigation measures were included. The project did not consider at design the risk of the high cost of coordinating a large number of agencies. 27 At the middle of the project, it opened a branch in Lilongwe, but its headquarters were in Blantyre. 28 This was identified at design to be focal point for the merger between MSB and MRFC, but in practice this work was concentrated at the Ministry of Finance. 29 The Steering Committee (where organizations were members) helped through time to bring implementers up to speed on Bank rules and procedures. 30 The ultimate example of this was a subcomponent added at restructuring and never completed because of lack of funds when it wanted to execute its activities (see Annex 1). 31 See for instance dashboard reports from Regional offices where targets of disbursement per year are set at the individual project level. 32 This is based on an assumption that every activity within project design is (i) effective, limiting scope for changes along the way, and (ii) is more effective if spent in a concentrated period of time. 33 The political support to donor activities in general seemed to have deteriorated somehow in the last two years of President Mutharika’s tenure (2010-2012). After his death, the new President brought-in a different style within the same political orientation of focusing on business environment reforms. 12 45. Given a number of interventions included capacity building initiatives of government officials, including Master’s programs and long placements to counterpart departments in other countries, the risk of training someone and losing her/him after was high. This risk was fully considered at design, but the only mitigation measure of involving the government during project design was a weak one. This risk of staff turnover was particularly high for specific posts such as judicial officers in the High Court. This risk materialized in practice during the project with the loss of trained personnel to other ranks of the justice34. Additionally, members of the PIU including fiduciary staff had to be replaced during implementation because they took other employment opportunities35. 2.2 Implementation 46. Implementation was initially constrained by two factors: (i) unfamiliarity with Bank procedures by some of the involved organizations because it was the first time they were implementing a Bank project, and (ii) resistance to change in specific departments. The quality of the initial ToRs was often weak, processes were not fully understood, there was mistrust of the program and its objectives, mid-level officials within agencies were learning about the implications for them of the change - the speed of implementation of the different sub-components varied. This and changes in the Bank’s TTL delayed implementation for the initial phase of the project, but there was a catch-up a couple of years into the project. 47. The largest issue during project implementation was the delay in securing the co- financing from the EC, which only became available in mid-2011, as well as the misunderstanding on how those funds could be used. The co-financing was led by the Access to Governance and Justice Unit of the EC. At design, the Bank team believed and made explicit in the PAD that the EC funds (equivalent to US$3.7 million) would cover all components and would be proportionally allocated between activities. However, the Administration Agreement signed between the Bank and the EU in 2008 showed that the EC funds would only be able to support Component 1, because the funds were only allowed to focus on access to justice. The grant agreement between the Bank and government, which was delayed 36 , and only countersigned in 2011, confirmed that allocation, which had been also reflected in the 2010 restructuring paper. This limit on activities that could be supported by the EC TF is argued by the government and the Bank team to have created an implementation problem because of the funds already 34 These officials were seen to have been recruited by other offices specifically because they received this training supported by BESTAP. But, one could argue that as long as these officials remain in the judiciary, including courts at the regional level, the capacity building exercise is benefiting the country. Although that is a fair consideration, the reality is that they were trained to improve the systems in specific departments. Once the trained official is taken away, the institutions often do not keep that knowledge. 35 The initial PIU Procurement Specialist joined the Bank, the PIU Financial Management Specialist joined the AfDB. 36 This delay resulted from the understanding that the IDA funds could be used first and then followed by the EC funds, as well as bureaucracy issues in getting the documents ready and signed by the recipient. 13 committed to other components thinking the combined envelope would be larger. But if this restriction was known since 2008, this over-commitment in certain subcomponents should not have happened, suggesting that there was a planning problem of not having incorporated that information in the revisions of the procurement plan. 48. The project’s implementation and supervision followed normal standards of Bank procedures. This included a mid-term review in October 2010, which led to the restructuring of the project in November 2010 following changes in government policy. At restructuring, the team believed the revised project would strengthen the legal and regulatory environment for promoting and supporting foreign and domestic investment in the country, as well as build capacity within government agencies and institutions. The restructured components were seen to better support the successful achievement of the (revised) PDO. Indeed, the restructuring focused on cutting resources from activities that had no progress and add to others with more potential. In that sense, it increased the likelihood of achieving the PDO. 49. Progress in implementation was affected during the period 2010-12 by the deteriorating political and macroeconomic conditions in the country, especially the shortage of fuel and foreign currency, and the rising inflation. Aid to Malawi in general was reduced due to political problems 37 , which affected the implementation of complementary interventions. The continuous social unrest increased the risk of the reforms being sufficient to generate new investment. 2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization 50. The initial PDO was to support capacity development and investment climate reforms in order to accelerate economic growth. The M&E framework rightly did not focus on the broad last part of this objective but rather just on more specific investment climate indicators. 51. The PDO indicator 38 throughout the project was the cost to formally start a business as a % of income per capita. This indicator was designed to be easy to monitor under the Doing Business Survey, and could be partially attributed to the project reforms. However, the choice of this indicator was too narrow, reflecting only one element of reform39, while the project sought to address a number of issues. In addition, the problem with this indicator is that the cost itself is only marginally correlated with the size of the economy; hence a change in the GDP could affect the ratio during the time of the project 37 http://www.bbc.co.uk/news/world-africa-15206437 38 At design, there was another indicator – turnover of firms supported by the matching grant program versus a control group – that was both a PDO and an Intermediate indicator. In ISRs, it was just reported as an intermediate indicator. In the restructuring paper, it became only an intermediate indicator. 39 At most, this indicator covers some 15 percent to 20 percent of the non-PIU activities proposed at design, if one considers the partial involvements of Components 1.1, 1.3, 2.1, and 2.3 towards achieving this outcome. 14 even without any reform. This indicator was hence wrongly designed as the PDO indicator and should have been replaced during the life of the project. 52. The investment climate intermediate indicators set at design were thought to be easily trackable: legislative backlog, time to settle commercial disputes, time to register a business, time to register property, time to obtain (export) licenses. Most of these were drawn from the yearly published DB reports. Notably absent was an intermediate indicator on the capacity building front, which was an aspect referenced in the PDO, and seen as important in generating additional investment climate reforms. Intermediate indicators on capacity building could have included the number of officials trained in the departments, specific intensities of capacity building efforts, and mechanisms of sustaining institutional knowledge. For a couple of cases of the intermediate indicators, the design did not fully incorporate the risk of attribution to the project. This was especially the case for the time to settle a commercial dispute. The team tracked the DB reports but the reforms in establishing the High Court had no direct influence on the results of that indicator, which measures changes in the Magistrate’s Court. The intervention was concentrated on the new Commercial Division of the High Court. Hence, although creating and supporting the High Court helped achieve the de facto objective of the project, it is not reflected in the DB ranking. 53. When using a proxy PDO – as described in section 2.1 - and when the PDO indicators are incorrectly designed, the ICR guidelines recommend using proxy outcome indicators to better evaluate progress towards achieving the PDO. Both for design and restructuring, the following would have been better PDO indicators to measure the impacts of the project: 1. Time to settle commercial disputes above MWK 1 million 2. Time to formally start a business40 3. Time to register real property 4. Number of enacted laws up to international standards originally drafted by the capacity provided in Component 2 (MITPSD, PPP, PPD) 5. Turnover in firms accessing the matching grant scheme (BUGS) versus control group 54. These indicators were actually mostly listed as intermediate indicators in the project (some not exactly as stated here), but could have been identified as the objective indicators of BESTAP. 55. The intermediate outputs could have included the computerization of systems, the achievements in terms of capacity building initiatives, the size of the teams seconded in the Ministries to help coordinate drafting of law reforms, the number of meetings between public and private sector, and the number of firms with access to finance solutions. 40 Cost to start a business in real terms could also be used, but suggesting time to be consistent with the time to register property. 15 56. The M&E implementation was supervised by a Specialist in the PIU. As discussed, this Specialist had to engage with each agency to obtain updated information on the progress of the subcomponent. The quality of the M&E reporting was high and facilitated supervision (as demonstrated in the Implementation Status Reports - ISRs). 57. No impact evaluation was incorporated at design of the project, though it was very rare at the time for a FPD project in Africa to have one. However, in early 2010, the BESTAP PIU, other members of the government, and the TTL participated in the DIME- FPD Impact Evaluation workshop in Dakar. In this seminar with participation of various project teams and World Bank researchers, the Malawi team learned about methodologies of conducting rigorous impact evaluations and worked with a researcher to design an impact evaluation for BESTAP. It was first considered to conduct an impact evaluation of the matching grant program in Component 3, which was critical in terms of SME development activities, seemed amenable for impact evaluation work, and there was significant debate on whether it was working or not. After a follow-up mission to Malawi, the research and government team realized that the adequate conditions did not exist to implement the impact evaluation of the matching grant program. The main reasons behind the difficulty in implementing this study are summarized in Annex 3, which explains that the problems were also common to other countries where similar studies were attempted. Given the problems in implementing this evaluation, the project team hired a consulting firm to conduct a process assessment of BUGS as well as a feedback survey on participants. The results are discussed in section 3.3 and more detailed in Annex 5. 58. In terms of impact evaluation, it was decided with the government in mid-2010 to focus on an impact evaluation of the business registration reforms, given that under BESTAP the government was streamlining its registration process. The impact evaluation was designed to assess the effects of firms becoming formal in terms of access to financial services, markets, and performance41. 2.4 Safeguard and Fiduciary Compliance Safeguards 59. The project was classified as a Category C, and no safeguard policies were triggered. Financial Management 60. The PIU kept good records of financial management. The project complied with the financial covenants stated in the financing agreement. The audits were made on time and followed Bank rules. All audits were unqualified42. Budgeting and procurement plans were of good quality and prepared in a timely fashion. The initial FM team had worked in 41 Expected benefits of a business registration are access to formal banking services, credit, government procurement and matching grants, and less harassment from government officials. 42 Source: FM Specialist, World Bank. 16 the previous FPD project – PURP – and knew the Bank procedures. However, the FM Specialist left in the middle of the project to work for AfDB, but the systems were kept in place. The overall performance of the financial management function is rated Satisfactory. Procurement 61. The procurement function was overseen by a consultant at the PIU, which had also been working with PURP. The first procurement specialist left in 2010 to work for the World Bank local office. During a period of over six months the Bank’s new procurement specialist (previously from BESTAP’s PIU) sent BESTAP requests for review to a colleague in the Zambia office43. The PIU ensured implementing agencies followed the agreed procurement guidelines and procedures. The overall performance of procurement is rated Satisfactory. 2.5 Post-completion Operation/Next Phase 62. Following the closure of the project, the government and other partners have continued to operate the main institutions created or developed under the auspices of BESTAP. Specifically, the Commercial Division of the High-Court is financed out of the state budget and carries the work that was successfully developed under BESTAP. The work on drafting laws to improve the doing business indicators has continued post- completion with a person attached to the MITPSD. The PPD Secretariat, which sits in the MCCCI, reduced the number of staff from two to one person after BESTAP’s closure but has continued with Chamber’s funds to promote dialogue between the government and the private enterprises (a high-level meeting took place in February 2013). The ICT systems that have been introduced continue to be used by the respective Departments. 63. Furthermore, the Bank and other donors have continued to work on follow-up activities within the business environment space. In 2011, the Bank approved the Financial Strengthening Technical Assistance Project (FSTAP), which tackles a complementary area of work – the financial sector – and can through regulatory reforms support the business climate. 64. Since 2011, the GoM expressed to the donor community interest in continuing to implement investment climate reforms post completion of BESTAP. After subsequent discussions led by the Bank with other development partners in the context of leveraging resources, the UK’s Department for International Development (Dfid) agreed to respond to the request received from the GoM to provide assistance. The Bank has been providing technical support in the preparation of this project, which aims to improve the business enabling environment in Malawi. The design focuses on the continuation of the work in BESTAP. 43 This to avoid any potential conflict of interest. 17 65. Additionally, following the closure of BESTAP, the new government has finalized and approved a set of additional laws and regulations that had been prepared under the auspices of the project. 3. Assessment of Outcomes 3.1 Relevance of Objectives, Design and Implementation Rating of Relevance - Original: Substantial Rating of Relevance - Revised: High 66. Given the Level 1 restructuring in 2010, we evaluate the relevance of Objectives, Design and Implementation splitting between the initial and the restructured project. The relevance rating is Substantial for the project at approval and High for the revised project. 67. The Substantial rating for relevance at approval reflects the High rating for relevance of objectives and Substantial rating for relevance of design and relevance of implementation. The High rating for relevance post-restructuring reflects the High rating for relevance of objectives and relevance of design and Substantial rating for relevance of implementation. Relevance of Original Objectives: High Relevance of Revised Objectives: High 68. As discussed above, the PDO of supporting capacity development and investment climate reforms in order to accelerate economic growth has some elements that are difficult to attribute to the project. The same shortcoming applies to the revised PDO, which aims to improve the ease of doing business processes in order to increase foreign and domestic private sector investment in the economy44. But, as explained in the Bank’s OPCS ICR guidelines45, “whenever the PDOs stated in project documents are so broad and/or vaguely worded as to preclude any meaningful evaluation, intended project objectives are inferred by the evaluator from key associated outcome targets (and/or project design features as relevant)�. Based on the discussion in Section 2.1, a more appropriate PDO to both the objectives at design and at restructuring would be: “The PDO is to improve Malawi’s investment climate, as measured by selected processes, and support the development of Micro, Small and Medium Enterprises (MSMEs). For the purposes of this project, investment climate reform comprises inter alia dealing with commercial disputes, improving the legislative framework, 44 A number of factors impact investment and only at the margin can the project contribute to those. In this case the revised Financing Agreement (2011) dropped the mention to investment focusing solely on business processes. 45 Appendix B of the ICR Guidelines. 18 facilitating PPPs and PPD, and improving processes of firm registration, land registration, and business permits.� 69. Against this backdrop, the objectives of the project remain relevant to Malawi today 46 because they are aligned with the country’s current development priorities as reflected in the recently approved CAS (FY13-FY16), which in turn is aligned with the Second Malawi Growth and Development Strategy (MGDS II) adopted in April 2012 and the subsequent Economic Recovery Plan launched in October 2012. It is also aligned with the National Export Promotion Strategy (NES). The CAS prioritizes Bank’s support around three themes: (1) Promoting Sustainable, Diversified and Inclusive Growth; (2) Enhancing Human Capital and Reducing Vulnerabilities; and (3) Mainstreaming Governance for Enhanced Development Effectiveness. The three priority themes of the CAS have been structured around six results areas, including Result Area 2: A business environment that promotes competitiveness and enhances productivity. 70. Furthermore, the investment climate reforms are at the forefront of the dialogue with the government. This has been further enhanced since the new government - which has brought a new reformist agenda - has been in power. As discussed above, the Bank has worked with the government and Dfid in preparing a new program of activities on business environment. The government organized a High Level Forum on Doing Business Reforms47 in March 2013. This Forum had the objective of assessing progress on implementing DB reforms in Malawi, identifying priority measures to be implemented by responsible ministries/agencies, identifying challenges in formulation of reform proposals, putting in place mechanisms for adequate support at political and technical level, and identifying ways of enhancing support from development partners. 71. MSME development – an element in the proxy PDO for this ICR - remains very relevant in the context of the development priorities for the government and the international community. Enhancing productivity is in one of the Results Areas of the new CAS. The AfDB approved in 2012 a US$18 million project with the MITPSD in order to improve the competitiveness of the private sector. This project complements the work on investment climate by directly targeting MSME development initiatives. The AfDB project also continues the firm-level work (through a different approach) started under the matching grant program of BESTAP. More details about this project are listed in Annex 10. 72. For these reasons, there are strong indications to suggest the high relevance of the project’s objectives, both originally and post-restructuring. 46 Furthermore, as discussed above, the project objectives were also aligned with the CAS and government strategy (MGDS) relevant in 2007 and guiding the strategy at the time of the restructuring. 