Completion Report. Project Number: Loan Number: 1990 October Viet Nam: Housing Finance Project

Transcription

1 Completion Report Project Number: Loan Number: 1990 October 2013 Viet Nam: Housing Finance Project

2 CURRENCY EQUIVALENTS Currency Unit dong (D) At Appraisal At Project Completion 15 November August 2011 D1.00 = $ $ $1.00 = D15,369.5 D20,628 ABBREVIATIONS ADB Asian Development Bank CCF Central People s Credit Fund CFI community-based financial institution HFF housing finance facility ICPMU International Credit Project Management Unit MIS management information system MOF Ministry of Finance MRA mortgage refinance agency PCB participating commercial bank PFI participating financial institution SBV State Bank of Viet Nam ULIH urban low-income household ULIPH urban low-income and poor household UPH urban poor household NOTE In this report, $ refers to US dollars. Vice President S. Groff, Operations 2 Director General J. Nugent, Southeast Asia Department (SERD) Director S. Hattori, Public Management, Financial Sector, and Trade Division, SERD Team leader Team members H. Aoki, Senior Financial Sector Specialist, SERD M. Mina, Project Analyst, SERD K.M. Sanchez, Operations Assistant, SERD In preparing any country program or strategy, financing any project, or by making any designation of or reference to a particular territory or geographic area in this document, the Asian Development Bank does not intend to make any judgments as to the legal or other status of any territory or area.

3 CONTENTS Page BASIC DATA I. PROJECT DESCRIPTION 1 II. EVALUATION OF DESIGN AND IMPLEMENTATION 2 A. Relevance of Design and Formulation 2 B. Project Outputs 3 C. Project Costs 5 D. Disbursements 6 E. Project Schedule 6 F. Implementation Arrangements 7 G. Conditions and Covenants 8 H. Related Technical Assistance 9 I. Consultant Recruitment and Procurement 10 J. Performance of Consultants, Contractors, and Suppliers 10 K. Performance of the Borrower and the Executing Agency 10 L. Performance of the Asian Development Bank 11 III. EVALUATION OF PERFORMANCE 11 A. Relevance 11 B. Effectiveness in Achieving Outcome 11 C. Efficiency in Achieving Outcome and Outputs 12 D. Preliminary Assessment of Sustainability 12 E. Impact 12 IV. OVERALL ASSESSMENT AND RECOMMENDATIONS 13 A. Overall Assessment 13 B. Lessons 14 C. Recommendations 15 APPENDIXES 1. Project Framework Gender Action Plan and Outputs Status of Compliance with Loan Covenants Policy and Institutional Action Plan 24 SUPPLEMENTARY APPENDIXES A. Revisions of Project Criteria 30 B. Disbursement 32 C. Actual Financing and Disbursement 33 D. Project Financing Mechanism 36 E. Comparison of Costs (Original vs. Actual) 37

4

5 BASIC DATA A. Loan Identification 1. Country 2. Loan Number 3. Project Title 4. Borrower 5. Executing Agency 6. Amount of Loan 7. Project Completion Report Number Viet Nam 1990 Housing Finance Project The Socialist Republic of Viet Nam State Bank of Viet Nam SDR million ($30.0 million) 1425 B. Loan Data 1. Fact-Finding Date Started Date Completed 2. Loan Negotiations Date Started Date Completed 3. Date of Board Approval 4. Date of Loan Agreement 5. Date of Loan Effectiveness In Loan Agreement Actual Number of Extensions 6. Closing Date In Loan Agreement Actual Number of Extensions 7. Terms of Loan Interest Rate Maturity (number of years) Grace Period (number of years) 8. Terms of Relending (if any) Interest Rate Maturity (number of years) Grace Period (number of years) 14 Oct Oct Nov Nov Dec Jul days after 31 Jul 2003 or 29 Oct Jan Jun Aug 2011 (physical completion) 3 May 2012 (financial closing) 3 1% p.a. during the grace period, and 1.5% per annum thereafter 32 8 Quarterly weighted average interest rate of all time deposits in the Viet Nam banking system 20 5

6 ii 9. Disbursements a. Dates Initial Disbursement 30 Aug 2004 Effective Date 6 Jan 2004 Final Disbursement 11 Jan 2012 Original Closing Date 30 Jun 2008 Time Interval 88 months Time Interval 54 months Category or Subloan b. Amount Original Allocation Last Revised Allocation Amount Canceled Net Amount Available Amount Disbursed Undisbursed Balance A. Credit SDR 16,754, SDR 16,115, SDR 16,115, USD 22,400, USD 24,900, USD 24,900, B. Credit SDR 3,740, SDR 1,343, SDR 1,343, USD 5,000, USD 2,010, USD 2,010, C. Non-credit SDR 1,944, SDR 1,231, SDR 1,231, USD 1,700, USD 1,831, USD 1,831, D. Interest on loan SDR USD 900,000 Total SDR 22,438,000 SDR 18,689, SDR 3,748, SDR 18,689, SDR 3,748, USD 30,000,000 USD 28,742, USD 1,257, USD 28,742, USD 1,257, Local Costs (Financed) - Amount ($) 26.9 million - Percent of Local Costs - Percent of Total Cost 44.5