47 More than 150 senior public and private sector CEOs and their representatives, as well as key members of the donor community attended the meeting. The meeting was chaired by the Minister of Commerce and Industry of Malawi. 19 Relevance of Original Design: Substantial Relevance of Revised Design: High 73. The original design of the investment climate reforms are aligned with the project objectives and through that aligned with the most recent CAS, MGDS, etc. The post- restructuring design in terms of business environment is also very aligned with the objectives of the project. The two new activities added at restructuring – the upgrading of the capacity to deal with BRPs and TEPs; and review and update of the Companies Act – are relevant components of the investment climate reform agenda. 74. Before and after restructuring, the interventions were concentrated on core components of improving the investment climate, as measured by the outcomes of interest of the project. For example, one of the objectives was to reduce the time to start a business. This aspect involves different processes in different agencies as outlined in Annex 11. The desired change may not occur if the project is not working directly or indirectly with the agencies that are able to provide the largest change in the outcome of interest48 (those associated with the largest columns in Figure 12 of Annex 11). In this case, BESTAP in practice worked on 5 of these 10 steps (see Annex 11). 75. The design of the components is also relevant in the context of the specific priorities of the government today. The planned Dfid/GoM project that the Bank is helping design in 2013 largely focuses on the continuation of BESTAP’s work: (a) improving ease of starting a business; (b) improving ease of dealing with construction permits; (c) improving ease of trading across borders; (d) improving ease of enforcing contracts; and (e) improving ease of resolving insolvency. 76. The design of Component 3 was originally weak. At design stage, there were already discussions about the lack of government’s commitment towards merging the MRFC and the MSB49. Also, the initial project design did not allocate enough resources to establish the SME Investment Fund, suggesting ownership of this subcomponent was low. None of these activities took place. While for the banks, the government is now considering internal reforms in these institutions – continuing the spirit of change of the project – for the SME Investment Fund, the government has changed its priority, looking towards creating a Development Bank instead. 77. Given the soundness of design of Components 1 and 2 and the weakness in initial design of Component 3, the rating of relevance of original design is Substantial. As the activities from Component 3 were changed to interventions relevant to the PDO and still relevant today, the rating of relevance of design post-restructuring is High. 48 In addition to the Department of Registrar’s General (DRG), it is necessary to tackle issues related to the time taking to start a business in City Assembly licenses and the Ministry of Labor. The project worked on legislative reforms for the City Assemblies. The new business license is waiting for Parliament approval and hence its effects are not yet reflected in the DB rankings. 49 According to interviews with people involved in the process, this was mostly an idea driven by the then Minister of Finance rather than by the project team or the technical teams in the Ministry. 20 Relevance of Implementation against original PDO: Substantial Relevance of Implementation against revised PDO: Substantial 78. Implementation of BESTAP was broadly consistent with the activities designed at approval, as well as with those approved at restructuring. The operation remained important throughout implementation to the country and Bank development objectives, which have not materially changed throughout this process. The activities were implemented responding to the proxy PDO, which is aligned with the current priorities, suggesting relevance of Implementation. 79. Policy changes were incorporated through a restructuring of the project in November 2010 that dropped non-performing subcomponents and added new activities with apparent potential. This restructuring could have happened earlier, because as early as January 2009 the government informed the Bank of the ongoing policy changes50. Given the project was downgraded at that time to Moderately Unsatisfactory due to these underperforming components, one could have expected a quicker turnaround, which did not happen51. 80. Furthermore, certain issues during implementation led to redirecting of funds within the project from the design stage to its implementation. These included among others (i) the delay in securing the co-financing from the EC, as well as the misunderstanding on how these funds could be used; (ii) the acceleration of early disbursement by some of the involved organizations knowing that otherwise some other subcomponents could potentially be competing for same funds52; (iii) the need to keep the PIU throughout the duration of the project, when it was only initially thought that two years would be sufficient. These issues did not materially change the relevance of the activities implemented; which were kept aligned with the PDO. It meant though that some other activities – including a core activity with the Department of Immigration added at restructuring - were not implemented, losing some relevance. 81. For these reasons, the relevance of implementation remained Substantial both against the original PDO and the revised one. 50 Delays in getting a formal request from the government help explain the delay of the process. 51 One potential explanation for this delay is that the low rating only lasted for 6 months and hence the pressure to change was not as intense as it could have been. 52 This is based on the rate of disbursement of funds and anecdotal evidence resulting from interviews with involved organizations. 21 3.2 Achievement of Project Development Objectives Rating of Efficacy - Original: Substantial Rating of Efficacy - Revised: Substantial 82. The achievement of the PDO is rated Substantial against both the original PDO and the revised PDO. The components’ achievements are discussed at length in Annex 2. 83. The PDO indicator before and after restructuring - Cost to formally start a business as a % of GNI per capita – was largely achieved (see Figure 1). Figure 1: Cost to formally start a business as a % of GNI per capita (comparison with BESTAP target) 245 215.7 210 188.7 175 140 125.9 108 108.4 105 90.9 83.7 Target 70 35 0 DB2007 DB2008 DB2009 DB2010 DB2011 DB2012 DB2013 Source: Adapted from DB reports. BESTAP target is for DB2012. 84. However, as discussed in section 2.3 above, this indicator was incorrectly designed – and incorrectly kept as the single PDO indicator throughout the project - because it can only be partly attributed to the project and only reflects one section53 of the reforms being supported. 85. Hence, it is appropriate to introduce proxy PDO indicators: (i) time to settle commercial disputes above MWK 1 million; (ii) time to formally start a business; (iii) time to register real property; (iv) number of enacted laws up to international standards originally drafted by the capacity provided in Component 2 (MITPSD, PPP, PPD); and (v) turnover in firms accessing the matching grant scheme (BUGS) versus control group. All these five indicators are core in achieving the proxy PDO of improved investment 53 At most, some 15 percent to 20 percent of the non-PIU activities proposed at design, if one considers the partial involvements of Components 1.1, 1.3, 2.1, and 2.3. 22 climate, as measured by selected processes, and in developing the MSME sector54. These indicators and proxy PDO are relevant at design and post-restructuring. 86. Out of these five proxy PDO indicators, the project largely achieved the set targets on (i), (iii) and (iv), and was not successful on (ii). Some of the effects of the program have yet to materialize including those related to the recently enacted laws, namely on starting a business. 87. In the matching grant program – indicator (v) - there are indications of a successful implementation, and positive feedback from the firms supported (see Annex 5), but in general it is not known if the program was able to improve the turnover of the targeted firms versus a valid control group55 (see Annex 3 for more details on attempt to conduct impact evaluation). In the absence of a rigorous impact evaluation, the team conducted a customer satisfaction survey. Based on this assessment, 68 percent of the firms considered the matching grant helped them improve skills, 47 percent of them to improve product quality, and 44 percent to increase sales. During the period of the project, beneficiary firms grew in employment (31 percent), investments (109 percent) and turnover (62 percent). These changes cannot be attributed to the BUGS support directly, but the general perception of the firms’ beneficiaries was positive. In any case, there is insufficient information on whether or not indicator (v) was achieved. 88. In this scenario, where three PDO indicators were successful, one is unsuccessful, and one cannot be attributed, this suggests a Substantial rating for achievement of the proxy PDO. Given these indicators are the most relevant both before and after the Level 1 restructuring, a Substantial rating is appropriate for achievement of the proxy PDO both for approval and against the revision of the project. 89. In more detail, the following are the main achievements of the project: (i) Financing, training, and placing the team that helped draft 24 core regulations and laws according to international standards to promote private sector investment. These laws were either enacted or just before Parliament. These laws included the Investment and Export Promotion Bill, Credit Reference Bureau Bill, Personal Property Security Bill, Business Registration Bill, Companies Act, and Business Licensing Bill. Additionally, this team developed a pipeline of an additional 22 laws in very advanced stages of preparation. The list of laws is detailed in Annex 2, as well as details on their consulting process of preparation. (ii) Putting in place in Blantyre and Lilongwe an operating Commercial Division of the High Court helping to address disputes of commercial matters above MWK 1 54 With equal weights, this second area is then less important than business environment because only one indicator is used versus four in investment climate. Given the name of the project and the importance given to investment climate, that is not an unfair assumption. 55 The government reaches the same conclusion in its ICR (see Annex 7). 23 million56. Over 1,000 cases have been judged by the Commercial Court since 2007 with an average of 96 calendar days per case 57 . This metric compares with 432 calendar days for the traditional Magistrate’s Court in judging commercial disputes. (iii) Establishing an operating IT system to register land. The days to register property have decreased from 118 in 2007 to 69 days in 2012. (iv) Establishing an Institutional Framework for Public Private Partnerships (PPPs). This included enacting a new PPP bill among other initiatives to strengthen the institutional support in this area. (v) Improving awareness of the matching grant program (BUGS) in Malawi and finishing the project with 1,130 cost share agreements and a number of firms assisted. (vi) Improving the institutional capacity of the MITPSD in implementing programs, including a core group of officials with solid competences for continuing the support to business environment reforms. (vii) Creating institutional knowledge in the MITPSD on the usage of counterfactual analysis in informing results-based policy making. 90. On the other hand, these were certain obstacles to better outcomes: (i) Limited progress in establishing an operating One-Stop-Shop (OSS) on Investment and Trade (MITC). The MITC bill was only passed in 2012. The OSS is still very incipient and no relevant progress was achieved on business licenses. (ii) In spite of the investment in IT, capacity building, and changes of processes, which resulted in having in place a new administrative system, the business registration is still done manually as of the time of the ICR assessment58. The system has been tested and is ready, but has not yet been launched. Furthermore, the delays in 2010-12 in enacting some of the regulations have limited the change in the main indicators of progress on business registration. Going forward, the legislative work funded by BESTAP on removing the need of a company seal, reducing the time it takes to get a business license of City Assembly from 29 to 7 days (Business License Bill), and reducing the requirement of inspections in every firm, should allow for improving the time to start a business target. (iii) The project supported the creation of an operating Public Private Dialogue (PPD) discussing issues such as tax reforms, inter-bank interest rates, labor reforms, 56 Approximately US$2,700 at current exchange rate, given the recent depreciation of the Malawian Kwacha. 18-24 months ago it was equivalent to approximately US$5,000. 57 Source: PIU 58 A dispute on the final training program – doing it in Malawi or outside of the country – is hindering the launch of the automation system that is already tested and in place. 24 electricity problems, water supply, and exports. This was useful for establishing the debate, but the capacity of the platform to generate change was limited. At best, it speeded up ongoing reforms. One of the issues seemed to be that the officials including some Ministers present in the PPD did not have enough power to generate change 59 . Following the conclusion of the project, the PPD platform is still operational with the support of the Chamber of Commerce. (iv) Two subcomponents of the project – establishment of a SME Investment Fund and merger of two banks – were dropped less than two years before closing the project. The policy change60 does not fully explain the problem here. The failure of these activities is related to project design, which may not necessarily be attributed to government or Bank shortcomings61. (v) One of the activities added at restructuring – support to the Immigration Department in preparing licenses for investors – never took off because of limited funds for implementation, suggesting inadequate preparation of the restructuring process. (vi) Issues in setting up the co-financing process have confused implementation and hindered the achievement of results – like improvement in DB ratings – during the duration of the project. (vii) The relative importance of institutional capacity building likely hindered the possibility of addressing earlier the immediate outcome measures of the project. 3.3 Efficiency Rating of Efficiency - Original: Modest Rating of Efficiency - Revised: Modest 91. The PAD presented in its Annex 9 an economic analysis for every subcomponent of the project. The analysis at approval suggested a NPV of US$20 million and an ERR of 49 percent. In spite of efforts with the original TTL of the project, it was not possible to recover for this evaluation the files that helped construct the cost-benefit analysis at design. Given the nature of the majority of the interventions – legislative reforms, new processes, new IT systems, capacity building – the figures presented at design were purely speculative. A NPV or ERR should not have been used to assess whether to go 59 Anecdotally, there were indications of power concentrated in the Presidency. 60 In the case of the SME Investment Fund, the government decided to pursue a development bank rather than a venture capital fund. This would imply for substantial seed capital and a different institutional set up that BESTAP was not ready to provide. In the case of the merger between MSB and MRFC, the government decided to keep the entities separate and pursue reforms internally. 61 In the case of the merger between MSB and MRFC, interviews for this evaluation seem to suggest that there was significant pressure from the then Minister of Finance in including this activity under the project. In the larger picture of things, some of the decisions on project structure often include accepting activities that the team is not fully comfortable with (especially when there is no clear acceptance that it will also go wrong), but that are needed to move other core set of activities forward. 25 ahead with interventions such as those in Component 1 and 2. Therefore, for the purpose of this ICR, a new NPV / ERR model for assessing the efficacy of these interventions was not applied. 92. The only component where it would be appropriate to conduct the economic analysis through a NPV approach is Component 3. Here, two subcomponents were dropped during implementation, while the matching grant program (BUGS) remained. It would have been appropriate to include a rigorous impact evaluation of this subcomponent comparing the actual effects with the economic analysis. That was attempted in 2010, but a set of problems including low take-up at that time of the impact evaluation (25 percent of the funds had been disbursed) and lack of commitment from the implementing agency limited the possibility of incorporating an impact evaluation. Nonetheless, a study was commissioned by the project at its end to conduct a review of BUGS. That assessment is summarized in Annex 5, showing that the matching grant program became gradually accepted by MSMEs in Malawi (in the end over 1,100 cost- sharing grants were awarded). Given the limitations at that stage in attributing causality, the study focused on a customer satisfaction survey. Based on this self-assessment, 68 percent of the firms considered the matching grant helped them improve skills, 47 percent of them to improve product quality, and 44 percent to increase sales. However, only 15 percent believed that BUGS had directly increased the income and employment levels of their businesses, while 28 percent thought there was no tangible impact yet, but felt more confident that things would improve. 93. The interventions in Component 1 and 2 are grounded on the assumption that they help facilitate local and international investment. For instance, with a better system of enforcing contracts, investors would at the margin be more attracted to invest in Malawi vis-à-vis other countries that do not have the same conditions. The same holds true if the legislation for developing businesses is transparent and clear. However, the actual investments are not solely dependent on these parameters, but could also be driven by good economic opportunities and other factors. In that vein, the important efficiency questions for this type of interventions are (i) whether there are more productive alternatives within the same scope of doing business activities and (ii) whether implementation was done at the least possible cost. 94. On the first question, we can take the example of the business registration reforms under the project – a new registration bill, digitization of existing records, and computerization of the application process – which are linked to the stated PDO indicator. After visiting the Department of Registrar’s General (DRG) offices in Blantyre, it is easy to argue that these reforms are necessary. But, the question is whether they make a difference in terms of generating new registrations and making an impact in SME development through formalization. The numbers from other countries suggest that sometimes – not always - improving processes generate increased demand for business registration, especially driven by new entrants62. In Malawi, the impact evaluation done 62 Bruhn, 2013. http://blogs.worldbank.org/allaboutfinance/one-stop-shops-do-they-or-don-t-they-increase- business-registration 26 in the auspices of BESTAP tested whether making it extremely easy to register can make a difference. Indeed, if registration process was totally costless to the firm, a high proportion of firms would accept registration (approximately 75 percent). In terms of the effects on firms’ development of becoming formal, it seems that these impacts in the short run are small. There are indications of better outcomes when combining formalization with access to business bank accounts (see more on this in Annex 3). 95. On the second question, a comparison of the cost of the activities with other countries conducting similar reforms seems appropriate. In the case of new IT systems for business and land registration for instance, it seems that the costs were in line with other countries doing similar investments63. 96. A least cost approach analysis of the commercial court may suggest though that other options could have been cheaper 64 , but these at the risk of not achieving the objective. Given that there was a decision to fully fund a new structure under the High- Court rather than making more judges available in the old Magistrate’s Court for commercial cases only, it is questionable that this was the cheaper option from a pure efficiency point of view. If the Magistrate’s court were to place three judges and 30 of their (new) staff solely working on Commercial Cases, it is possible that the same results – 96 days on average to solve commercial disputes - could have been achieved. However, this higher cost that the project incurred seems to be justified on the risks that the efficient approach would have had a higher variance of results: a non-negligible chance of not performing. A new agency with separate space reduced the political-economy institutional risks associated with implementation. 