7 iii C. Project Data 1. Project Cost ($ million) Cost Appraisal Estimate Actual Foreign Exchange Cost Local Currency Cost Total Financing Plan ($ million) Cost Appraisal Estimate Actual Implementation Costs Borrower Financed ADB Financed Counterpart Funding (Beneficiaries) Other External Financing Total IDC Costs Borrower Financed ADB Financed Other External Financing Total ADB = Asian Development Bank, IDC = interest during construction. 3. Cost Breakdown by Project Component ($ million) Component Appraisal Estimate Actual Component A ADB Counterpart Component B ADB Counterpart) Component C ADB Counterpart Interest on Loan ADB Counterpart Nil Nil Total Project Schedule Item Appraisal Estimate Actual Project physical completion date 30 Jun Aug 2011 Project financial closing date 3 May 2012 Date of contract with consultants 6 Apr Sep 2004 Completion of work by consultants 31 Dec May 2007

8 iv 5. Project Performance Report Ratings Implementation Period Development Objectives Ratings Implementation Progress From 1 Jan 2010 to 31 Dec 2010 S PS From 1 Jul 2009 to 31 Dec 2009 S S From 1 Jan 2007 to 30 Jun 2009 S S From 1 Jul 2004 to 31 Dec 2006 S S D. Data on Asian Development Bank Missions Name of Mission a Review 1 Review 2 Dates Nov Apr 2005 No. of Persons 1 3 No. of Person-Days 5 6 Specialization of Members b g b, d, g Review Mar d, g, h Midterm review 6-19 Sep d, g, h Review Sep d, g Review Jan g Review Nov g Review Jul d, g Review Aug g Project completion 2-5 Jul g, h a Include identification, fact-finding, pre-appraisal, appraisal, project inception, review, special loan administration, disbursement, project review mission. If more than one of each type of mission, number consecutively as review mission 1, 2, etc. b May use reference letters in table, e.g., a = engineer, b = financial analyst, c = counsel, d = economist, e = procurement consultant or specialist, f = control officer, g = programs officer, h=staff consultant

9 I. PROJECT DESCRIPTION 1. In December 2002, the Asian Development Bank (ADB) approved the Housing Finance Project. 1 The project aimed to improve the quality of life for urban low income and poor households (ULIPHs) through improved housing. The specific objective is to promote the establishment in Viet Nam of a sustainable housing finance market accessible to ULIPHs. The project aimed to provide approximately 27,500 housing loans for more than 137,000 people in the first round of lending, disbursing these through participating financial institutions (PFIs). 2. The project consists of three parts. Component A encompasses mortgage lending to urban low-income households (ULIHs) through participating commercial banks (PCBs). Component B involves housing microfinance for urban poor households (UPHs) via communitybased financial institution (CFIs). Component C addresses institutional strengthening and capacity building. Component C aimed to support the establishment of a housing finance facility (HFF) to become the apex institution for the housing finance sector, as well as implementation of a policy and institutional reform action plan to facilitate development of the housing finance system and a sustainable and effective housing finance market in Viet Nam. 3. The project s rationale rested on the assumptions that rapid population growth and rural urban migration had made access to housing difficult in Viet Nam s urban centers. In 2002, almost 40% of households in urban areas occupied usable housing space of less than 36 square meters (m 2 ) each, with 16% of households occupying less than 24 m 2. Urban LIPHs, in particular, lacked access to affordable options to finance house improvements or new housing in the absence of an efficient housing finance system. The lack of access to long-term credit funds prevented financial institutions from providing long-term loans for housing. The lack of long-term loans made access to housing finance unaffordable for almost 65% of the urban population. Short-term loans required high debt service payments. To make housing loans affordable to ULIPHs, it was deemed necessary to establish an efficient housing finance system that could provide long-term financing for purchase of housing and financing options for the poor to upgrade or build their own houses. The development of an efficient housing mortgage market in Viet Nam also required the establishment of an HFF that will refinance housing loans originated by qualified PFIs. The HFF is to become an independent mortgage refinance agency (MRA) serving as a conduit to attract additional funds to the sector through the issue of bonds or other securities. 4. The ADB loan was needed to (i) provide the HFF initial resources with which to start up its operations on a pilot basis, and (ii) provide capacity building for the HFF and PFIs with the objective of developing human resources in the national housing finance sector. The establishment of an HFF provided an opportunity to test mechanisms for loan disbursement to ULIPHs and acted as a catalyst to facilitate the progress of policy and legal reforms in the sector. ADB s involvement in housing finance is justified to (i) develop an institutional and policy framework for housing finance that integrates housing and finance issues, (ii) make housing loans affordable to ULIPHs and so reduce poverty and enhance the quality of people s lives, and (iii) create employment business opportunities in the construction and banking sectors. 1 ADB Report and Recommendation of the President to the Board of Directors: Proposed Project Loan to the Socialist Republic of Viet Nam for the Housing Finance Project. Manila.