97. A critical aspect that is important to consider in this assessment is the cost of implementation and coordination. During BESTAP’s implementation, the risk of having a high number of implementing agencies was somehow controlled because the PIU was effective in supervising the activities of the different organizations. However, this came at a cost to the project. The PIU and M&E (Component 4), which had been budgeted at approval and restructuring to cost 15 percent65 and 19 percent of the envelope, ended up representing 27 percent of the project expenses, 35 percent of IDA funds. By restructuring, Component 4 represented 30 percent of the disbursements until then. 98. Component 4’s expenses comprised US$2 million for fees to project staff, US$1.2 million for operating costs, US$625 thousand for equipment, US$600 thousand for PIU 63 According to Investment Climate IT Specialist at the WBG - Mozambique: Information System with no customized development, US$130 thousand; Liberia: Business registration system, US$1.5 million; Rwanda, Business registration system, collateral registry, IP, US$2.5 million. 64 The state continues to pay for the operating expenses of the court including the maintenance costs of the buildings and their opportunity cost after the conclusion of the project. 65 At design, it was thought the PIU would only last for the initial two years – and then the project would move to the MITPSD - but budget already considered more because by restructuring it was known that the PIU would stay for the overall project and the difference in budgeting was not so high. 27 training, and US$550 thousand for coordination meetings. Three percent of the Component 4’s expenses were devoted to M&E (excluding the team’s specialist salary). 99. The costs of Component 4 are considered high even if not uncommon to other projects in the region with large number of counterparts: for instance, the costs of the PIU for the Malawi’s Community-Based Rural Land Development Project (P075247) were 32 percent of the total budget. 100. In sum, given the difficulty in measuring effectiveness of the project subcomponents, there is a need of focusing this analysis on cost efficiency for both the original and the restructured PDO. In that line, the cost of the PIU throughout project implementation suggests that a Modest rating is appropriate both against the original and restructuring efficiency. The problems in efficiency were also raised by the implementation partners during stakeholder workshops held in preparation of this ICR (see Annex 6 for more details). 3.4 Justification of Overall Outcome Rating Rating: Moderately Satisfactory 101. The overall outcome rating of Moderately Satisfactory 66 is derived from individual outcome ratings using the guidelines for projects with formally Revised Project Objectives. 102. Table 1 demonstrates the weights calculated to generate this result. This takes into account that the project outcome is assessed against both the original and the revised project objectives. The outcome rating against the original PDO is Moderately Satisfactory reflecting the Substantial ratings on relevance and efficacy and Modest on efficiency. The outcome rating against the revised PDO is also Moderately Satisfactory reflecting the High rating on relevance, Substantial on efficacy and Modest on efficiency. Table 1: Outcome of BESTAP given formal revision of objectives Against Original Against Revised Overall PDO PDO Rating Moderately Moderately Satisfactory Satisfactory Rating Value 4 4 Weight (% disbursed 64% 36% before/after PDO change) Weighted valued 2.56 1.44 4.0 Final rating (rounded) Moderately Satisfactory 66 This rating is in line with the final Implementation Status and Results Report (ISR). 28 103. The project was overall deemed relevant for Malawi’s development strategy and achieved a large number of important outcomes, most importantly in conducting a set of legal reforms, improving the judicial system, improving ease of doing business systems, and developing government’s capacity. These achievements would unlikely have happened with the same success in the absence of the project. However, the project’s efficiency was Modest due to the costs of implementation. 3.5 Overarching Themes, Other Outcomes and Impacts (a) Poverty Impacts, Gender Aspects, and Social Development 104. Both the original and the restructured PDOs indicate that the project’s target group was local and international investors. Through new investment in the country, BESTAP aimed to increase employment and eventually affect poverty, but that was not stated as an objective because the link would be too far-fetched. In that vein, the project reached in generic terms its target group. In the short-term, the effects on the poor may have been limited but that was known from the onset. In terms of the target group, the only way of having a leakage of benefits would be through the overhead costs of the interventions. Putting in place a large project with fixed costs is likely to generate some fixed expenses that do not go directly to the benefit of the main target group. 105. The M&E Indicators were gender neutral. Even the matching grant program was not designed with the objective of disproportionally targeting women, hence there is limited information on the relative success in developing through this project women- owned enterprises for instance. (b) Institutional Change/Strengthening 106. A large emphasis of the project was on building the institutional capacity of the departments involved. A lot of effort was put on training programs and bringing the right set of skills and tools to (i) improve ways of doing things and (ii) continue pursuing future reforms. This was part of the original PDO. Based on individual interviews for this evaluation, as well as assessment of achievement post-project, there are indications of a stronger MITPSD, continuing the reforms agenda beyond the closure of BESTAP. 107. In the first aspect, BESTAP provided instruments for facilitating processes but it also conducted very intense capacity building exercises. Some officials spent long periods – 1 to 3 months – working directly with someone in a similar position in a different country. The objective was to go beyond in-class training and move to more intense learning mechanisms that could have lasting effects on a better way of doing the job. A total of 50 short-courses were also taken by officials attached to the MITPSD. 108. In the second dimension, the project supported Masters’ programs of two government officials, who returned to the Malawi public sector after completion. This has the immediate effect of removing officials from their jobs during the project, but it aimed to generate a strengthened department a couple of years down the line that could continue to develop reforms beyond project’s closure. 29 3.6 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops 109. These are detailed in Annex 3, Annex 5 and Annex 6. 4. Assessment of Risk to Development Outcome Rating: Moderate 110. The overall risk at the time of the evaluation that the development outcomes or the expected outcomes will not be maintained or realized is Moderate. 111. At the individual level, the main risks to the development outcome are macroeconomic, institutional, and political. The likelihood that the core achievements of the project are reversed is low to moderate and it is mostly concentrated in areas with limited relative importance. 112. The risk of Malawi having macroeconomics problems is high. In 2012, the depreciation of the exchange rate was over 100 percent, inflation has increased sharply, and fuel shortages became the norm. This reflected both political and economic issues in the final stages of the previous government. The problem for the sustainability of the development outcomes of BESTAP is that in a scenario of weak macroeconomics, the business environment as measured by the DB becomes less relevant for investment. The contribution to the investment climate from the project would still remain, but the relevance of that contribution would lose some of its importance. 113. The future evolution of the economy may have an impact as well in the effectiveness of the matching grant component. In this case it can go both ways - for instance, a recovering economy may mean that all businesses do well irrespectively of having received the cost-sharing grant. This would mean that the impacts of the matching grant program are reduced. A study in Tunisia67 has shown large impacts of a matching grant program on exports in the early years after the program, but then the impact was not sustained in the medium-term. 114. The institutional risk can take effect through two means. The more problematic one would be if some of the institutions that were created with the project and have committed to carry on work of the project are dissolved due to lack of funds. In the case of the Commercial Division of the High Court, there is little risk that the institution will close because there is strong ownership and to some extent it is self-sustainable68. For the PPD secretariat, the private sector has already continued supporting it, and it will continue as long as it sees value in this structure. 67 Cadot et al., 2012. “Are the Benefits of Export Support Durable? Evidence from Tunisia�. World Bank Working Paper 6295. 68 The court fees go the state’s budget but the government will recognize that it is a source of income that should be kept. 30 115. The second means of the institutional risk materializing is through the lack of maintenance of the systems installed. This is the case for instance for the (Information and Communication Technology) ICT systems put in place in Registration offices. It will be very important to keep monitoring its sustainability. 116. The risk of a political change that would lead to modifications in the legislation that was enacted is negligible because the reforms are not politically contentious in Malawi, and there is a strong commitment from both the new government and the supporters of the old government in pursuing these policies. These results of the project were the core of the development outcome achievements and will to some extent be irreversible. Furthermore, the new government has shown a tremendous enthusiasm in completing some of the reforms – for instance in legislations – that had been started under BESTAP. 117. Table 2 below summarizes the relevance of each risk for the main types of areas of intervention of the project. Table 2: Risks to achievements of the project BESTAP Political risk Macroeconomic Institutional risk Proxy PDO risk Indicators Commercial division Risk of sustainability of High-court Likelihood: Low Impact: Substantial Time to start a Risk of new processes business / register / systems being property dropped Likelihood: Low Impact: Substantial Legislative reforms Risk of change in Risk of reforms political course becoming irrelevant in an adverse economic Likelihood: Low environment Impact: Substantial Likelihood: Moderate Impact: High Impacts on firms of Risk of not being able the cost-sharing grants to tackle new opportunities Likelihood: Moderate Impact: Substantial 31 5. Assessment of Bank and Borrower Performance 5.1 Bank Performance (a) Bank Performance in Ensuring Quality at Entry Rating: Moderately Satisfactory 118. At entry, the Bank facilitated preparation by drafting a number of supporting documents that informed project design. These included the 2004 DTIS, the 2006 ICA, the annual Doing Business surveys, the 2006 Malawi Country Assistance Evaluation (CAE), the 2006 Governance and Anticorruption Report (GAC), and the Portfolio Improvement Plan (PIP) prepared by the CMU jointly with the Quality Assurance Group (QAG). 119. This knowledge was incorporated in the PAD and in the discussions with the government and other donors on what to do. The PAD is detailed in explaining the problems investors faced, and why it was thought the Bank could support the government in addressing private sector development issues. The appraisal was also substantive in identifying risks to implementation, and incorporating lessons from previous projects. The fiduciary aspects were well considered from the entry stage including a very detailed procurement plan with a timeline of activities. 120. The shortcomings at entry may have included some of the technical solutions identified by the team, particularly the very strong focus on business environment issues, in detriment of other constraints to private sector development like access to finance and access to markets. The Bank’s leadership in this debate certainly had an influence in the final design of the project. The problems in these early assessments were also present in the economic analysis of the project, which was based on very optimistic figures in areas where this type of economic analysis is not appropriate. There were also weaknesses in the preparation of the M&E framework notably in setting the PDO indicator. (b) Quality of Supervision Rating: Moderately Satisfactory 121. The supervision missions took place regularly and they adequately reported the developments in implementation of the project through ISRs and aide-memoires. The quality of the information prepared by the PIU and its usage by the Bank facilitated the conclusions and recommendations to government provided during supervisions missions. The quality of the supervision processes was high and detailed analysis of fiduciary issues continued throughout the life of the project. Given that three TTLs managed the project during its life time, this affected supervision early, but the consistency of approaches across time was a laudable feature of the project. 32 122. The M&E framework and the PDO were only revised at the margin – despite the restructuring – during the duration of the project, which hindered the possibility of more clearly establishing goals and assisting the government in achieving those goals. 123. BESTAP was rated Moderately Unsatisfactory in early 2009 when it became evident that two of the subcomponents in Component 3 would not be implemented and there were some delays in other aspects of the project such as the creation of the MITC. The team was quick to respond at that stage, but was also quick in migrating the project back to Moderately Satisfactory in September 2009, significantly ahead of the restructuring that only happened in late 2010, more than 18 months after the alert was raised. The reasons raised in the supervision mission69 of September 2009 for the upgrade in rating included that the team thought that the pace of implementation had improved significantly with the majority of the components registering notable progress. At that time, the team had worked with the government on a draft restructuring paper that did not include change in the PDO. Discussions over the next year led to a Level 1 restructuring and a request by the government of restructuring that was sent to Bank much later than the initial flag in early 2009. 124. Furthermore, in other activities that were not progressing as well or were progressing but not generating significant changes, the Bank remained relatively hopeful that they would turn around before the end of the project. There could have been scope to be more outspoken in raising flags of areas that were not achieving the goals. The project was often rated Satisfactory when only a limited number of subcomponents were achieving the desired results. 125. One issue of difficult judgment is the speed of disbursing of funds. While a high disbursement rate is often supported by the Bank’s structure, this may not always be aligned with the project’s success. The allocation of the expenses to where they make more sense should be priority. The Bank’s focus on keeping high disbursements rates can lead - like in the case of BESTAP - to overspending by some subcomponents at the cost of others, which is sometimes not in the interest of the project’s objectives. Certain activities with potential were not even implemented due to shortage of funds at the time that they were ready. (c) Justification of Rating for Overall Bank Performance Rating: Moderately Satisfactory 126. This rating combines the score for quality at entry and quality of supervision. Both of these ratings were Moderately Satisfactory, thus the overall Bank performance is also kept at Moderately Satisfactory. 69 Restructuring Appraisal & Implementation Support Mission, September 16-24, 2009. 33 5.2 Borrower Performance (a) Government Performance Rating: Moderately Unsatisfactory 127. While the Bank’s main shortcoming in implementation was not acting fast enough when there was a need for restructuring, the government’s shortcomings also included: the deterioration of the institutional environment affecting at the margin the importance of the project (see Section 4) and the speed of implementation of reforms during the critical years of 2010-12. 128. The commitment of the government to the project and in more general terms to the agenda of business environment was high at design. The government incorporated the doing business agenda into their policy goals. The government was also prudent when it decided to establish a PIU instead of running the operation from the MITPSD. The government’s support to the project went well in the first years, but there were political problems70 mid-way through implementation. Simultaneously, there were fuel, electricity and foreign currency shortages associated with these political issues. This was in part linked to the international crisis and its effects on the declining prices of commodities, but that does not explain the full scope of instability, as other countries in the region did not suffer to the same extent. 129. While BESTAP was implemented in consultation with other donors, in 2011 some donors including the UK suspended new aid 71 to Malawi following a series of political, social and economic controversies 72 . This reinforced an already worrisome economic and social unrest, which may have hampered the full success of the project. 130. While the first couple of years of implementation and the end of the project went well from the point of view of overall government performance, the instability created in the focal years of 2010 to mid-2012 combined with the delays in putting in place the changes in the project leads to a Moderately Unsatisfactory rating of the government. (b) Implementing Agency or Agencies Performance Rating: Moderately Satisfactory 131. Implementation of the project was managed by the PIU, part of which including the Project Coordinator, had migrated from the previous FPD project (PURP). The previous experience, high professionalism, and knowledge of Bank procedures and standards facilitated project implementation. This helps explain the fact that there were 70 http://www.bbc.co.uk/news/world-africa-15206437 71 Reengagement was obtained with the new government in 2012. 72 This included expelling the British ambassador after he criticized Malawi's leadership as autocratic in a leaked diplomatic cable (linked to the wikileaks case). 34 no significant delays in project disbursement. The fiduciary arrangements were also secured within the PIU and despite the changes in the FM and Procurement Specialist, they were kept to high standards of Bank procedures. 132. As discussed in the efficiency section, the costs of having an effective PIU are considered high. The effectiveness of the project investment is undermined when project management costs, which do not directly contribute to achieving PDO, exceed US$5 million out of US$18.7 million. 133. In addition to the PIU’s overall management of the project, ten implementing agencies were involved. The Steering Committee where these agencies were present had regular supervision meetings throughout project implementation but did not meet in 2012, allegedly for lack of funds. It seems that the relationships between agencies and between agencies and the PIU were not always smooth 73 . Ownership, alignment with the objectives of their subcomponents, readiness for implementation, and adequacy of monitoring of results, varied across implementing agencies. This helps explain why a better rating is not achieved for the overall outcome. 134. The implementing agencies’ performance is hence rated Moderately Satisfactory. (c) Justification of Rating for Overall Borrower Performance Rating: Moderately Satisfactory 135. This rating combines the score for the government (Moderately Unsatisfactory) and implementing agencies’ (Moderately Satisfactory) performance. Appendix A of the ICR Guidelines 74 recommends that when the rating for one dimension is in the Satisfactory range and the other is in the Unsatisfactory range (like is the case here), the overall Borrower performance is rated Moderately Satisfactory if the Outcome is rated in the Satisfactory range (as is the case here as well). 6. Lessons Learned 136. The lessons learned from preparation and implementation of BESTAP are diverse and include: (i) Projects can have valuable outcomes even when their objectives are not transformational. Given BESTAP’s focus on processes (systems and regulations), it raises the question whether individual projects should always aim at higher-level transformational outcomes such as poverty, economic growth, or investment, or whether it would be valuable to sometimes clearly accept that the project’s objective is just an interim outcome. Certain systems and processes, when implemented will not answer the 73 This is based on individual and group consultations to all agencies involved, including specific examples of how these may have affected completion of the proposed activities. 74 Page 41 of the OPCS ICR Guidelines (August 2006, updated 2011). 