10 2 II. EVALUATION OF DESIGN AND IMPLEMENTATION A. Relevance of Design and Formulation 5. At appraisal, the project design and objective appeared consistent with worldwide experience that a broad, market-based system is the most effective vehicle through which to provide financial resources to low-income and poor households in urban areas. Lessons learned from developing countries experience with ULIPHs indicate that interest rate subsidies on loans to ULIPHs often end up going to higher income groups, despite there being income-qualification criteria in place. The project s design was also based on the premise that any housing solutions for lower income segments at a national level could not be addressed by the central government and/or financial institutions acting alone. It requires the active participation of several public and private actors, including relevant government agencies and bodies. There remains an open question whether the HFF should have been located within the State Bank of Vietnam (SBV), although in the long term it is intended for HFF to become an MRA (mortgage refinance agency) to draw additional funds to the sector. It is therefore important to have an appropriate exit strategy in place. As per the project design, it was appropriate to have the HFF act as a refinance facility (revolving fund) in charge of disbursement to the PFIs and repayments on a regular basis. The establishment of an HFF is justified to the extent that it (i) promotes common standards, procedures, documentation, and good practices among PFIs toward developing an efficient and transparent primary mortgage market accessible to ULIPHs; and (ii) builds up basic requirements for the emergence of a secondary mortgage market. 6. The country strategy and program had suggested that ADB operations address dimensions to make growth and the structural transition pro-poor in its orientation and sustainable. 2 To this end, ADB aimed to focus its operations selectively on three pillars: propoor, sustainable economic growth; social development; and good governance. Further, project preparatory technical assistance of $400,000 was included into the 2002 country programming and the related Housing Finance Project was included in the ADB lending program for Viet Nam for The HFF was to be financed in the total amount of $30 million from Asian Development Fund Special Sources. The loan was classified under poverty intervention, as about 32% of the total beneficiaries are expected to be poor. Approximately 43,700 urban poor were expected to benefit from the project directly through the microfinance component. 7. However, given the significant weaknesses in Viet Nam s housing policy, the legal framework with respect to land and property rights, the institutional framework for implementation and the banking sector, the implementation of the project as well as its longterm sustainability were indeed challenging. It is important to recognize that secondary mortgage markets are unsustainable unless a primary mortgage market has been developed on a sound foundation of land titling, valuation, and legal mortgage loan recovery provisions. The project preparatory technical assistance was completed in June 2002 and the project was processed between July and November Given its design the project could have benefited from more in-depth consultations with the stakeholders. 8. The disbursement under the ADB loan was slower than envisaged at appraisal, as there were significant delays in the loan becoming effective and in recruiting and mobilizing the consulting firm under component C. After a brief initial mobilization of the consulting team in late November through to 21 December 2004, the consultants resumed work from 16 January Amendments were required in the loan agreement to build more flexibility into the project scope 2 ADB Country Strategy and Program: Socialist Republic of Viet Nam Manila. 3 ADB Technical Assistance to the Socialist Republic of Viet Nam for Housing Finance. Manila.

11 3 and implementation arrangements in order to reflect changes in both the policy environment and market reality. As described in Supplementary Appendix A, the changes made in the loan agreement (i) raised the subborrower income eligibility limit; (ii) increased the maximum subloan size; (iii) introduced an annual update mechanism for (i) and (ii); (iv) added flexibility to the accreditation criteria for PFIs to qualify for the HFF; (v) created flexibility to include any qualifying PFI without being limited to those PFIs identified in the loan agreement; and (vi) required that CFI loans of D14million and beyond be secured by land use rights certificates. An updating mechanism provided through the amendments to the loan agreement in 2006 recognized that future adjustments may be necessary due to inflation or other factors but stipulated that such adjustments be based on independent market data to ensure that only the intended sector (i.e., urban low-income and poor households) continue to be served. As the result of the aforementioned amendments providing for changes in the qualification criteria, 10 PFIs were accredited under the project. 9. Given the delay in project commencement and associated issues, the project s completion date had to be extended three times: the first being from December 2007 to June 2009, the second to July 2010, and the final extension until August The disbursement picked up during 2010 and By the date of the loan s physical completion, the project disbursement had exceeded $28 million, which was more than 95% of the disbursement plan. B. Project Outputs 10. Of 17 output targets included in the project framework (see Appendix 1), 10 have been fully achieved as of the date of this completion report. There were accomplishment delays given the belated start of the project. Of those remaining, 3 output targets relating to the number of ULIPH and ULPH loans for ownership and home improvement could not be met as envisaged at appraisal due to cost escalation since appraisal, both for new housing units as well as for upgrading. Nevertheless, the disbursement under the two credit components (components A and B) was $26,911,338, equivalent to 98.2% of the funds allocated for these components. In particular, disbursement of the ADB loan for component A reached 111.2%. Disbursement under component B was only 40%, however, due to problems associated with identifying poor households and because only one of the two CFIs originally included at appraisal actually participated in the project. As a result, no loan was extended under component B to acquire new houses for the urban poor. In view of the above, the HFF proposed to ADB to increase the credit line for component A and correspondingly to decrease that under component B. 11. The project s three components are discussed individually below. 1. Component A: Mortgage Lending to Urban Low-Income Households 12. Housing mortgage loans to ULIHs via PCBs were to be provided through two subprojects, as agreed at appraisal: (i) mortgage loans for housing purchase (costing up to D100 million per housing unit), and (ii) mortgage loans for housing upgrade (costing up to D20 million per housing unit). At appraisal, ULIHs with monthly household incomes in the range of D0.75 million to D3.5 million ($50 $233 equivalent) who could afford to finance a new or existing housing unit at a cost of D100 million ($6,666 equivalent) with 70% financing (30% to come from the borrower) or who could upgrade their housing at a cost of up to D20 million ($1,333 equivalent) at commercial interest rates and maturity of up to 20 years were to benefit from the project. The PCBs were to bear 20 percentage points of the 70% financing component from their own sources. Subsequent to periodic revision of the qualifying criteria based on agreed parameters laid out in the loan agreement as amended in 2006, disbursements under this component picked up considerably such that 11,761 sub-loans