35 bigger questions, but they are needed especially in IDA countries. The question remains whether the Bank’s role can be to get involved in these areas, particularly if the changes lay the groundwork for more impactful far-reaching reforms. Currently, in the absence of the Bank’s support, these processes are often not improved, delaying other aspects in the development process. Hence, the Bank could have more candid approach to the question and acknowledge more often in project design that it may just want to successfully improve a new system or process, which is needed for more important things to happen. (ii) Applying a theory of change is important for understanding which assumptions are needed to achieve the project objectives. There are possible critiques of the Doing Business indicators, but their simplicity facilitates preparing a theory of change. It is possible with the DB to identify how certain outcomes such as time taken to process X can or cannot be changed by the activities of the project. In some cases, the team did not do early enough this kind of analysis, making it difficult to act swiftly in generating the desired change. (iii) Partnerships are good but do not underestimate time and costs of coordinating with other donors. Co-financing helped implement the project, but between the decisions in 2007 to work together and the release of the EC funds in 2011, there were four years of negotiations, delays in processes, and a number of activities started under the assumption of a larger envelope. For future, more discussion needs to happen between donors to review project documents and procedures. Any space for facilitating processes – ensuring the other donors’ processes are also followed – would naturally be welcomed. (iv) It is important to incorporate the new information in a timely manner. Once information affecting project design is available, it is important not to delay the restructuring process. Similarly, it is critical to secure that the funds allocated at restructuring for a new subcomponent are actually available for those activities. (v) Need for effective mechanisms of controlling coordination costs. Working with many government agencies was an objective in itself because one of the core goals of BESTAP was to increase the capacity of local agencies. A strong PIU was able to deal with this issue, but with high costs of coordination. It would be important to identify mechanisms of working with multiple players and at the same time keeping coordination costs low. (vi) More complex programs such as matching grants may require a pilot phase and commitment to learning from all levels of the project. When little is still known about traditional matching grant programs, it is difficult to make a case about their implementation or against it. Given the multiple options of implementation, it would make sense to pilot variations and then scale-up the most effective approach. In a country like Malawi where the culture of using matching grant programs is limited and the number of services providers available is low, it takes time to get a program up and running smoothly and effectively. Hence piloting within the initial phases of BESTAP would facilitate future implementations. 36 (vii) Public-Private Dialogue is important but should be as much as possible linked to results. The objectives of the PPD platforms need to be clear from the onset to avoid open discussions that may or may not generate results. The expected areas of reforms and ideally targets in terms of number of reforms should be set from the onset to guide expectations and lead to more productive results. (viii) Not all components of a project are amenable to an NPV/ERR economic analysis. In the case of BESTAP, the team prepared economic analysis using a net- present value methodology for all interventions of the project. This is an incorrect use of the methodology and questions the review process of project management. One way of addressing this is to have reviewers more focused on individual sections, in this case on the economic analysis. 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners (a) Borrower/implementing agencies The Borrower prepared a completion report, which is attached in Annex 7. It also provided the following comments on the draft ICR. The Bank’s response is noted in italics:  Given the recent progress in laws and regulations prepared under the BESTAP, the government will be available to update the report before its submission with the bills passed recently into Law. The Bank will confirm the status of the list of Bills in various stages (Annex 2) before finalizing the ICR.  One of the lessons learned mentioned in the report is that there is a need for acting in a timely manner once information affecting project design is known. BESTAP was certainly a victim of this. Despite the early recognition to discontinue the two activities (Merger of MSB and MRFC and the creation of the Malawi Development Fund) it was agreed that the restructuring should coincide with the mid-term review. (b) Cofinanciers The European Commission, which co-financed BESTAP, provided the following comment on the draft ICR. The Bank’s response is noted in italics:  Thanks for the report which is well written. It appears that our Finance and Contracts Section does not have comments at this point in time. One observation on the investment made through BESTAP vis a vis sustainability of the initiatives: ICT systems at the Registrar of Companies, Lands Registry among others. My reading of the report (especially page 35) indicates that the performance of Implementing Agencies was rated as Moderately Satisfactory. How committed are the agencies to keep the systems up and running? Are there long term maintenance plans or assurances from GoM that they will continue to maintain the systems? I know that these are side issues but critical to the objectives of BESTAP. The Bank agrees with the risk and has made it clear in the risks section. 37 There is for the time being strong commitment from the government in continuing using the systems put in place, but considering that could change, it would be important to set in place long-term maintenance plans. 38 Annex 1. Project Costs and Financing (a) Project Cost by Component (in USD Million equivalent) Actual / Actual / Actual PAD PAD Restructuring Component 1 Streamlining the Regulatory Environment for 811 1,100 74% Business (Ministry of Justice) Improving Access to Commercial Justice (High 1,841 1,300 142% Court, Commercial Division) Improving Services at the Business and Lands 3,443 1,850 186% Registries Department of Immigration (appears also as 6 0 Component 2 in the M&E Framework) 6,100 4,250 144% 93% Component 2 Establishing a One-Stop-Shop at the Malawi 351 900 39% Investment and Trade Centre (MITC) Establishing an Institutional Framework for 1,018 1,800 57% Public Private Partnerships (MoF/PC) Improving Private Sector Development policy 2,296 1,000 230% analysis and delivery (MITPSD) Supporting a Sustainable Framework for Public- 727 900 81% Private Dialogue (MCCCI) 4,393 4,600 95% 95% Component 3 Establishment of a sustainable SME venture finance institution / investment fund 0 600 0% (MoF/private sector) Business Growth Scheme 3,298 3,200 103% Merger and transformation of MRFC and MSB 0 800 0% into a viable rural finance institution (RBM) 3,298 4,600 72% 89% Component 4 Project Implementation Unit and M&E 5,209 2,900 180% 144% Unallocated 2,349 Total Expenditures 18,999 18,699 102% 102% 39 (b) Financing Appraisal Actual/Latest Type of Estimate Estimate Percentage of Source of Funds Cofinancing (USD (USD Appraisal millions) millions) Borrower 0.00 0.00 .00 EUROPARTNERS BANK 3.70 3.58 97% IDA Grant 15.00 15.44 103% Annex 2. Outputs by Component Project Objective The main PDO indicator - Cost to formally start a business as a % of GNI per capita – was partially achieved. The progress was obtained through a steady decline in this cost in relative terms to income per capita, but as discussed in the main section this indicator had problems. Figure 2: Cost to formally start a business as a % of GNI per capita (comparison with BESTAP target) 245 215.7 210 188.7 175 140 125.9 108 108.4 105 90.9 83.7 Target 70 35 0 DB2007 DB2008 DB2009 DB2010 DB2011 DB2012 DB2013 Source: Adapted from DB reports. BESTAP target is for DB2012. 40 Figure 3: Cost to formally start a business as a % of GNI per capita (comparison with region) Generic goals In general terms, it is also relevant to track the DB ranking, although there is an ongoing review to assess its relevance. As argued in different sources 75 , the country ranking classification is under extreme scrutiny following China and other countries concerns about it. During the implementation of the project, Malawi’s DB ranking kept falling as more countries embarked on reforms and the country was not as fast in generating change. The country started at 110th out of 175 economies in the DB2007 and ended in 145th out of 183 economies in the DB2012 (figure 4). 75 http://www.project-syndicate.org/commentary/making-the-most-of-the-world-bank-s-doing-business- report-by-ricardo-hausmann 41 Figure 4: Malawi’s Progress in the Doing Business Ranking 157 160 145 134 132 133 127 120 110 80 40 0 DB2007 DB2008 DB2009 DB2010 DB2011 DB2012 DB2013 Figure 5: Malawi’s Distance to DB frontier76 Furthermore, the PDO both at approval and at restructuring mentions a link between the easiness of doing business and a major outcome variable. Initially that link is economic growth, while in restructuring this is revised to local and international investment. 76 The distance to frontier measure shows how far on average an economy is from the best performance achieved by any economy on each Doing Business indicator since 2005. The measure is normalized to range between 0 and 100, with 100 representing the best performance (the frontier). The overall distance to frontier is the average of the distance to frontier in the 9 indicator sets shown in the figure. See the data notes for more details on the distance to frontier measure. 42 Although, as discussed at length in the main text, the changes in these variables cannot be attributed to the project, it is important to monitor how these have evolved during its duration. This is done in Figure 6 to provide a more general picture of the environment happening during the project. The government has indicated that at a minimum Malawi ought to grow at 6 percent to have effects on reducing poverty. During the period of the project77 (2008-2012), the Malawian economy grew at a compound average growth rate of 6.4 percent (EIU, 2013). This compares with 6.9 percent in Mozambique, 6.5 percent in Zambia, and 0.0 percent in Zimbabwe. The investment share in the economy remained relatively stable during the period, indicating that it has followed the trend of economic growth. Net FDI at an average of US$100 million per annum is still small. Net inflows of foreign investment in Malawi represented 15 percent of the FDI in Zambia and 7 percent of the one in Mozambique78. Figure 6: Malawi’s GDP Growth and Net inflow of Investments (as % of GDP) 26 26 25 24 23 23 12.1 10.8 8.3 5.1 5.2 1.6 2007 2008 2009 2010 2011 2012 Investment (% of GDP) GDP growth Components The project supported four components: a) Component One: Strengthening private property rights institutions and business facilitation; b) Component Two: Strengthening private sector development support institutions and services; 77 Here we are computing calendar years 2008-2012. However, the project started in mid-2007. 78 These are of course different countries with different size, but the population numbers are not very different between Malawi and Zambia. Malawi is also only one third smaller than Mozambique in number of people. 43 c) Component Three: Promoting access to finance and productivity enhancement d) Component Four: Capacity building and implementation support Component One: Strengthening Private Property Rights Institutions and Business Facilitation This component supported activities that aimed to improve the regulatory environment in which the private sector operates, to strengthen the institutions that protect private property rights, and to allow speedy and low cost business facilitation. It included three sub-components: (i) streamlining the business regulatory environment, including addressing the large legislative backlog of economic laws and introducing key new legislation; (ii) providing support for strengthening and expanding the capacity of the newly established Commercial Division of the High Court to relieve the large backlog of commercial cases in Malawi’s court system and to strengthen contract enforcement; (iii) enabling the business and land registries to improve their effectiveness in registering businesses and facilitating the registration and securitization of land-based assets. (i) addressing the large legislative backlog of economic laws The project financed the team of people attached to the government with the objective of helping to prepare regulations and updating outdated laws to facilitate investment. Table 3 describes the laws enacted or in the process of being enacted during the project duration. BESTAP aimed initially to update/prepare 40 laws up to Parliament/Cabinet stage. According to Table 3 below, 19 out of 46 proposed laws are already enacted / in force and five others are before the Parliament or Cabinet. A number of others are in the process of being commented by the respective authorities. Hence, significant progress was done in this area, but the work has not been concluded. Overall, this was one of the strongest subcomponents of the project with significant progress in revising the legal environment for firms and investors operating in Malawi. Among the laws listed below, there is the review of Companies Act and the Investment Act, which were added at restructuring as core activities of the project. Part of the funds released from the components dropped in Component 3 was used to assist this legislative review. Table 3: List of Bills/Acts drafted or revised under BESTAP Bill / Act Approval Enacted 1 Investment and Export Promotion Bill 2012 2 Public-Private Partnership Bill 2011 3 Employment Amendment Bill 2010 4 Credit Reference Bureau Bill 2010 5 Financial Cooperatives Bill 2011 6 Microfinance Bill 2010 7 Securities Bill 2010 44 8 Business Registration Bill 2012 9 Financial Services Bill 2010 10 Reserve Bank of Malawi Amendment Bill 2011 11 Banking Bill 2010 12 Insurance Bill 2010 13 Pension Bill 2011 14 National Registration Bill 2010 15 Personal Property Securities Bill 2013 In force 16 Companies (Winding-up) Rules 2010 17 Road Traffic Registration and Licensing Amendment 2010 Regulations 18 Road Traffic Miscellaneous Fees Regulations 2010 19 Road Traffic Prescribed Offences and Penalties 2010 Amendments Before Parliament 20 Business Licensing Bill 21 Companies Amendment Bill Before Cabinet 22 Export Processing Zones Amendment Bill 23 Tenancy Labour Bill 24 Insolvency Bill Submitted to relevant authority for comments 25 Tax Administration Bill 26 Malawi Revenue Authority Amendment Bill 27 Taxation Amendment Bill 28 Customs and Excise Amendment Bill 29 Value Added Tax Amendment Bill 30 Export Incentives Amendment Bill 31 Copyright Bill 32 Arbitration Bill 33 Malawi Road Traffic Authority Bill 34 Malawi Ports Authority Bill 35 Control of Goods Amendment Bill 36 Labour Relations Bill 37 Conveyancing Act Being drafted / waiting instructions 38 Tenancy Labour Forms Regulations 39 Consumer Protection Bill 40 Competition and Fair Trading Amendment Bill 41 Cotton Bill 42 Tobacco Bill 43 Corrupt Practices Bill 45 44 Land Act 45 Registered Land Act 46 Money Laundering, Proceeds of Serious Crime and Terrorist Financing Act The policy choices to select the laws of focus were based on the Malawi Investment Climate Assessment 2006 and on the Prioritization of Economic Laws for Review 2008. The drafting of the laws received support from international experts. For instance, the Personal Property Security Bill 2013 and the Insolvency Bill 2013 are related to areas that are fairly complex and where the level of local expertise was deemed not sufficient. Following the Doing Business Reform Memorandum for Malawi prepared by the World Bank Group in 2009, the WB Foreign Investment Advisory Services (FIAS) provided technical assistance in reviewing and reforming these two laws. For the first, there was a need to carry out a diagnostic of the personal property security system and the insolvency framework. These were benchmarked against international instruments such as the UNCITRAL Legislative Guide on Insolvency and the WB’s Principles for Effective Insolvency and Creditor Rights Systems. The consultants carried out extensive discussions and interviews with various stakeholders in the credit system and insolvency systems. These included practitioners, the judiciary, financial institutions, etc. Furthermore, the Personal Property Security Bill was presented and extensively discussed at stakeholders’ forum where a large group of participants, including commercial law judges, registrars, business owners and creditors, voiced their opinions. The Insolvency Bill used aspects of the Malawian legal environment, best practices from the Mauritius Insolvency Act of 2010 and aspects of the English Insolvency Act amended in 2002, while at the same time following best practices from the UNCITRAL Legislative Guide on Insolvency. (ii) providing support for strengthening and expanding the capacity of the newly established Commercial Division of the High Court Another strong subcomponent of the project was the development of the Commercial Division of the High Court. Given the time taken to resolve commercial disputes in the traditional Magistrate’s Court, the establishment of an alternative Commercial Court starting from scratch and fully concentrated in these cases was seen as a good opportunity. According to the DB 2007, it took 337 calendar days to settle a commercial dispute. However, for cases disputing more than MWK 1 million79, the establishment of the Commercial Division of the High Court allowed for usage of a dedicated team, including three judges in Blantyre and 1 judge in Lilongwe80. A study conducted by the GoM suggests that on average it takes 96 calendar days to settle a commercial dispute in the Commercial Division of the High Court versus 432 days still in Magistrate’s Court. 79 The Magistrate’s Court is focused on cases below MWK 2 million. The exchange rate is currently at MWK 370 to dollar, but was during the project mostly at MWK 165 to dollar. 80 The objective is to have four judges in Blantyre and two in Lilongwe. 46 In Lilongwe, the Court has addressed over 200 cases since 2009, and in Blantyre, that figure is over 1,000 cases. Large majority of the cases still pending are those recently started. Therefore, it is possible to keep the levels of three months to deal with a case. The problem with the Commercial Court is that it is more expensive than the Magistrate’s Court, hindering some people from using it. Additionally, information does not flow perfectly and some lawyers still advise people in using the Magistrate’s Court as a way of keeping control of a longer-duration case. The project supported the High Court with training including extensive study tours, management information systems, vehicles, furniture and equipment, and workshops. Given the small team – currently there are 34 people working in the Blantyre court – a major issue is staff turnover, especially after receiving customized training. Table 4: Cases in the Commercial Division of High Court81 Source: PIU (iii) Improving Business Registration and Land Registration Services Business Registration For the Department of Registrar’s General (DRG) in Blantyre – responsible for overseeing national business registration - the project supported its institutional development with IT systems (computerization of existing records, business registration of application platform), and capacity building (study tours, training). The total investment amounted to US 1.7 million, which is in line with the cost of these activities 81 Cases in 2012 refer to activity only until July 2012. 47 in other countries82. Activities were mostly completed but the new system has still not be launched83 allegedly because of a dispute between the PIU and the staff of the DRG over the final capacity building program. The objective of this intervention as measured in the M&E framework was to improve the processes of registering firms. At restructuring, where the number of indicators was increased and targets were updated, the objectives and their results included the following: Table 5: Objectives of the Business Registration reforms Indicator Objective Achieved? Automated business registry Operational Yes database at DRG % of electronic transactions 20% in 2010, 50% in 2012 No, not operational Time to start a business 5 days No, still at 39 days in 2012 (see Figure 10) Cost to register business US$80 Depends on firm, cost is US$60.8 plus 1% of nominal capital 82 According to Investment Climate IT Specialist at the WBG - Mozambique: Information System with no customized development, US$130 thousand; Liberia: Business registration system, US$1.5 million; Rwanda, Business registration system, collateral registry, IP, US$2.5 million. 83 This was verified not to be in operation in late February 2013. However, the system was ready to be used. 48 Figure 7: Time to start a business84 50 39 39 39 39 40 37 37 37 30 30 20 15 10 5 Target 0 DB2007 DB2008 DB2009 DB2010 DB2011 DB2012 Projected Actual In some indicators like time to start a business, this subcomponent alone would not be in conditions to change the outcome. There are ten steps included in the registration process and this subcomponent only dealt with two of them accounting for 8 of the 39 days needed to register a firm. There are other subcomponents including the backlog of economic laws that can have an effect on this target. A total of five steps are addressed by the project. Going forward, the legislative work funded by BESTAP on removing the need of a company seal, reducing the time it takes to get a business license of City Assembly from 29 to 7 days (Business License Bill), and reducing the requirement of inspections in every firm, should allow for improving the time to start a business target. Land Registration Similarly to DRG, the project also supported the computerization of the process of land registry at the Department of Lands (DoL), including an application platform and database of existing records, as well as training programs. Total investment was also of approximately US$1.7 million. Progress in this subcomponent was faster than in DRG with the system fully operational by the end of BESTAP85. In a visit to the DoL in preparation for this ICR, the staff’s capacity in using the new system was uneven, but the appreciation of the project’s work was high. 84 The term “starting a business� is used rather than “business registration� because registration is only one of many steps of “starting a business�. These steps include obtaining a City Assembly business license and registering for taxes (TPIN) among others. See Annex 11 for more details. 85 This was verified in a visit to the Department in February 2013. 49 In terms of the subcomponent’s outcomes, the main objective that was part of the list of indicators since approval was time to register property. The target of 60 days (from 118 days in 2007) was met by the DB2011, but then a review in 2012 brought it up again to the DB2012 level of 69 days. This is shown in Figure 8, which indicates a strong progress in this indicator. Figure 8: Time to register property 140 118 118 120 100 100 88 80 70 69 80 Target 60 60 40 49 20 0 DB2007 DB2008 DB2009 DB2010 DB2011 DB2012 Projected Actual The restructuring paper again added two indicators - % of electronic transactions and costs to user for public services. The first was fully met with all new requests being done electronically at the end of the project. The second was difficult to assess because the target at the end of the project was initially set at the same level as baseline86, implying no change. The target is also dependent on the value of the land, which makes it difficult to monitor. Actually these restructuring indicators were not monitored during neither in the supervision missions nor in the PIU’s regular reporting. Component Two: Strengthening Private Sector Development Support Institutions and Services This component focused on building capacity of institutions that provide services to the private sector in order to improve the quality and volume of services delivered, and to strengthen institutions that provide policy direction on PSD issues. The proposed activities included (i) supporting the institutional capacity of the newly established Malawi Investment and Trade Centre, including setting up the legal and regulatory 86 If not a typo, this would suggest that the objective would be in practice to keep the cost constant in nominal terms despite inflation. This indicator was not monitored in ISRs post restructuring, thus it is not clear whether the team actually followed it. 50 framework for a one-stop business licensing, investment and exporting promotion centre; (ii) establishing an institutional framework for Public Private Partnerships; (iii) build capacity in the Department of Private Sector Development of the MITPSD, including hiring of a small team dedicated to working on DB reforms; (iv) establishment of a public-private dialogue (PPD) secretariat in the Malawi Confederation of Chambers of Commerce and Industry (MCCCI) by funding two full time positions. At restructuring, it was added a fifth subcomponent on computerizing the process for issuing Temporary Employment Permits and Business Residency Permits. (i) One-stop-shop at the Malawi Investment and Trade Centre The investment in this sub-component ended up being only one third of the planned at approval 87 . This reflected the lack of progress in merging the Malawi Investment Promotion Agency (MIPA) and the Malawi Export Promotion Council (MEPC), which only happened in 2011. In practice, the bulk of the support (82 percent) comprised staff training. The project also supported an ICT Needs assessment, an Investor Road Map Study, a National Investment Policy, and an Investment Incentives Review. Of the planned activities at approval, the project did not support a business process re-engineering and restructuring of internal operations, as well as IT systems with capability of investor tracking and online license applications. The main outcome to measure success for this subcomponent was established at approval to be the time to obtain (construction) licenses. The team used for the PAD the DB metric on this, which indicated 185 days were needed to have a construction license. The target by the end of the project was 30 days, but that was not achieved. At restructuring though, the indicator on time for a construction license was dropped and replaced with the time to obtain a Business Residence Permits (BRP) and Temporary Employment Permit (TEP). This would be monitored in the Department of Immigration’s subcomponent, which never took off. 87 US$350 thousand instead of the US$900 thousand planned at approval. 51 Figure 9: Time to obtain construction license 300 268 250 213 213 213 185 200 200 150 185 90 100 70 50 50 30 Target 0 DB2007 DB2008 DB2009 DB2010 DB2011 DB2012 Projected Actual (ii) establishing an institutional framework for Public Private Partnerships This subcomponent achieved its objectives of developing initial competences and a framework for implementing PPPs in Malawi. With half the investment proposed at approval, the subcomponent incorporated workshops on PPPs, training of the team at the MoF, and support for a new PPP Act 88 . The goals of the subcomponent were not extremely ambitious but that allowed as well keeping it achievable. This intervention helped to develop a small structure in the government, which in addition to working on a limited number of PPPs, is supervising the privatization of some companies such as Air Malawi, a bus company, and a transport terminal. (iii) build capacity in the Department of Private Sector Development of the MITPSD This subcomponent was very much aligned with Component 1.1 on addressing the backlog of legislation because the team set at the MITPSD with the support of the project – the Doing Business Unit (DBU) – was especially concentrated in advancing the legislative framework. The list of reforms already incorporated in the Doing Business Surveys and the involvement of this unit are presented in Table 6 below. 88 This law was in conjunction with the subcomponent on enacting new legislation. 52 Table 6: Doing Business Surveys reforms (DB2008-DB2013) Indicator set Reform Year DBU involved Registering Malawi eased property transfers by cutting the DB2011 Property wait for consents and registration of legal instruments by half. Getting Credit Malawi improved its credit information system DB2012 Yes by passing a new law allowing the creation of a private credit bureau. Paying taxes Malawi decreased the time for tax compliance DB2010 by encouraging the use of electronic systems. Malawi introduced a mandatory pension DB2013 contribution for companies. Trading across The implementation of a risk-based inspection DB2010 Yes borders regime and a post-destination clearance program for pre-approved traders has reduced the delays for clearing goods in Malawi. Trading across borders in Malawi became easier DB2013 Yes thanks to improvements in customs clearance procedures and transport links between the port of Beira in Mozambique and Blantyre. Enforcing Malawi has made enforcing contracts easier by DB2008 contracts opening a commercial court and hiring new judges. Malawi simplified the enforcement of contracts DB2011 by raising the ceiling for commercial claims that can be brought to the magistrates court. Resolving Malawi introduced a new law limiting the DB2010 insolvency liquidator's fees during insolvency procedures. Malawi adopted new rules providing clear DB2012 Yes procedural requirements and time frames for winding up a company. At restructuring, the team added a target under this subcomponent on the number of reforms measured by the DB reports. The objective was to move from two recognized reforms per year to three reforms per year. However, the number of reforms per year remained at two per year during the final stages of the project. It is expected though that the recent legislative reforms will generate recognized reforms on indicator sets such as starting a business. (iv) establishment of a public-private dialogue (PPD) The project supported the creation of a PPD secretariat, which helped drive the agenda of the dialogue between the private and the public sector. The secretariat was established at the Malawi Confederation of Chambers of Commerce and Industry (MCCCI), the main hub of business associations, and involved the development of ongoing relationships with the MITPSD. 53 There were more than ten PPD meetings between the private and the public sector until 2011 89 . PPD meetings discussed cross cutting issues such as access to finance, mechanisms of promoting export diversification, firm productivity, regulations, and constraints in accessing utilities such as energy and water. The PPD comprised the creation of sub-committees dedicated to discuss specific issues. Although the PPD was seen as a good opportunity of leveling the playing field of the discussion between the private and the public sectors, it was kept mostly at the high-level, eventually driving follow-up discussions on particular issues that may have helped generate specific reforms. Hence, its link to changes in business environment is so far been relatively limited. The sustainability of the secretariat post project was kept with MCCCI taking it over, funding one of the two core members of the team. Post project conclusion, there was already a PPD meeting in 20 February 2013. (v) Business Residency Permits (BRP) and Temporary Employment Permits (TEP) At restructuring in November 2010, the government decided to change priorities and drop two sub-components on access to finance under Component 3 and among other aspects added a subcomponent on working with the Department of Immigration to improve the processes of obtaining BRPs and TEPs. These permits are requested by international investors and workers of international companies and the process was (and still is) manual, lengthy (approximately 40 days) and cumbersome (steps include that each of the permits needs approval by the Minister of Labor and the cases go by letter from one department to the other). The goal of the subcomponent was to computerize the process for issuing BRPs and TEPs, including providing ICT infrastructure, as well as supporting capacity building in skills development in using the new system. This intervention was seen at restructuring to be very aligned with the reforms being done in business registration, property registration, and in creation of the one-stop shop. Given the limited progress in the latter, the focus on additional permits was seen as a good opportunity to continue to push for changes in things that could be easily controlled by the government. However, the implementation of this intervention did not take place. The main reason was the depletion of funds by other subcomponents before this one could be implemented. ToRs for the planned activities were prepared but there were no funds left to contract the firm. It was argued that this subcomponent was part of Component 2 and the funds available from co-financing of EC were only available for Component 1. But the restructuring paper is not clear on the Component to which this intervention belonged to. On one hand, it mentions that it is part of Component 1 (page 21, and the allocations in page 24). On the other hand, it is placed in Component 2 in the M&E framework. 89 In early 2012, there was a change in government and the dialogue restarted more formally in 2013. 54 In both situations, this problem reveals that the restructuring should have made this aspect more clear and should have ensured that there were funds available for the activities that were being planned. Component Three: Promoting Access to Finance and Productivity Enhancement The objective of this component was to support the growth and development of micro and small enterprises, into a “missing middle�, given few enterprises were located in between the many micro and informal sector businesses, and the larger multinationals and conglomerates. This component aimed (i) the establishment of a sustainable SME Investment Fund to increase access to finance for small-scale enterprises; (ii) the introduction of a matching grant scheme aimed at supporting business development services; and (iii) merging of two state-owned banks - the Malawi Rural Finance Company (MRFC) and the Malawi Savings Bank (MSB) to improve the service delivery of rural finance. (i) establishment of a SME Investment Fund This subcomponent was dropped at restructuring. The objective at approval was to help create a venture capital type fund partly 90 financed by the MoF and that would be providing loans, equity, quasi equity and guarantees to SMEs. But in the PAD, it was also mentioned that it could be just a loan guarantee fund, suggesting that there was no full indication of which intervention would be pursued. The lack of progress in the creation of the SME Investment Fund led to the right decision in late 2010 to drop this subcomponent, which did not have any PDO indicator attached to, but had an intermediate indicator - the number of SMEs benefiting from the fund91. The reason for the change of course was mentioned in the restructuring paper to be the government’s decision in January 2009 to move to the creation of a development bank rather than a venture capital fund. This would imply for substantial seed capital and a different institutional set up that BESTAP was not ready to provide. In practice, this suggests that the design of this subcomponent was not well discussed at approval and the decision to keep a placeholder for this subcomponent of the project was likely not the right approach. (ii) introduction of a matching grant scheme for business development services (BDS) This component comprised the creation of a unit – Business Growth Scheme (BUGS) - responsible for managing a matching grant program to support firms with 50 percent of the cost of accessing BDSs such as training programs, participation in international fairs, 90 The rest would be financed by the private sector. 91 The target was 200 SMEs by the end of the project. 55 business plans, consultancy services on inventories, new products, etc. The unit also provided advisory services to firms. BUGS was first established in Blantyre and then expanded to Lilongwe. The table below summarizes the take-up of the program at the end of the project. It was worth noting a significant pick up in number of activities in the last few years of the project. Annex 3 explains the situation in May 2010 when there were efforts to conduct a rigorous impact evaluation. Annex 5 goes through a summary of the assessment of the program conducted in 2012 in advance of this ICR. Table 7: BUGS performance Source: PIU (iii) merging of two state-owned banks - the Malawi Rural Finance Company (MRFC) and the Malawi Savings Bank (MSB) The objective of this subcomponent was to support the merger of two state-owned banks that were identified at approval not to be sufficiently effective in developing access to finance in rural communities. The objective was to streamline operations of these two banks by creating a joint bank that could serve clients with more products. While MRFC was not licensed as a bank, MSB was restricted to grow by its inability to achieve the required minimum share capital. 56 This subcomponent was dropped at restructuring and no funds were expended, in this case because the government decided not to go ahead with the merger. Differently than in the case of the SME Investment Fund, the action of what to do was well known at approval, but in hindsight, key informant interviews suggest that the decision was not fully owned by the government. 57 Annex 3. Economic and Financial Analysis Component 1 and 2 The PAD presented in its Annex 9 an economic analysis for every subcomponent of the project. The analysis at approval suggested a NPV of US$20 million and an ERR of 49 percent. In spite of efforts with the original TTL of the project, it was not possible to recover for this evaluation the files that helped construct the cost-benefit analysis at design. Given the nature of the majority of the interventions – legislative reforms, new processes, new IT systems, capacity building – the figures presented at design were purely speculative. An NPV or ERR should not have been used to assess whether to go ahead with interventions such as those in Component 1 and 2. Therefore, for the purpose of this ICR, a new model for assessing the efficiency of these interventions was not replicated. The interventions in Component 1 and 2 are grounded on the assumption that they help facilitate local and international investment. For instance, with a better system of enforcing contracts, investors would at the margin be more attracted to invest in Malawi vis-à-vis other countries that do not have the same conditions. The same holds true if the legislation for developing businesses is transparent and clear. However, the actual investments are not solely dependent on these parameters, but could also be driven, for example, by good economic opportunities. In that vein, the important efficiency questions for this type of interventions are (i) whether there are more productive alternatives within the same scope of doing business activities and (ii) whether implementation was done at the least possible cost. On the first question, we can take the example of the business registration reforms under the project – a new registration bill, digitization of existing records, and computerization of the application process. After visiting the DRG offices in Blantyre, it is easy to argue that these reforms are necessary. But, the question is whether they make a difference in terms of generating new registrations and making an impact in SME development and poverty reduction through formalization. The numbers from other countries suggest that sometimes – not always - improving processes generate increased demand for business registration, especially driven by new entrants92. Furthermore, the project incorporated under its auspices an impact evaluation of business registration (see below). The objective of the ongoing study is to understand the usefulness of the business registration reform from the perspective of the firms at the margin of becoming formal. In particular, it studies whether firms would respond to a scenario where the process of registration was extremely easy and whether firms would benefit from becoming formal, as often argued. 92 Bruhn, 2013. http://blogs.worldbank.org/allaboutfinance/one-stop-shops-do-they-or-don-t-they-increase- business-registration 58 On the second question (regarding the cost), a comparison of the cost of the activities with other countries conducting similar reforms seems appropriate. In the case of new IT systems for business and land registration for instance, it seems the costs were in line with other countries doing similar investments93. A least cost approach analysis of the commercial court may suggest though that other options could have been cheaper, but these at the risk of not achieving the objective. Given that there was a decision to fully fund a new structure under the High-Court rather than making more judges available in the old Magistrate’s Court for commercial cases only, it is questionable that this was the cheaper option from a pure efficiency point of view. If the Magistrate’s court were to place 3 judges and 30 of their (new) staff solely working on Commercial Cases, it is possible that the same results – 96 days on average to solve commercial disputes - could have been achieved. However, this higher cost that the project incurred seems to be justified on the risks that the efficient approach would have had a higher variance of results: a non-negligible chance of not performing. A new agency with separate space reduced the political-economy institutional risks associated with implementation. Business Registration Impact Evaluation (BRIE)94 The informal sector is usually defined as the subset of firms that are not registered with the government and/or with a tax paying authority and is often large in developing countries such as Malawi. According to the World Bank Doing Business Report, “informality comes at a cost: firms in the informal sector typically grow more slowly, have poorer access to credit and employ fewer workers�. While an informal enterprise avoids paying taxes and complying with other regulations, it “loses access to public goods and other benefits of official status, such as external finance� (La Porta and Shleifer, 2008). Governments around the world have spent much of the past decade extending a helping hand to informal businesses by trying to make it cheaper and less burdensome to formalize. Since 2004, 75 percent of the countries included in the Doing Business survey have adopted at least one reform making it easier to register a business (IFC, 2009). Yet, despite these efforts, the majority of firms in most developing countries remain informal, with studies which have examined the impact of these regulatory reforms finding that much of the action comes from increases in entry of new firms, rather than from formalization of existing firms (e.g. Bruhn, 2011, Klapper et al., 2006). Knowledge is particularly lacking on the ability of facilitation efforts to induce formalization in a low-income African context. Furthermore, although the benefits of business registration may include increased access to finance (business bank accounts 93 According to Investment Climate IT Specialist at the WBG - Mozambique: Information System with no customized development, US$130 thousand; Liberia: Business registration system, US$1.5 million; Rwanda, Business registration system, collateral registry, IP, US$2.5 million. 