12 4 (999 for new homes and 10,762 for upgrades) were provided to 40,575 ULIH beneficiaries as of the date of the project s physical completion in August If one takes into account the cost escalation since appraisal for new housing units or for upgrading, this compares favorably with the estimate at appraisal that approximately 4,100 ULIHs would receive loans to buy or build houses and about 14,600 ULIHs to upgrade or improve existing houses. Total disbursements under component A were $24.9 million, which is beyond the original allocation of $22 million. The additional allocation for component A was made at the request of the executing agency by reducing that for component B. In total, 10 PFIs participated under the project and availed themselves of the refinance from the HFF to onlend for subloans under component A. With the agreed periodic revisions, the qualifying average monthly income criteria for ULIHs adjusted to market realities as of the project s closing date was D4 million to D10.9 million per household in Ho Chi Minh City and Hanoi and D3 million to D9.83 million in other urban centers. The maximum value of qualified household units at project completion was D708.9 million for units in Ho Chi Minh City and Hanoi while it was D600.6 million for other urban centers. Table 2 in Supplementary Appendix B details disbursements between the project completion date and the date of this report. 2. Component B: Housing Microfinance to Urban Poor Households 13. Through established community-based financial institutions (CFIs), microfinance housing loans to existing UPH depositors were to be provided through two subprojects: (i) loans for small home improvement (up to D15 million per housing unit), and (ii) building material loans for own-build housing. The CFIs were to focus on UPHs beneath the poverty line, which at appraisal meant in the monthly income range of D350,000 to D750,000 ($23 $50 equivalent). Affordable options for the poor consist of (i) a building materials loan for self-help housing, and (ii) smaller housing improvement at a cost of up to D15 million ($1,000 equivalent) at commercial interest rates but at maturities of 5 10 years. At least 6,700 UPHs were targeted to receive loans to upgrade or improve existing houses, and 2,100 were targeted to receive loans to own-build new houses. While the original allocation for component B had been $5 million, as of the project s completion $1.20 million had been disbursed for home upgrades only through 1,324 sub-loans with 4,638 beneficiaries. No loan had been extended for new home acquisition. The low utilization was due to a lack of supply of suitable housing units for UPHs and to the fact that only one CFI (Central People s Credit Fund [CCF], which has since converted into a cooperative bank) participated in the project. With periodic adjustments as provided under the loan amendments, the average monthly household income range as of project closing for UPHs was D1 million to D4.37 million in Hanoi and Ho Chi Minh City and D0.75 million to D3.28 million for other urban centers. This is consistent with the Ministry of Labor s definition of urban poor based on income criteria. 14. There were a number of constraints upon meeting the demand for housing from lowincome people in urban areas. The price of units in social housing projects is not affordable for the majority of low-income and poor people. A lack of transparency and information in the real estate market made it more difficult for buyers and sellers. During most of the project period, real estate businesses focused on luxury apartments as these were seen to be more profitable in the short term. This period also saw limited investment in building apartments for low-income and poor people, even though this is the segment with the largest demand. Land and housing prices became potentially unstable and unreasonably high in comparison with real incomes.

13 5 3. Component C: Institutional Strengthening and Capacity Building 15. The project supported institutional strengthening and capacity building to develop the housing finance system in Viet Nam. This included to (i) support establishing an HFF that would become the apex institution for the housing finance sector, (ii) provide training and capacity building for HFF and PFIs, and (iii) develop a housing finance awareness and marketing program. The establishment of a housing demand database for ULIPHs at the HFF did not occur as envisaged, nor was the intended project performance monitoring system within the HFF and SBV s International Credit Project Management Unit (ICPMU) implemented. Difficulties and delays in implementing a management information system (MIS) contributed to these shortcomings. The allocation at appraisal for component C was $1.7 million, of which $1.3 million was utilized before the consultants ceased work in A significant level of training was delivered to HFF staff by the consultants based on a training program schedule approved by the HFF. This training covered broader issues relating to the project s implementation, including subjects such as overall policy issues, use of hedging instruments and mortgage insurance, and the need to implement the project in such manner as to enable the formation of a secondary mortgage market in future. Most scheduled training courses were completed as planned and additional workshops were delivered. ICPMU and HFF staff also participated significantly in overseas study tours. 16. The consultants prepared a report on creating an MRA and supplied to the HFF draft documents for its reference. Those drafts, however, were not sufficiently detailed. The HFF, in coordination with related SBV departments and other ministries, later prepared and in 2010 completed a draft MRA report for submission to the Governor of SBV. The draft is being revised and is scheduled to be submitted to the Governor by the fourth quarter of Housing finance awareness and marketing programs were carried out in 2006 and 2010 through advertisements in Viet Nam Banking Times and distribution of leaflets. C. Project Costs 18. The total cost of the project estimated at appraisal was $51.8 million equivalent, including all the investments into new and improved housing expected to be generated through the catalytic effect of the ADB-financed subloans. Table 1 compares the assumptions made at appraisal to the actual figures. Utilization under component A exceeded estimates while only 40% of the estimated amount for component B was utilized. The allocation for component C was only 66% utilized because the consultants ceased work in September 2006 as relations became strained between the executing agency and consultants. 19. The cost estimates had assumed that the ULIPHs receiving housing loans through the project would invest an amount equal to at least 30% of the value of the total housing expenditure. Given the cost increases in the realty sector since 2002, the original assumptions did not hold true either for the ADB contributions or counterpart funding. Increasing costs over the implementation period also resulted in reduced numbers of both subloans and beneficiaries being funded under component A while the borrowers contributions from their own sources came to 43.58% of the total cost. ADB s contribution as a percentage of total cost, at 46.07%, was lower than the original estimates of 57.92%, given the increase in project cost over the implementation period and the fact that end borrowers share of project cost, at 43.58%, was in excess of the original estimates of 28.8%.