94 Campos, Goldstein, and McKenzie, findings from ongoing study. 59 and loans), access to markets (government procurement, export licenses, customers who demand receipts), access to government support programs, and access to networks (chamber of commerce), other development obstacles are complementary to informality. It has recently been argued that increasing access to bank accounts have promising effects for households and micro firms (Ashraf, Karlan and Yin, 2006; Brune et al., 2011; Dupas and Robinson, 2009). In that vein, this is an area of important debate and one where targeted interventions can complement the value for firms of becoming formal. One specific problem for SMEs in Malawi is that household and business resources tend to be strongly intertwined. Entrepreneurs tend to muddle up finances, which then frequently results in depletion of working capital. According to baseline data for this study, 82 percent of the targeted group of firms takes money from the business for own or household expenses whenever needed. Hence, training on the specific benefits of separating household and business money coupled with offering business bank accounts may have strong positive effects in adding value to business registration. Evaluation questions The Malawian government is interested in evaluating whether or not formality improves enterprise performance and, at the same time, in exploring potential reasons for high informality. This study aims to answer the following main questions:  What is the value for SMEs of becoming formal, particularly among the larger of the informal firms? This study measures the effects of business registration on access to finance, access to markets and networks, investment in the business, growth and profitability, employment, harassment in regards to the business, standards of living of the entrepreneurs, and household decision making.  Do both male and female-owned enterprises gain equally from registration? Are the effects of registration heterogeneous on other dimensions including for certain sectors or for younger (older) firms?  Does separating business from household money complement the value of becoming formal? With increased separation of household from business money, firms may (i) start tracking revenue and costs accurately, with effects on their budgeting, (ii) manage working capital better, (iii) reduce the risk of lack of self-control on the usage of money, and (iv) protect the budget devoted to the enterprise from appropriation by other household members/network of contacts. Intervention Under BESTAP, the Malawian government helped streamline its registration process in order to increase the registration rate amongst small-to-medium enterprises. BRIE is a direct response to the government’s interest in evaluating whether or not business registration improves enterprise performance. The government will therefore use the results to create incentives for registration in the future (if positive) or to identify other bottlenecks that constrain enterprise performance (if negative). 60 In an ideal experiment for testing the impact of business registration on enterprise performance, a randomly selected subset of a sample of unregistered firms would be automatically registered. To replicate this ideal experiment as closely as possible, the intervention that we have been implementing tries to make registration totally costless. That is, a team hired by this study visited business owners in the treatment group and offered help in registering their businesses, while conveying to them a single-page information flyer with the potential benefits offered by registration. For those that were interested, the team helped them fill out the Business Registration form, took the required photo, and delivered their entire application to the DRG, also paying the registration fee. Once ready, the team delivered the Business Registration Certificates (BRC) to these firms. We invited through this process 2,250 firms of our sample of informal MSMEs to register at the DRG. Out of these, we offered a random group of 300 of them to also register for taxes and thus obtain a Tax Payers Identification Number (TPIN). The formalization interventions help answering the two initial questions of this study. Additionally, we assess the complementary value of an intervention to increase the separation of household from business money. The additional intervention involves informing entrepreneurs on the benefits of separating business from household money and offering of business savings accounts. Evaluation methodology As the graph below illustrates, the study intervention consists of four experimental groups, including a control sample. In order to test the impact of business registration on business performance, informal firms get randomly assigned to costless registration with either just the Department of the Registrar General (750 firms) or to both the Department of the Registrar General and Malawian Tax Authority (300 firms). In order to assess the complementary value of separating household finance from business finance, a random subsample (1,200 firms) is also invited to training on business bank accounts. We are targeting in this study the informal MSMEs that could potentially benefit the most from business registration and that the government has said that would be their first group of interest for a future road-show on business registration. We are targeting firms in urban Lilongwe and Blantyre, the major commercial cities in the country. 61 Short-term effects The first main conclusion of the intervention is that making registration costless works in terms of getting firms registered. According to preliminary figures, the take-up of the offer to help registering was of circa 75 percent. Take up is higher when business owners are also invited to an information session by a commercial bank (85 percent) but lower for the Business Registration Certificate when firms were offered at the same time to register for taxes (69 percent). As expected, the refusal rate for the registration for taxes part was extremely high. Only 4 percent of this group took TPIN on top of Business Registration. On the other hand, the demand for the commercial bank information sessions and business bank account was high. Preliminary results suggest around 71 percent take-up of the information sessions, with over 90 percent of these participants opening a bank account in the name of the business in the course of these sessions. Following these interventions, which were conducted in 2012, the team will follow these entrepreneurs beyond the closing of BESTAP for a total period of 2 years. This follow-up will be done through firm-level surveys, eliciting information such as business performance, access to finance, access to markets, harassment, etc. The first of this follow-up surveys was conducted between late November 2012 and March 2013. It was important to assess the immediate short-term effects of the interventions. The results of the short-term analysis suggest that immediate effects of business registration are modest in changes in behavior beyond a small usage of the certificate. On the other hand, coupling the intervention with access to bank accounts seems to generate increased savings in formal accounts – any type of account – better usage of financial records, and reduction of the intertwine between household and business money. _______________________ Component 3 – BUGS (matching grant program) The only component where it would be appropriate to assess the economic analysis through a NPV approach is Component 3. Here, two subcomponents were dropped during implementation, while the matching grant program (BUGS) remained. It would have been appropriate to include a rigorous impact evaluation from the onset of the project to compare the actual effects of the matching grant with the economic analysis, but that was not been attempted before 2010. A set of problems limited the possibility of incorporating an impact evaluation. The reasons for not completing the rigorous impact evaluation are summarized in this short summary below about attempts in Malawi and other countries in Africa to conduct randomized controlled trials of matching grant programs. Still, a study was commissioned by the project at its end to conduct a review of BUGS. That assessment is summarized in Annex 5, showing that the matching grant program became gradually accepted by MSMEs in Malawi (in the end over 1,100 cost-sharing grants were awarded). Given the limitations at that stage in attributing causality, the study focused on a customer satisfaction survey. Based on this self-assessment, 68 percent of 62 the firms considered the matching grant helped them improve skills, 47 percent of them to improve product quality, and 44 percent to increase sales. However, only 15 percent believed that BUGS had directly increased the income and employment levels of their businesses, while 28 percent thought there was no tangible impact yet, but felt more confident that things would improve. It is difficult to tell what these numbers mean in practice for an economic analysis and hence it is impossible with current information to rigorously compare the real impacts with the objective at the start95. Lessons for Matching Grant Programs from Failed Attempts to Evaluate Them96 A typical matching grant consists of a partial subsidy - most commonly covering 50 percent of the cost - provided by a government program to a private sector firm to help finance the costs of activities to promote exports, innovation, technological upgrading, the use of business development services, and, more broadly, firm growth. Matching grant programs are one of the most common policy tools used by developing country governments to actively facilitate micro, small, and medium enterprise competitiveness, and have been included in more than 60 World Bank projects totaling over US$1.2 billion, funding over 100,000 micro, small and medium enterprises. Yet despite all the resources spent on these projects, there is currently very little credible evidence as to whether or not these grants spur firms to undertake innovative activities that they otherwise would not have done, or merely subsidize firms for actions they would take anyway. Since firms self-select into whether they apply for such programs, and then the programs decide which applicants receive funding, attempts to compare outcomes for matching grant recipients to non-recipients are likely to be biased. Attempted Experiments We set out to design randomized experiments to prospectively evaluate seven matching grant programs in six African countries including Malawi. Five were to be supported through World Bank loans and technical support (including BESTAP) while two stemmed from a direct engagement with the government. In theory, matching grants satisfy a number of conditions that make randomization a possibility: i) they involve selection of individual firms; ii) the numbers of firms involved can be large enough to potentially generate enough statistical power for measuring 95 The objective in the M&E indicators and incorporated in the economic analysis was to grow turnover of firms supported by the matching grant by 20 percent when compared to a valid control group. 96 Campos, F., A. Coville, A. Fernandes, M. Goldstein, and D. McKenzie (2013). “Learning from the experiments that never happened: Lessons from trying to conduct Randomized Evaluations of Matching Grant Programs in Africa�. Journal of the Japanese and International Economies. forthcoming. 63 impacts, and iii) data on key outcomes may be measured reasonably well through firm surveys. Given that the government is effectively giving away free money to firms, one might expect significant demand for this funding, resulting in the need for projects to select which firms receive it. Since we believe there is substantial uncertainty over which firms would best benefit from receiving these funds, our suggestion was for randomized evaluation based on an oversubscription design. The idea here would be to make the matching grant programs open for all firms meeting certain basic eligibility criteria, and then randomly select which firms would be awarded the grants from among eligible applicants. What happened? Out of the seven projects that we discussed impact evaluations with, five initially agreed to implement projects with an oversubscription-based randomization experiment included, while the other two had encouragement designs planned. However, in practice, we were unable to implement any of the randomized experiments successfully. The main reasons were: 1. Lack of applications: despite giving away free subsidies, programs struggled to get enough applicants97, resulting in insufficient eligible applicants to randomize among. 2. Repeated delays and changes in personnel: implementation delays of over a year or more led to changes in government personnel, reversing some of the buy-in for the evaluation; it also meant we ran up against impact evaluation funding. Why is it so hard to give away free money?  Political economy and capture: these subsidies were viewed by some governments and partners as something to give their constituents: so we found Chambers of Commerce lobbying to keep eligibility conditions such that only their members would be eligible, while local governments competing with national governments in programs.  Overly strict eligibility criteria: requiring firms to be registered with audited tax accounts tended to exclude most firms in many African countries: criteria were set on the idea of which firms would grow fastest, not which firms would see the greatest benefit from the program. 97 In the case of BUGS, this was true at the time of the attempt to conduct the impact evaluation (first half of 2010), but the project got traction later on – after the impact evaluation had been dropped - to achieve its goal in number of activities supported. 64  Last mile issues and red tape: many of the conditions imposed by the World Bank and governments made it difficult and burdensome for even eligible firms to apply and receive money. Examples include: requiring firms to get letters from the tax department to show they were current on taxes; requiring firms to go through procurement processes like getting three bids in writing for services, and paying upfront for services with delayed reimbursement; and restricting what grants could be used for.  Incentives facing project staff: project staff was typically on fixed wage contracts, giving no incentive to try and get more applications. Implications Matching grant programs need to change the mindset from picking winners to picking positive treatment effects: the latter is even harder to judge, making impact evaluation even more important. Focus more on eligibility criteria and making it easy for firms to apply, and to get the money once they are awarded it. There are also several lessons for designing impact evaluations of such projects: these include using methods for small samples; having more realistic expectations on time frames; and conducting “little IE� or impact evaluation on different program design features. _______________________ Component 4 A critical aspect that is important to consider in this assessment is the cost of implementation and coordination. During BESTAP’s implementation, the risk of having a high number of implementing agencies was somehow controlled because the PIU was effective in supervising the activities of the different organizations. However, this came at a cost to the project. The PIU and M&E (Component 4), which had been budgeted at approval and restructuring to cost 15 percent98 and 19 percent of the envelope, ended up representing 27 percent of the project expenses, 35 percent of IDA funds. By restructuring, Component 4 represented 30 percent of the disbursements until then. Component 4’s expenses comprised US$2 million for fees to project staff, US$1.2 million for operating costs, US$625 thousand for equipment, US$600 thousand for PIU training, and US$550 thousand for coordination meetings. Three percent of the Component 4’s expenses were devoted to M&E (excluding the team’s specialist salary). Some of these costs99 concern the expenses of the team managing the matching grant 98 At design, it was thought the PIU would only last for the initial two years – and then the project would move to the MITPSD - but budget already considered more because by restructuring it was known that the PIU would stay for the overall project and the difference in budgeting was not so high. 99 The exact figure was not possible to obtain in preparation of this ICR. 65 scheme (BUGS) right after its early closure in 2011100. This followed completion of the cost-sharing grants in a period when the future of the scheme was still under discussion. Also, throughout the project there was a need to have accounting activities both in Blantyre and in Lilongwe, when initially it was thought that one accountant would be sufficient. In any case, these costs are considered high even if not uncommon to other projects in the region with large number of counterparts: for instance, the costs of the PIU for the Malawi’s Community-Based Rural Land Development Project (P075247) were 32 percent of the total budget. 100 After the completion of all the cost-sharing agreements, it was decided to have an early closure of BUGS. In the period between that closure and ending all commitments associated with BUGS, there were expenses that were secured by Component 4. 66 Annex 4. Bank Lending and Implementation Support/Supervision Processes (a) Task Team members Responsibility/ Names Title Unit Specialty Lending Constantine Chikosi Manager EACTF Sylvester Kofi Awanyo Lead Procurement Specialist EASR2 Yeshareg Dagne Program Assistant AFTFE Wedex Ilunga Senior Procurement Specialist AFTPE Esther Angellah Lozo Executive Assistant AFMMW Khwima Lawrence Nthara Senior Economist EASPT Gert Johannes Alwyn Van Der Lead Financial Management Spec AFTME Linde Dileep M. Wagle Consultant AFTFP Supervision/ICR Irene F. Chacon Operations Analyst AFTFW Ingrid R. Chikazaza Consultant AFMMW Grace Ingrid Chilambo Program Assistant IEGDG Simon B. Chenjerani Chirwa Senior Procurement Specialist AFTPE Yeshareg Dagne Program Assistant AFTFE Pauline Mbombe Kayuni Temporary EASPR Esther Angellah Lozo Executive Assistant AFMMW Samuel Munzele Maimbo Lead Financial Sector Speciali ECSF2 Steven Maclean Mhone Procurement Specialist AFTPE Francis Kanyerere Mkandawire Financial Management Specialis AFTME Brian G. Mtonya Senior Private Sector Developm AFTFE Rekha Reddy Economist LCSPF Gert Johannes Alwyn Van Der Lead Financial Management Spec AFTME Linde Dileep M. Wagle Consultant AFTFP (b) Staff Time and Cost Staff Time and Cost (Bank Budget Only) Stage of Project Cycle USD Thousands (including No. of staff weeks travel and consultant costs) Lending FY07 3.56 39.92 Total: 3.56101 39.92 Supervision/ICR FY08 17.95 86.75 FY09 23.88 108.51 FY10 30.68 147.69 101 The preparation of the project was likely linked to another code (including analytical work), hence the limited number of staff weeks using the lending code. 67 FY11 22.68 96.16 FY12 19.78 71.94 FY13 22.04 72.56 Total: 137.01 623.53 68 Annex 5. Beneficiary Survey Results Summary of Impact Assessment of BUGS on the Development of MSMEs in Malawi Salephera Consulting Ltd, June 2012 The Government of Malawi (GoM), through the Ministry of Industry and Trade (MoIT) has, since May 2008, been implementing a five-year World Bank-sponsored Business Growth Scheme (BUGS) within the framework of the Business Environment Strengthening Technical Assistance Project (BESTAP). The main purpose of BUGS is to facilitate maximum possible growth and to increase productivity within the private sector in Malawi by providing: (i) Direct technical assistance (ii) Matching grant assistance (iii) Grants for on-site HIV and AIDS counseling and testing This study was conducted with the underlying objective of assessing the impact of BUGS on the development of participating micro, small and medium scale enterprises in the country. A combination of approaches was employed as follows: (i) A survey of 158 business firms (with a sample of 108 beneficiary firms and 50 non-beneficiary firms) was conducted (ii) Interviews with 20 Business Development Service (BDS) providers (iii) Consultations with stakeholders (iv) Consultations with officials of BUGS, BESTAP and Ministry of Industry and Trade (v) Extensive desk study of key project and national documents Subsequent sections here-under summarize the study’s findings, conclusions and recommendations. Characteristics of BUGS Beneficiary Firms Business firms supported by BUGS operate in several sectors of the economy. Relatively, most of the firms supported are in the service sector (27 percent) and agriculture (23 percent); a few are in general trading (13 percent); a few (<10 percent) in education, manufacturing, associations and ICT; while very few firms (<5 percent) are in engineering/construction, tourism and mining/natural resources management. BUGS was mandated to focus on MSMEs, though some large firms also benefited in the process. The micro, small, medium and large enterprises respectively comprised 5 percent, 29 percent, 18 percent and 48 percent of the beneficiary firms. From these statistics, it is obvious that the large firms benefited the most which shouldn’t have been the case at all, given the stipulations of the project design whose beneficiary target was MSMEs. However, this was so because at the beginning of the project, MSMEs didn’t seem to have adequate individual financial capacity to meet their contribution requirements under the matching grant facility, such that this determined the small amounts that they could qualify for under the facility. 