14 6 Item ADB Table 1: Project Costs (original estimates and actual) ($ million) Allocations Utilized Unutilized Counter- Counter- Counterpart part part Funds Total ADB Funds Total ADB Funds Component A (2.5) (13.3) (15.8) Component B Component C Interest and fees Total (12.0) (10.6) Note: Negative values within parentheses denote the amount by which actual costs exceed those estimated at appraisal. Source: SBV (ICPMU) D. Disbursements 20. The proceeds of the loan were drawn in accordance with ADB s standard disbursement procedures. Two imprest accounts were used, one for the credit components (A and B) and the second for component C. Statement of expenditure procedure was used to liquidate advances into the imprest accounts or for reimbursements. The ceiling for the imprest account was raised from $3 million to $5 million on 20 April 2010 to facilitate cost-effective disbursements. The initial disbursement was on 30 August 2004 and the final disbursement on 11 January While the original allocation had been SDR22,438,000, the disbursed amount was SDR18,689,640.62, and thus there remained an unutilized amount of SDR3,748, The executing and implementing agencies had no issues in following the prescribed disbursement procedure. 21. Disbursement under the ADB loan was slower than envisaged at appraisal, as there were significant delays in the loan s becoming effective as well as in recruiting and mobilizing the consulting firm. By the loan completion date, the project disbursement had surpassed $28 million, accounting for more than 95% of the disbursement plan (see Supplementary Appendix C, Table 1). Disbursement under the two credit components totaled $26,911,338, equivalent to 98.2% of funds allocated for these components (see Supplementary Appendix C, Table 2). 22. About $8 million has so far been repaid by the PFIs and is available for refinancing under the revolving fund. Disbursements from August 2011 to the date of this completion report were $3.78 million. Table 3 in Supplementary Appendix C provide lending data by category, PFI, and time period. E. Project Schedule 23. The loan was approved in December 2002 but became effective only on 6 January Thereafter, there was some delay in recruiting and mobilizing the consultants under component C. After brief initial mobilization of the consulting team during inception in late November through 21 December 2004, the consultant resumed work from 16 January The loan agreement was amended on 5 June The project s completion dates were extended three times, and the final completion date was 31 August Financial closing was 3 May Total

15 7 F. Implementation Arrangements 24. The project was to be implemented over 5 years and expected to be completed by December As stipulated in the loan agreement, the Ministry of Finance (MOF) and SBV entered into a 20-year financing agreement on 28 November This enabled the establishment of an HFF project management unit. The ADB loan was routed through MOF to SBV and the HFF in dong. Qualified PFIs concluded a housing finance company loan agreement with the HFF before accessing housing loan refinancing for beneficiary subloans. The HFF initially provided credit lines for each housing finance company with a maximum of $2 million for PCBs and $1 million for CFIs. The maximum cumulative total housing finance company amount for any PFI was $7 million. PFIs provided from their own resources capital equal to 20% of all subloans to borrowers from PCBs and 10% of all subloans to borrowers from CFIs. Subborrowers were required to contribute 30% of the costs to ensure they had significant equity in their housing projects. Given the escalation in market prices the subborrowers ended up contributing about 44.5% of the project costs from their own sources. The subloans for new home acquisition were on average made for a maximum tenor of 15 years. The average rate of interest charged by the PFIs was between 15% and 18% per annum. The nonperforming loans are reported to be very low, at less than 1% of the overall loan portfolio. The principal repayments (about $8 million as of the date of this completion report) from PFIs to SBV are held in a revolving fund established by SBV to continue refinancing PFIs subloans. SBV is allowed to retain 1% per annum of the amount of the outstanding housing finance company loans to PFIs to cover costs of training and institutional capacity building (see Supplementary Appendix E). 25. The loan agreement specifies that the HFF will be managed by a project director and a full-time project manager and be staffed with at least 15 full-time professional staff including the project director and project manager. The HFF s initial structure was outlined in a Decision of the SBV Governor No. 304/2004/QD-NHNN, dated 24 March 2005, which accommodated an important requirement for an effective HFF, which is that full-time personnel report through the HFF s project manager and ICPMU s director. As of the date of this completion report, HFF and ICMPU together had total staff strength of 38 personnel to carry out their overall functions, although there was no longer any unit dedicated exclusively to housing finance activities. 26. Implementation of the project MIS has been difficult and there were many delays during mid The PFIs are submitting reports in hard copies which are then entered manually into the HFF operating system to generate internal reports. Basically, the MIS now performs only accounting functions. No database on housing demand and supply is in place, and there is no performance monitoring system. The HFF maintains only consolidated data as to the number of loans made under components A and B and not for each subproject under the two components. 27. In June 2005, the consultants provided a report to the HFF entitled Housing Finance Demand Database Update Manual. Two training sessions were provided to HFF staff to familiarize them with this web-based database and demonstrate how the relevant HFF staff could update it on at least a quarterly basis. However, the database is yet to be established. 28. A project steering committee was set up. Chaired by the SBV s Deputy Governor, it was comprised of representative from the ministries of Construction, Finance, Planning and Investment, Natural Resources and Environment, and Justice, as well as from SBV s Bank and Nonbank department. The meetings were irregular and rarely were chaired by the Deputy Governor. Two inter-ministerial committees were also established through an SBV decision dated 18 November 2005 to address policy and legal issues related to housing in Viet Nam.

16 8 29. There was also delay in activities under component C. According to the original action plan, the National Housing Finance Policy should have been completed and approved in Due to delays described above, it was eventually approved by the Prime Minister only on 13 July Moreover, the plan to establish an MRA based on the HFF and put it into operation in December 2007 was not realized inasmuch as the necessary premises for establishing and operating such agency were not yet in place. A draft paper for the establishment of an MRA had been completed and submitted to the Governor by ICPMU and HFF in October This required additional work, and it is now targeted for submission in the fourth quarter of Due to delays in the implementation of project activities the project completion date was extended three times. Closure was on 31 August In the original report and recommendation of the President establishing the project, gender issues were not considered significant for the project. Only a limited gender action plan was prepared. Performance relative to that plan is summarized in Appendix 2. G. Conditions and Covenants 32. There was some delay for loan effectiveness as the Governor s directive to establish the HFF was issued only in October 2003 while the financing agreement between MOF and SBV was executed only on 28 November The HFF established a financial management system acceptable to ADB as well as prepared draft housing finance company agreements between the HFF and PFIs with terms and conditions acceptable to ADB to satisfy the conditions for disbursement. Due to the delay in the start of the project, many of the action plans could not be completed within the agreed time frame. All qualified PFIs are in compliance with the eligibility criteria agreed upon by the government and ADB, including the prudential norms issued by SBV. Each accredited PFI submitted an opinion of an independent auditor to the effect that such PFI is in proper financial condition. However, the PFIs expressed their discomfort with the requirement for additional audit certification annually beyond the standard audit report on the compliance with various financial ratios stipulated in the project although they are willing to have the ratios certified by their authorized signatories. Most of the accredited PFIs have a housing finance project implementation unit and separate accounts for project-related housing lending. PFIs have formulated loan underwriting, services, and field audit procedures to ensure that subloans are made only to ULIPHs. Each PFI observed subloan financing criteria that include maximum ADB financing of 80% of the subloans provided by PCBs for ULIHs and 90% of the subloans provided by CFIs for UPHs. The HFF-ICMPU also confirmed that subloans do not entail any involuntary resettlement arising from land acquisition or changes in land use. The consultants as well as HFF carried out training programs as originally envisaged at appraisal (para. 15) including preparation of a housing finance manual detailing lending guidelines to PFIs. Based on this manual, regular monitoring was carried out by the HFF. PFIs as well the HFF are staffed with a fair proportion of female staff and credit officers. Almost all loans require the joint signature of both partners. The women s union at CCF provides care and support for female employees, equipping them with social knowledge and enhancing their professional skills. Certificates of excellence are regularly awarded to teams and individual female staff. Workshops are held at regular intervals at CCF, including participation by female staff in overseas trips for institutional strengthening and product development. CCF has been the principal borrower of the ADB loan funds through SBV and is keen to further avail of the refinance available through the recycling of used funds. Refer to Appendix 3.