69 According to BUGS management, many MSMEs could only manage contributions as small as MK20,000.00 102 or MK30,0000.00 103 , which meant that individual and subsequently total financial disbursements would be very small vis-à-vis the total project funding that was to be disbursed. On realizing this, BUGS management decided to incorporate large firms as beneficiaries to the scheme as well in order to effectively meet their financial target. Nevertheless, management observed that MSMEs starting showing overwhelming interest over time such that given the trend, they should have exhausted the project funds by themselves within the stipulated project timeframe. However, BUGS had already invited large firms and couldn’t turn them back, having already committed to them. At the time of the assessment, most firms (64 percent) received support for training/workshops/seminars, a substantial proportion received support for market research (17 percent) and business plans/feasibility studies (14 percent), and very few firms (5 percent) received support for study tours. A number of firms that did not benefit from the scheme (non-beneficiaries) were randomly included into the study. The majority of the non-beneficiaries (86 percent) indicated that they were not participating in the scheme because they were not aware of its existence, and some (14 percent) did hear about the scheme but did not follow up further. Financial Performance of BUGS Beneficiary Firms Average annual sales revenue of beneficiary firms was MK 79.3 million104 from 2008- 2010 (3-year average at mid-term valuation). The average sales revenue graph of these beneficiary firms indicated an increasing trend, which is a promising indication of gradually increasing financial benefits as a result of BUGS. During the project period (2008-2010), beneficiary firms in the Southern Region of Malawi realized significantly higher sales revenue than those in the Northern and Central Regions, with those in the Centre realizing the least sales revenue. Annual sales revenue of beneficiary firms in the North, Centre and South averaged MK43.1 million, MK19.4 million and MK174.5 million respectively during this same period, resulting in an average total of MK79.3 million. Impact of BUGS on Business Performance: Beneficiaries’ Perception A substantial number of the beneficiary firms were not sure of the direct contribution of BUGS to the improvement of their businesses: only 15 percent believed that BUGS had directly increased the income and employment levels of their businesses; 35 percent believed their businesses had improved but could not attribute this directly to BUGS support; 28 percent thought there was no tangible impact yet, but felt more confident that 102 US$120 at the time of the intervention. 103 US$180 at the time of the intervention. 104 Circa US$480,000. 70 things would improve; 6 percent indicated their businesses had improved, but stated that they would still have improved even without BUGS support; 6 percent indicated BUGS support had no impact; and 11 percent did not give any response. All in all, there was general uncertainty among beneficiary firms of the contribution of BUGS support to the improvement of their business performance. A most likely cause for this uncertainty could be because BUGS had not collected performance indicators of these beneficiaries prior to BUGS support in order to compare those trends with the new indicators after BUGS support. However, most beneficiary firms (62 percent) were satisfied/very satisfied with the performance of their businesses; while very few (5 percent) were disappointed/very disappointed with the performance of their businesses. Business Outcomes as a Result of BUGS Interventions Inasmuch as beneficiaries are not certain with regard to BUGS contribution to the positive shifts in their business activities (as noted above), it however becomes apparent from the survey responses that the support provided by BUGS has yielded important outcomes for some beneficiary firms. According to the survey responses, BUGS support translated into improvement of skills (68 percent of beneficiary firms), improvement of product quality (47 percent), increased sales (44 percent), increased customer care (29 percent), increased advertisement (22 percent), new markets (16 percent), cooperation with other enterprises (16 percent), substantial business expansion (16 percent), broader product range (14 percent), investment technology (9 percent), increased exports (5 percent), increased outsourcing of inputs (4 percent), different enterprises (4 percent) and reduced outsourcing of inputs (1 percent). A notable outcome was the improvement of skills arising from training activities. These skills were reportedly relevant and used on a daily basis, thereby empowering the beneficiaries in the process. The vast majority of large beneficiary firms were satisfied with the 50-50 cost-sharing arrangement, and very few of them were asking for an alternative proportion. The vast majority of these large beneficiary firms (96 percent) expressed interest to use BUGS again. On the contrary, however, the majority of micro and small firms desired a higher ratio figure on the grants side, a preferable ratio being a 75:25 cost-sharing arrangement. Analysis of BUGS Activities Scheme Management The initial thrust of the BUGS implementation process was hindered by a delay in recruitment of key scheme personnel, such as the second BUGS advisor and Accountant who were recruited one and two years respectively after project inception. This initial incapacity, coupled with uncertainty amongst Malawian SMEs about BUGS services, affected the scheme’s opportunity to capture the trust and confidence of the market from inception. 71 Generally, beneficiary firms were positive about the management and delivery mechanism of the scheme. The period it took BUGS to process and respond to applications was considered effective. Furthermore, BUGS provided the relevant information requested by clients usually within a 10 working days period, which was also considered effective. The approval process and communication was also thought to be effective. However, a number of firms felt that the application process could omit common information already in the hands of BUGS from previous applications when applying for assistance a second or third time. Strengths, Weaknesses, Opportunities and Threats A main strength of BUGS was the independence it had from the project implementation unit and the Ministry with regard to operations. BUGS was given the liberty to plan, implement, monitor, implement and review operations, a significance that made it possible for them to undertake processes, such as application approvals and financial reimbursements, with effective speed. BUGS’s focus was to place the client first, by giving priority to a fast and efficient service and by cutting down on bureaucracy. The Scheme also realized the relevance of supporting micro-enterprises. The only constraint is that this category was very limited with regard to the financial contributions that it could make under the 50-50 matching grant, such that it couldn’t significantly take advantage of facility in such a way as to make an impact on their individual enterprises. Nevertheless, the fact that micro-enterprises comprise the bulk of the informal sector in Malawi (Ministry of Industry, Trade and Private Sector Development, SME section) cannot be dispelled. According to the Ministry, this category, if properly guided and deliberately developed, can effectively provide the impetus for potential national growth and competitiveness. Adopting such a deliberate strategy which positions BUGS to deliberately guide and develop micro-enterprises (and SMEs as well) through a series of graduation processes to finally bring them into the formal sector at a well-organized level, would not only develop Malawi’s business and industrial capacity, but would also develop it from an indigenous view point. Furthermore, BUGS would become the primary planning tool for business development in Malawi and would be the specialist entity with regard spearheading MSMEs development, working hand in hand with government and the private sector. The Scheme has the potential to broaden its clientele base even presently because of its proven delivery efficiency, speedy response to requests, and commitment to achieving its objectives, therefore enabling it to quickly build credibility. For example, according to its plans, BUGS had targeted to achieve 1,000 clients. But because of their aggressive stance and the market demands for the service, they achieved 1,600 clients in the same time-frame. Another strength of BUGS was its flexibility to reduce the claims threshold from 50 percent to smaller amounts such as 10 percent, as long as invoices were submitted, thereby enabling SMEs take advantage of the matching grant at affordable terms, though 72 this didn’t reduce the 50 percent total contribution requirement. Beneficiaries who were assessed as to be struggling, but whose projects had potential and viability, were allowed to claim several times, an administrative arrangement designed to allow fluidity in their cash flows. One of the major weaknesses of BUGS was the inability to follow up on clients in districts in the three regions where it had launched its program and conducted workshops. After creating the awareness, BUGS did not undertake monitoring and aftercare services as it should have. This would have been an opportunity for it to identify key operational constraints amongst beneficiaries, undertake impact assessments and further support requirements for beneficiary firms. Nevertheless, there were some constraining factors for BUGS as well whose roots are in the program design. For example, their very limited human resource outlay meant that an officer had to strategize, go out and find clients, administer applications, process them, disburse funds, monitor and review operations. It’s therefore understandable, with a view to the limited human resource vis-à-vis the given workload, that they just couldn’t cope with providing aftercare services particularly to far-away places. There were opportunities for BUGS to create networks and synergies with various organizations whose objectives were poverty alleviation and capacity building, which were the ultimate over-arching objectives of the scheme. In both urban and rural areas, BUGS had opportunities to synergize with District Development Committees, District Councils, MASAF, business associations, cooperatives and other SMEs support structures/institutions such as NABW, NASME, DEMATT and financial institutions in order to create a self-sustaining, far-reaching and an integrated approach toward business development and capacity building of the private sector in Malawi, insofar as they couldn’t personally reach some of these places due to their limited human resources (limited by the project design). Creation of local networks was critical for the sustainability of the scheme. A critical threat for the current and future of BUGS is the uncertainty on how the scheme will continue to be funded. At the time of this assessment, funds for new grant agreements were unavailable despite the fact that the scheme’s performance during the period of the assessment was the highest it has been since the program commenced four years ago. Relevance, Efficiency, Effectiveness, Impact and Sustainability of BUGS Relevance One of the goals of the Malawi Growth and Development Strategy 1 (MGDS 1) was to make the private sector the engine of economic growth in the country. The BUGS aimed to facilitate maximum possible growth and to increase productivity of private sector firms. Consequently BUGS has, in a way, contributed to the development goals of GoM. From a perspective of enterprises, BUGS was highly relevant. The Scheme was quick and responded to BDS demand of MSMEs in the country. BUGS triggered reasonably positive effects on firms’ growth and facilitated the creation of an environment for the generation of employment (See Chapter Six). 73 Efficiency Efficiency of BUGS was assessed in terms of cost efficiency, operational efficiency and synergy efficiency. The grants allocation was US$2.0 million which included US$0.2 million for the HIV element, while the total budgeted operating costs for the Scheme were $2.05 million. As at 31st May 2010 (mid-way of the project life) the scheme had disbursed a total of US$475thousand which represented 24 percent of the total grants allocation. BUGs had slower speed of progress at its start because of the constraints it faced in its inception stages, such as limited human resource, lack of adequate publicity (because there was no publicity budget in the design), among others. Nevertheless, businesses were becoming aware of the advantages and benefits of BUGS support, such that the demand for its services were increasing. Consequently, by January 2012, BUGS had dispensed 62 percent of the grant allocation. From the side of operational expenditures, the amount disbursed by 31st May, 2010 was $1.168 million representing 58 percent of the operational budget. The figure of operating costs includes all vehicles and equipment for the scheme. As indicated, BUGS had disbursed only 24 percent of the matching grant mid-way through the project (May, 2010). This expenditure level though above 50 percent of the total operational budget at its mid-term of operations, would be considered prudent at this level, particularly considering the fact that BUGS had to undertake certain necessary cost activities such as advertising and publicity, which were not included in the design. This explains the relatively higher levels of expenditure on the operational budget during this same period. Secondly, another major reason for the low disbursement of monies – further to the constraints mentioned above- was the bureaucracy involved for BUGS to access the monies from BESTAP for both operations and grants. BUGS indicates that it would always have to make request for funds to BESTAP, and BUGS always faced delays in receiving the funds, which also affected their business; both administrative and operational. Effectiveness Effectiveness of BUGS services was assessed with regard to results to final beneficiaries, BDS service providers, and on institutional development and networking. Beneficiary firms were satisfied with the way services were delivered. In several cases, BUGS sensitized SMEs to the value of good consultancy, in a way that such firms continued to contract services on their own after having seen the value of it. Impact BUGS had only operated for 2.5 years when this assessment was done. As such, the impact of the scheme could not be effectively measured at this point, particularly with the view that the project had take-off constraints during its inception period (elaborated 74 above) such that optimal delivery of services delayed substantially. Nevertheless, beneficiary firms grew in employment (31 percent), investments (109 percent) and turnover (62 percent) during this same period. While these indicators could not be fully attributed to the BUGS support directly, the general positive perception of the firms’ beneficiaries was good enough indication that the contribution of BUGS was felt. Positive changes in employment, investments and turnover had direct positive effects on the socio-economic livelihoods of the entrepreneurs’ families as well as the employees. As firms grow, their contribution to taxation also grows, thus oiling the engine of growth for the Malawi economy. BUGS services had certainly facilitated sales growth among beneficiaries- sufficient evidence for this was received during the field work. The assessment found that 54 percent of beneficiaries’ businesses had grown by 55 percent from 2008 to 2010. In terms of employment and capital investment, BUGS beneficiaries projected that the two factors would generally grow by 85 percent and 9 percent in the next three years respectively. Had BUGS been in a better position to link recipients with investment finance, the impacts might have been much greater. The positive assessment of impacts was a strong point for BUGS. BUGS supported some large scale businesses that needed initial support to develop bankable business plans one of which received a loan of K150 million for a hotel construction project in Blantyre. This is an example of a significant BUGS impact in supporting start-up stages of large scale businesses. Sustainability Sustainability of BUGS heavily hinged on the availability of funds from development partners as well as the ability of the beneficiary firms to achieve sustained growth. BUGS’s dependence on donors was obviously not sustainable. However, if its services could generate growth among the beneficiary firms, the skills gained will enable those beneficiaries to grow. Integration of MSMEs into both the formal national and world markets required support services that by their nature cannot be sustainably provided by the beneficiaries themselves. BUGS should therefore be regarded as a role model for the delivery of such services in Malawi. A sustainable exit route for BUGS would have been to prepare Government of Malawi to regard subsidies as an indispensable factor of a competitive MSME promotion system, and to help them establish and manage such a system. This requirement called for intensive coordination and cooperation with the World Bank, GoM and similar programs of other donors - directed at both the micro and macro levels. It should also support policy dialogue. BUGS' experience in raising competitiveness should be advanced in this dialogue. Conclusions The following were the lessons drawn from the study: 75 1. There is evidence that the Scheme contributed to the MGDS goal of economic development through wealth and job creation by promoting SMEs and, therefore, was in line with the GoM national development policies and those of the World Bank; 2. BUGS increased the capacity of MSMEs to use BDS consultancy services to a considerable level especially among medium and large-scale enterprises, and played an instrumental role in helping most MSMEs in accessing BDS; 3. As a result of the activities of BUGS it was concluded that there was demand for BDS support services in Malawi; 4. The 50 percent cost share grant concept, which was fairly new, was gradually being accepted by some MSMEs that were previously demanding higher reimbursement percentage; 5. There was urgent need to invest in developing skills in the BDS industry (business consulting skills); 6. The Scheme registered low grant uptake in the first two years of implementation due to lack of awareness among MSMEs and the longer than planned time that it took BUGS to be fully operational; 7. BUGS lacked capacity in monitoring progress and impact of support resulting in the scheme operating without timely feedback from beneficiary firms. This was obviously a deficiency of the project design because it did not provide for a budget line on aftercare services; 8. Operationally, BUGS demonstrated good operational efficiency through professionalism and fast turn-around time from applications and disbursements. However, inadequate staffing continued to pose a challenge to BUGS operations as demonstrated by the failure to effectively monitor progress of supported activities; 9. The scheme was well-designed in terms of operational proficiency, but its design ignored the promotional component particularly advertising through print and electronic media on regular basis. This again was due to exclusion of a budgetary line on publicity and promotion in the design. This exclusion resulted in the scheme missing enormous opportunities to reach more MSMEs. Recommendations In view of the foregoing lessons learnt, the following recommendations are presented as a strategic way forward: 1. The next program should expand outreach to district level in collaboration with district councils in order to include participation of the MSMEs that are located away from the major towns and urban centers. 2. On monitoring and evaluation of the scheme’s activities, the assessment team presents the following options: a.The scheme needs an M&E Officer, specifically for the scheme. In addition, the scheme needs to develop a good M&E system. In order to establish this, suitable indicators measuring impact need to be formulated, together with methods of information collection that do not cost very much and that can be organized without effort. b.Maintain the current staffing but engage consultants to help with other tasks such as monitoring. 76 3. Next project design should include a baseline study to be conducted at the beginning of the project in order to establish benchmarks that would be used in establishing targets and determining indicators for monitoring and evaluation. 4. The scheme needs to develop a more comprehensive package for MSME support other than support for BDS consultancies only. The package could include finance, technology, equipment, skills, bid security or equity contribution to SMEs accessing finance from the banks. 5. The evaluation team strongly recommends that government should lobby for and mobilize support and partners to establish a development bank or fund. This bank or fund will then receive bankable project proposals from SMEs that were supported by BUGS. The development bank or fund should offer an integrated and complete support process, from the diagnostics level up to financing the MSME. 6. MSMEs supported by BUGS and with growth potential should be deliberately drawn out of the informal sector and guided and grown into the formal sector in a systematic manner. This, along with business performance monitoring and evaluating, can be an on-going role of BUGS. 7. There may be a need to vary the percentage contribution that the scheme can make in the light of the size of the enterprise. With, for example, matching grant reimbursement percentage being larger for micro and smaller firms. 8. Any future BUGS project should only focus on BDS and matching grants and not include an HIV and AIDS components of the nature that was in the implemented project. However, in the event that the HIV/AIDS component needs to be incorporated, then it will need to be redesigned. Through subcontracting arrangements, the funds could be channeled to organizations that have expertise in the management of HIV/AIDS programs. 9. The implementation of the scheme’s interventions needs to be guided by deliberate strategic planning for systematic achievements. 10. Physical presence should be established in the northern region of Malawi in order to improve the outreach function of BUGS. This will also lead to efficient operations as the need for staff to travel from Blantyre to Mzuzu will be eliminated unless otherwise determined from time to time. 77 Annex 6. Stakeholder Workshop105 Report and Results In preparation for this evaluation, the team conducted stakeholder workshops with all implementing organizations. Approximately 20 government officials participated in this workshop including the Project Coordinator (PIU) and one representative of the MITPSD. In the days that followed this workshop, the evaluation team met individual with the implementing agencies, as well as other donors, private sector, and other stakeholders. This was the agenda of the stakeholder workshop (replicated in Blantyre for agencies with headquarters in Blantyre): World Bank’s Implementation Completion and Results Report BESTAP – Business Environment Strengthening Technical Assistance Project Consultation Workshop Lilongwe, 19 February 2013 Agenda 9:00-9:10 Introductory remarks MoIT 9:10-9:30 Introductions and objectives Shadreck Ulemu, Francisco Campos, and Efrem Chilima 9:30-10:30 Feedback form Francisco Campos 10:30-10:45 Coffee break 10:45-11:30 Presentation of project and overview of results Shadreck Ulemu 11:30-12:30 Open discussion on specific topics Francisco Campos 12:30-13:45 Lunch 13:45-15:00 Continuation of open discussion on topics Francisco Campos 15:00-15:15 Coffee break 15:15-16:00 Continuation of open discussion on topics Francisco Campos 16:00-16:15 Next steps Francisco Campos 16:15-16:30 Closing remarks MoIT, Shadreck Ulemu and Efrem Chilima The workshop comprised detailed open discussion about the objectives, achievements, and problems of BESTAP, including written feedback forms shared only with the evaluator. The results of this feedback are summarized in the following graphs. The 105 The Stakeholder workshop was organized together with Efrem Chilima, Senior Private Sector Development Specialist in the Lilongwe office. 78 graphs reflect the opinion of the interviewed counterparts in the government, not necessarily of the evaluator or the institutions they belong to. Initial design of activity subcomponent is according to the needs back then 12 3 3 Yes Partly No Timing of implementation of subcomponent 9 8 1 Fast Satisfactory Slow Implementation of subcomponent 10 6 2 Fully Partially Not implemented implemented implemented (the workshop did not include representatives from two subcomponents that were dropped at restructuring) 79 Scope of subcomponent relevant during implementation 13 4 Just as it was Increase in scope Subcomponent efficient (mostly based on cost) 8 4 Yes No The feedback forms that were completed before the open discussion included the following questions: Design 1. Was the initial design of your component (in 2007) appropriate to the needs of Malawi back then? 2. How would you have designed your component differently knowing what you know now? 3. Do you think this component should not have been included in the project? Why/Why not? Implementation 4. Timing of implementation: was it satisfactory, slow, or fast? Why? 80 5. Scope of implementation: was the design of your component implemented fully, partially, or not implemented? Why? 6. Do you think the project should have considered during implementation changes to your component? Which ones – increase in scope, decrease in scope, deletion? Of which type 7. Do you think your component could have been done more efficiently? At a cheaper cost? Why/Why not? 8. How was the World Bank performance during the implementation of your component? What went well and what were the problems? 9. How did the implementation of your component go? What went well and what were the problems? Final feedback 10. What were the main results / impacts of your component of the project? 11. What were the main lessons of implementing this component of the project? 12. How would you have done your component differently? 13. What are the main takeaways of your component for other WB projects in the future? In the feedback forms, the government officials also provided individually the following suggestions / recommendations: Design:  Need for better assessment and scope of the project;  More involvement of beneficiaries in design stage of the project;  Incorporate in project design more training of officials, get more placements, specialized rather than general training; more capacity building beyond training;  Have only one donor per project to avoid problems of coordination; Implementation:  Coordination between implementing agencies is very important;  Need to publicize the results; high-level sensitization through stakeholder public awareness consultation, workshops, use of media;  Programs like BUGS should be all over the country in order to succeed;  Minimize changes in the Bank team especially in the TTL role;  Reduce time to process non-objections;  Some activities take time and so the project should be longer or be followed by a new one; 81  Include in the project a Customer Service Center to respond to queries;  Reduce the number of contracts with (international) consultants;  Risks of the project team assuming that additional financing will arrive;  Government needs to be bold to deal with slow moving components;  Need for a champion at the higher level for implementation;  Need to identify mechanisms of staff retention for ministries that have benefited from valuable specialized training; In addition, the open discussions during the workshops provided additional insights into the achievements and problems at all stages of the project that informed this evaluation. 82 Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR Summary of Borrower’s ICR: The Malawi Growth and Development Strategy (MGDS) plans to achieve 6 percent economic growth - the minimum necessary to have a meaningful impact on poverty levels in Malawi. For this growth to be achieved, it is essential that the private sector invests and creates jobs. The World Bank program under the Country Assistance Strategy (CAS), was fully aligned with the country’s MGDS and focuses its support in four key strategic areas: (i) improving smallholder agricultural productivity and integration into agro-processing; (ii) putting in place a foundation for longer term economic growth through improved infrastructure and investment climate; (iii) reducing vulnerability at the household level, especially from HIV/AIDS and malnutrition; and (iv) sustaining improvements in fiscal discipline, budget execution, and accountability of the civil service. The BESTAP project was an integral part of the CAS and contributes to achieving Malawi’s development strategy by addressing the constraints to PSD related to the business environment, inadequate enterprise skills and institutional support to PSD. Key constraints that were the driving force into the development of the BESTAP included: (a) The poor regulatory environment for business which impedes investment and constrains PSD; (b) Weak contract enforcement and limited access to commercial justice; (c) Low capacity in property rights institutions prevents the private sector from unlocking the value of fixed capital; (d) the process of registering a company or business name is subject to ineffective single centralized and manual system that was based in one location; (e) A cumbersome business licensing regime; (f) Weak policy analysis, formulation and implementation creating uncertainty in the private sector and; and (g) Limited access to finance. The project had four main components namely: (i) Strengthening Property Rights Institutions and Business Facilitation; (ii) Strengthening Private Sector Development Support Institutions and Services; (iii) Promoting Access to Finance and Productivity; (iv) Capacity Building and Implementation Support. Under component one; the project embarked on drafting 38 prioritized Economic Laws. This was done to reduce a backlog of economic laws which were out dated and needed wholesale revision. The project hired a full time experienced drafts person who was dedicated to reviewing these selected laws in order to enhance the doing business environment. Additionally, a High Court Commercial Division (HC-CD) was established with two divisions (in Blantyre and Lilongwe). Both Court officials have been trained and are able to undertake resolving of commercial disputes in 96 calendar days as compared to 377 calendar days at the onset of the project. The Commercial courts are hence fully functional. Besides the regulatory environment, the project also aimed at improving capacity in property rights institutions in order to enhance capabilities of unlocking the value of fixed 83 capital. Two major reforms that were undertaken were the computerisation of the Land registration process in all regions of the country and the computerisation of the business registration process at the Department of Registrar General (DRG). Both systems have been installed and commissioned. Under component two, Strengthening Private Sector Development Support Institutions and Services, the project had four sub-components. It aimed at building the necessary capacity in institutions that provide essential services to the private sector in order to improve quality and volumes of service delivery and also to provide policy direction on private sector development issues. The project planned to merge the Malawi Investment Promotion Agency and Malawi Export Promotion Council (MEPC) into the Malawi Investment Trade Centre (MITC). The merger only took place in 2011 and the MITC bill was passed in 2012. The initial target was therefore to reduce the number of days it takes to issue out licenses from 185 to 30 calendar days. However, there was not much that was done under the subcomponent as the MITC is not fully functional hence the indicator was at 200 calendar days in the 2012 DBS report. In addition, the Private Public Partnership (PP) was established to enhance the doing business environment and the PPP Act was approved by June 2012. The PPP is being implemented by the Privatisation Commission. Under component two, the project also embarked on building capacity of the Ministry of Industry and Trade (MITPSD). Several short courses were attended as well as two master degree level courses were attained. Additionally, a Doing Business Unit was also established at the MITPSD to handle all the reforms that the Ministry has been facilitating with the assistance of the World Bank Doing Business team. Under component two, the project provided support for the establishment of and operations of the Public Private Dialogue (PPD) secretariat under the MCCCI. This forum was a success as the private sector was provided with a podium to discuss matters that were crucial to the growth of the industry in the country. A minimum of two forums were held every year and they were attended by high level government officials at Minister Level. Most resolutions that were made at these meetings were followed up by the secretariat until they were implemented. Under component three which aimed at improving access to finance, only one subcomponent was pursued and this was the Business Growth Scheme (BUGS). The other two subcomponents were discontinued at midterm as a result of policy changes by the GoM. The BUGS was established to facilitate maximum possible growth in the private economic activity through offering direct technical assistance to firms and/or; part fund (50-50 matching grant) the use of outside expert or specialist. Business was slow at the beginning of the Scheme so much so that by September 2010, the scheme had used about 30 percent of the total allocation for grants but it had signed almost 735 cost share grants. However, by 2011 the scheme had gained popularity and more businesses were now aware of its existence hence the disbursement improved tremendously. By December 2011, the scheme had signed 1,134 cost share grant agreement and re- imbursement was at 63 percent totaling MK184, 014,092 (US$1,250,907). 84 Overall, BUGS implementation has been satisfactory overall, over 1,300 cost share agreements had been signed out of a target of 1000. However a study conducted in 2011 by an independent consultant was unable to establish causality and actual changes in percentage growth between firms that have received assistance from BUGS and average enterprises in the economy. Nevertheless, results showed that there was an increase in sales revenue thus 44 percent of the firms, 68 percent improvement in skills, 47 percent improvement in quality, 29 percent improvement in customer care among other achievements. General project implementation has been Satisfactory. Most of the project rating was Satisfactory save for an Implementation Support Mission (ISM) that was conducted in January 2009 which rated the project “Moderately Satisfactory� resulting from two subcomponents under component two that were yet to commence implementation. This resulted into project restructuring whose aims were; (a) to change the PDO; (b) dropping two components that were idle and the introduction of two new activities (the review of the Companies Act and improving the permit section of Immigration Department and; (c) reallocation of funds to support the two new activities that were added to the project. At the beginning of the project, the Doing Business Study (DBS) ranked Malawi number 110 out of 175 countries in terms of the overall enabling environment for business. It was costing 134.7percent of GNI per capita to formally start a business. The project planned to reduce this cost to 70 percent of GNI per capita. According to the DBS; the cost was 188.7 percent in 2008, 125 percent in 2009, 108 percent in 2010, 108.4 percent in 2011 and 90.9 percent in 2012. Even though the cost trend has reduced over the years, it is still above the average Sub-Saharan Africa cost ceteris paribus. Nevertheless, BESTAP cannot claim 100 percent causal linkages to the drop but it is clearly understood that the project reforms have contributed to the drop. Financial management was prudent throughout the project life time. For the IDA funding, the project cumulatively utilized US$15,444,321.56 by June 2012 and the available balance from the Grant was US$1,597.72 which represented a 99.99 percent disbursement rate for the IDA resources. For EC funding, the project has disbursed US$3,586,316.73 representing 96.85 percent and the balance was US$113,533.83. Several challenges were experienced during the five years and some of the most notable ones were: � the depletion of IDA funds before project completion date which affected implementation of activities that are still in progress; � lengthy procurement procedures; � utilization of EU part funding took longer due to some administrative hitches beyond the control of the project; � delays in government giving direction on two components that were later discontinued affected the implementation of these components and also overall performance of the project before midterm; 85 � Low absorption of BUGS grant fund by SMEs in the first three years of the project life � Delays in the development and implementation of an all-inclusive communication strategy at the MoIT which hindered public awareness of most of the reforms that have been done in the project. 86 Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders The European Commission, which co-financed BESTAP, provided the following comment on the draft ICR:  Thanks for the report which is well written. It appears that our Finance and Contracts Section does not have comments at this point in time. One observation on the investment made through BESTAP vis a vis sustainability of the initiatives: ICT systems at the Registrar of Companies, Lands Registry among others. My reading of the report (especially page 35) indicates that the performance of Implementing Agencies was rated as Moderately Satisfactory. How committed are the agencies to keep the systems up and running? Are there long term maintenance plans or assurances from GoM that they will continue to maintain the systems? I know that these are side issues but critical to the objectives of BESTAP. 87 Annex 9. List of Supporting Documents 1. Project Appraisal Document, April 27, 2007 2. Financing Agreement 3. Administration Agreement with European Commission, 2008 4. Grant Agreement with European Commission, 2011. 5. Restructuring Paper 2010 6. Restructuring Paper 2011 7. Restructuring Paper 2012 8. Aide Memoires and Implementation Status Reports – 2007-2012. 9. Impact Assessment of the Business Growth Scheme (BUGS), Salephera Consulting Ltd 10. Learning from the Experiments that never happened: Lessons from trying to conduct Randomized Evaluations, 2012. 11. Business Registration Impact Evaluation Concept Note. 2011. 12. Government of Malawi ICR, 2013. 13. Doing Business Reports – 2007-2013. 14. World Bank Country Assistance Strategy FY13-FY16. 88 Annex 10. Other relevant information a) Doing Business Indicators for Malawi (2007-2013) 2007 2008 89 2009 2010 2011 90 2012 2013 91 b) Description of Afdb’s Competitiveness and Job Creation Support Project (2012-2017) 92 Annex 11. Figures Figure 10 – Project’s summary theory of change (as approved) Figure 11 – Project’s summary theory of change (as restructured) Figure 12 – Time and Cost to register a business (DB 2013) 93 Table 8: Steps in registering a business in Malawi Procedure Processes Agency BESTAP How? involvement 1 Initiate a company Department of Registrar’s Yes IT system at name search. General, Ministry of Justice DRG 2 Submit application Department of Registrar’s Yes IT system at for a Certificate of General, Ministry of Justice DRG Incorporation 3 Register for payment Malawi Revenue Authority of income tax 4 Obtain a company Yes Business seal registration bill106 5 File an application City Assembly Yes Business license form to obtain a bill107 license 6 Inspection of City Assembly Yes Business license premises for the bill108 issue of the license 7 Pay license fee upon City Assembly approval of license and obtain license 8 Apply for a Occupational Safety, Health, registration of the and Welfare Department of workplace the Ministry of Labor 9 Inspection of Occupational Safety, Health, premises and Welfare Department of the Ministry of Labor 10 Register for PAYE Malawi Revenue Authority and fringe benefit tax by mail 106 It will not be compulsory from now onwards. 107 Proposal to reduce from 29 days to 7 days. 108 Inspections are proposed to be done randomly rather than to every firm. 94 IBRD 33440R1 32°E 34°E 36°E To To Song we Mbeya TANZA TA NZA NI NIAA Tunduma Chitipa MALAWI Karonga 10°S Chisenga 10°S To Muyombe Nykia Chilumba (2,606 m) Chelinda Mkondowe To Muyombe Livingstonia Katumbe Rumphi Ruarwe s. Kafukule Mzuzu Mtn Euthini Nkhata NORT HERN NORTHERN Bay ZAMBIA a Viphy Mzimba Chinteche (MALAWI) Lake Mala 12°S 12°S Luwawa To Lundazi Nkhunga wi Kaluluma Nkhotakota Kasungu MOZAMBIQUE Ntchisi C E N T R A L Dowa Makanjila Bua To Mchinji Chipata Salima 14°S LILONGWE 14°S Namitete Monkey Bay To To Cuamba Furancungo Dedza Mangochi To Ulongwe SOUTHERN To Cuamba Ntcheu Balaka MOZAMBIQUE 32°E Machinga ire Sh Lake Neno Chilwa MALAW I To Lirangwe Zomba Tete Mwanza Chiradzulu CITIES AND TOWNS Blantyre Phalombe DISTRICT CAPITALS* 16°S Chikwawa Mulanje Sapitwa 16°S (3,002 m) REGION CAPITALS Thyolo To NATIONAL CAPITAL Liciro N’gabu RIVERS To Morire MAIN ROADS RAILROADS Nsanje 0 20 40 60 80 100 Kilometers DISTRICT BOUNDARIES REGION BOUNDARIES This map was produced by the Map Design Unit of The World Bank. 0 20 40 60 Miles The boundaries, colors, denominations and any other information To INTERNATIONAL BOUNDARIES shown on this map do not imply, on the part of The World Bank Vila de Sena *District names are identical to the District Capitals. Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. 36°E MAY 2012