17 9 33. Of the 8 policy and institutional action plans, 4 can be considered as fully complied with, albeit with delay, 1 can be treated as in progress, while 3 are not complied with as of the date of this completion report. The 4 completed actions are the following: (i) Develop a national housing policy document. The National Housing Finance Policy was approved by the Government under Decision No 105/2007/QD-TTg of 13 July (ii) Improve legal procedures for calling up defaulting mortgages and selling mortgaged land and houses. There has been good progress over time under this policy action. A number of legal measures have been taken to date toward improving legal procedures, as detailed in Appendix 3. A draft amendment of the Land Law 2003 is expected to be completed in 2013, though actual instances of enforcement of complex and wide ranging security packages in Viet Nam are extremely rare. (iii) Improve valuation system of land and houses to facilitate the development of the primary mortgage market. Decrees 188/2004 and 101/2005 and the valuing standards and implementing practices are making considerable improvements to the valuation system for land and houses. This improvement should facilitate the development of both the primary mortgage market and a subsequent secondary mortgage market. As per Decree 163/2006/ND-CP, on secured transaction, financial institutions are free to perform their own land valuations and take all associated risks. (iv) Increase number of urban land use certificates issued and improve registration procedures. As of December 2011, 3,683,411 land use rights certificates had been issued to urban households, accounting for approximately 64% of urban land area, which is beyond project expectations. 34. The capital market in Viet Nam is still in the early emerging stage, and at present there is virtually no investor interest in mortgage investment neither directly nor via mortgage bonds or other instruments. The policy action to attract additional funds to the capital market through development of a secondary mortgage market has been initiated through efforts by MOF to develop the bond markets in Viet Nam. This has been facilitated through the ADB Financial Sector Development Programs for Viet Nam. 4 A proposal to establish an MRA in Viet Nam was drafted and submitted to the Governor in October Currently, ICPMU, in coordination with departments within SBV and experts in the banking sector, is reviewing and updating the draft for resubmission in the fourth quarter of Noncompliance with three action plans is reflected in the situation that (i) not much improvement has been seen in the banking sector performance, and there is an increasing level of nonperforming loans; (ii) there are no particular products in the banking system to link formal savings with mortgage lending for housing; and (iii) there has been no progress in developing mortgage credit insurance, as there is currently no market need or demand for mortgage insurance products, particularly due to the low loan-tovalue ratio currently existing in the housing finance market. A more detailed status of compliance with the policy and institutional action plan is provided in Appendix 4. H. Related Technical Assistance 35. No separate technical assistance was provided under the loan. Project preparatory 4 ADB Report and Recommendation of the President to the Board of Directors on a Proposed Program Loan Cluster to the Socialist Republic of Viet Nam for the Financial Sector Program Loan II. Manila. ADB Report and Recommendation of the President to the Board of Directors on a Proposed Program Loan Cluster to the Socialist Republic of Viet Nam for the Financial Sector Program Loan III. Manila.

18 10 technical assistance (footnote 5) was provided in an amount not exceeding the equivalent of $400,000 for preparing the Housing Finance Project. 5 Since the initial delay of implementation of the Project was caused by insufficiently flexible design, the performance is rated less than successful. I. Consultant Recruitment and Procurement 36. The consultants were selected and engaged by the HFF under component C in accordance with ADB s Guidelines on the Use of Consultants and other arrangements acceptable to ADB for selecting and engaging domestic consultants. According to the contract for consultants services, 38 person-months of international and 26 person-months of domestic consulting services were contracted to provide technical support to HFF, PFIs, and the project steering committee in project implementation. There was a delay in recruiting and mobilizing consultants. According to the schedule at appraisal, the consultants should have been recruited to support project implementation within 3 months of the loan s becoming effective (1st quarter of 2004). Recruitment and mobilization of consultants could take place only by November 2004, however, given the time required for the executing agency to comply with local procedure for recruiting consultants and to obtain ADB s endorsement. The consultants commenced work on 16 January This impacted project implementation and disbursement. The consultants ceased work in September 2006, with funds of $397,073 remaining unutilized as provided in Supplementary Appendix E J. Performance of Consultants, Contractors, and Suppliers 37. The consulting services were engaged under component C. Given the delayed implementation, the consulting services were not fully utilized and $397,073 of component C funding was not used. The tasks assigned to support the project performance monitoring system and the project completion report could not be undertaken. Further, the housing finance database could not be established. The consultants ceased work in September 2006, as relations between the executing agency and the consultant had become strained. The performance of the consultants was otherwise stated to be satisfactory. Continuation of the consulting service beyond 2006 could have benefited HFF and ICPMU in gaining more capacity in operations, as the disbursements picked up only from 2008 onwards. Implementation of the project s MIS was difficult and encountered many delays during mid The project consultants provided support in developing the MIS tender package and were required to assist in the tender procurement process. The total cost for the MIS component that included hardware and software development was $40,000. It was procured in accordance with ADB guidelines and consistent with Vietnamese regulations. The MIS functions were eventually very limited, however, and the interlink between the PFIs and HFF did not function. This is attributed to lack of budgetary allocations for regular maintenance, including for hardware, software, and for upgrading the operating software. K. Performance of the Borrower and the Executing Agency 38. The borrower s part in program implementation is rated less than satisfactory. The HFF was set up as a separate dedicated unit within SBV under Decision No. 1386/QD-NHNN of October 2003 by the Governor of SVB, as required under the project. Currently, however, the HFF team is also involved in supporting other projects under ICPMU. The project benefit and 5 This technical assistance was financed initially as a grant, but it was subject to the reimbursement arrangements set forth in the Board papers on Technical Assistance Operations (Doc. R51-77, dated 20 May 1977) and Streamlining Technical Assistance Operations (Doc. R44-88, dated 21 March 1988).

19 11 monitoring system and the housing finance database were not established within the HFF. The MIS role has been very limited. Efforts in pursuing the policy and institutional action plans could have been better. Nevertheless, reports on project performance were regularly submitted to ADB. Formation of the project steering committee and the two inter-ministerial committees helped in establishing good linkage among the different government agencies to address the issues related to land titling and registration. L. Performance of the Asian Development Bank 39. ADB s part in program implementation is rated satisfactory. Appropriate corrective measures were undertaken in 2006 to address issues in disbursement and participation of PFIs. The loan agreement was amended in time to provide more flexibility in project implementation and the qualifying criteria to enable disbursements to reach 98% of the loan amount. Regular review missions (16) were undertaken over the implementation period to address implementation issues. The executing agency acknowledges ADB s active support and close coordination in project implementation. The executing agency further appreciates the timely resolution of project-specific issues by allowing flexibility in some of the criteria originally stipulated in the loan agreement. This facilitated accrediting 10 PFIs as well as disbursement. The executing agency also acknowledges the advisory support provided by ADB over the project implementation period, which helped the HFF in its operations. III. EVALUATION OF PERFORMANCE A. Relevance 40. The project design is rated relevant. It was relevant at appraisal, as measured against the government s development strategy, ADB s country strategy and program, and international good practice (paras. 5 6). Given that it was a pilot project for developing a market-based housing finance system, the project design was a bit ambitious. Though disbursement reached 95% of ADB s loan amount, some of the policy and institutional actions plans proved difficult to accomplish within the project implementation period. The design at appraisal could have been more flexible, as it delayed the project disbursements and required amendments to the loan agreement. The project is considered relevant, as it has created a base for development of the housing finance system and could help in the development of a secondary mortgage market with more progress in the policy and institutional action plan. Mobilizing savings to support housing for low-income and poor households remains high on the government s strategic agenda. However, no loans could be extended for new house acquisition for the urban poor under component B (CFI being the only lending PFI) even as late as the date of this completion report. This relates to the supply of suitable housing units and the disinclination of lending institutions to consider lending to the urban poor as a fitting line of business. B. Effectiveness in Achieving Outcome 41. Rated less effective is the purpose as per the project framework. This was to establish a sustainable housing finance market in Viet Nam accessible to ULIPHs through a housing finance facility (HFF) within SBV and which would become the apex institution for the housing finance sector in Viet Nam and an independent mortgage refinance agency. Through the project, an HFF was established in SBV. Though delayed, a draft proposal to set up the MRA is in the process of inter-ministerial consultation for submission to the Governor of SBV by the fourth quarter of The lack of access to credit was a major constraint upon ULIPHs, and the project helped in establishing common underwriting standards and due diligence among the

20 12 PFIs. Nevertheless, the project could not extend any loans to the urban poor for acquiring new homes as envisaged at appraisal. C. Efficiency in Achieving Outcome and Outputs 42. The project can be considered efficient considering the fair level of accomplishments in outcome and outputs (paras 10 17) Project activities translated into project outputs with a good level of ADB support throughout implementation. The outputs could have been better had the consultation support been more evenly phased out and available through the implementation period. D. Preliminary Assessment of Sustainability 43. The project has initiated the process to develop a sustainable market-based facility to provide housing finance to the ULIHs, although there still are issues with lending to the UPIHs as evidenced by underutilization of component B. This raises the issue of whether such marketbased mechanism can be applied to the urban poor or if special government-sponsored schemes are needed to address their housing needs. The use of collaterals and the need for property insurance for the urban poor were other constraints. The use of market-based interest rates is expected to lead to the project s sustainability, although the end borrowers could benefit more with a fixed-rate interest modality. The establishment of a revolving fund at HFF for refinancing mortgage loans from PFIs should ensure the sustainability and reuse of the funds, although consideration should be given to making the process more acceptable to the PFIs as only two PCBs and one CFI have expressed interest to further use the line of credit under the second phase. PFIs cite high administrative cost and the requirement for additional annual audit certification beyond the standard audit reports as issues. The willingness of PCBs and CFIs to provide 20% and 10%, respectively, of the total loan value and to standardize the housing loan underwriting procedures, practices, and documentation should also ensure sustainability. In the medium-term, will facilitate the establishment of appropriate control mechanisms and legal reforms, and in the long-term attract capital to the sector for the development of a secondary market. The HFF is likely to be sustainable. 44. The government in May 2013 announced a scheme for financing urban low-income and poor housing. Under the scheme, SBV will refinance loans totaling up to $1.5 billion to be extended through the four state-owned commercial banks and the Viet Nam Agriculture and Rural Development Bank for people who are not liable to pay income tax. The loans will be for acquiring housing units costing not more than D15 million and with area less than 70 m 2. They are to be provided at 6% per annum, with SBV extending an interest subsidy up to a maximum of 1.5%. Developers will also be provided financing for building low-cost housing units where land use rights will be provided and while requiring that at least 20% of the land be used for social housing. This government-sponsored scheme may reduce demand for the HFF, although it is available only through the state-owned banks, which are not participants in the HFF. E. Impact 45. The goal of the project as per the original project framework is to improve the quality of life for ULIPHs through improved housing. As of the project loan completion date, the total cumulative amount disbursed under the two credit components had reached nearly $27 million (over D510.3 billion), with more than 13,000 subloans and 45,000 beneficiaries. In comparison with the total of 5.8 million low-income households in urban areas across the country, the project beneficiaries accounted for a relatively modest 0.8%. Considering that this was a pilot project in the housing finance sector and with a modest loan amount, however, the project has