Project Evaluation Summaries
Prepared by the Policy, Planning and Evaluation Unit (PPEU)

---

Ethiopia

---

I. Basic Project Data

Project Number:	ETH/89/C07
Project Title:	Lafto Urban Development Project
UN Cooperating Agency:	UNOPS
Government Executing Agencies:	Ministry of Urban Development and Housing (now Ministry of Works and Urban Development); Urban Development Project Office (UDPO); Project Implementation Office (PIO); Housing and Savings Bank (HSB) - renamed Construction and Business Bank (CBB)
Sector:	Urban Development
Sub-Sector:	Housing
UNCDF Budget:	US$     5,266,233
Gov't. Budget:	2,623,000
Total Budget:	7,889,233
UNCDF Expenditures at Evaluation:	3,675,146
Date Project Approved:	June 1990
Date Project Began:	Dec 1992
Date Project Evaluated:	Aug-Sept 1996
Type of Evaluation:	Mid-term

II. Background

Because of rapid population growth, a declining housing supply and rapid deterioration of existing houses, a growing percentage of Addis Ababa's population is compelled to live in overcrowded and unsanitary quarters. Several studies indicate that the city is threatened by a massive urban crisis that could affect hundreds of thousands of people.

This project follows the formulation by the World Bank of a project (URBAN 1) to develop new housing for 3,000 households from the slum area of Teklehaimanot and to use the space conserved by this consolidation to upgrade urban services. As a result, the Ethiopian Government requested that UNCDF finance a similar project resettling 1,500 additional low income households from Teklehaimanot to another site, known as Lafto. A concept paper was developed by the UDPO in 1986. The project was formulated in 1989 and signed in 1990.

III. The Project

The Project was designed to strengthen the financial capacity of the Housing and Savings Bank (now called the Construction and Business Bank, CBB) and enable the production of more affordable housing units for low-income beneficiaries.

The project was geared to two primary goals: 1) to provide loan assistance through the CBB for self-help construction of individual houses to 1,500 low-income households organized in 64 housing cooperatives and in three phases; and 2) to equip the Lafto site with necessary urban infrastructure such as drinking water, electricity, pedestrian and vehicular access, and street lightning. Space already released by the benefiting households would have permitted construction of social amenities, more sanitation and adequate access in Teklehaimanot under the prior World Bank project. In addition, four satellite projects were proposed dealing with income generation, kebele development, building materials, and rain water catchment, with funding secured through bilateral assistance and the help of UNDP. The Government contribution included staff and other local expenses as well as US$ 2,211,000 to the revolving fund (UNCDF allocated US$ 3,316,000). The Government was also expected to cover off-site infrastructure development and solid waste disposal.

IV. Purpose of the Evaluation

The objectives of the midterm evaluation were to take stock of the project achievements, shortcomings and effectiveness in delivering outputs and targeting poor households; to identify possible options for improvement; to consider the project's relevance in the new national policy framework; and to extract programming and policy lessons. The mission was asked to survey loan repayment and housing use by beneficiaries of the first phase, and to document the efficacy of the relationship with the World Bank project.

V. Main Findings of the Evaluation

>A. Assessment of Results Achieved

The Lafto Urban Development Project is successfully addressing chronic housing and infrastructure shortages by strengthening local capacity through the compatible involvement of cooperative members.

Duration

The scheduled three-year implementation period has been extended to more than six years due to a shortage of construction materials and large price increases at the beginning of the project. In addition, decentralization and the resulting transformation of the implementing agency into the Urban Development Support Services (UDSS) resulted in a significant loss of momentum. Lastly, the long and tedious procedures of loan acquisition and disbursement at HSB (now CBB) contributed to setbacks.

Construction activities

During Phase I, 568 houses were completed by the beneficiaries with the help of the project. A total of 624 others are almost finished (as part of Phase I), and 308 are 60% complete (those which began in Phase III). Each house is arranged as part of a four-unit complex, with four households sharing a common dry pit tank. Some members have installed shower sets which may have negative consequences for the water levels of the pits. As anticipated in the project agreement, about one-third of the houses are of the least expensive type, composed of a general-purpose room, kitchen and toilets. The remaining two-thirds feature an additional room and a partitioned main room. The structural design, choice of materials and workmanship respond adequately to the dual purpose of structural stability and low cost, except for the doors and windows (which are flimsy) and the absence of drains around the buildings (which could cause structural damage). Site development is lagging: only 30% of Phase I electrical works and about 85% of the water lines are completed. The roads are only now being contracted.

Credit component

The Bank established the UNCDF Trust Fund in a separate account, which unfortunately is not being independently audited. A total of US$ 2,600,000 has been transferred to the fund by UNCDF, with the final transfer expected soon. Exchange rate fluctuations offset the observed inflation in construction costs. The Government has not yet contributed to the Fund, but the UNCDF contribution is able to cover the loans approved by the Bank. Loans are incurred by cooperatives at a 4.5% annual interest rate and paid back during a maximum term of 29 years. The Bank nevertheless inspects all individual applications for housing. Income requirements in national currency terms were raised in 1992 to take inflation into account. Other requirements included a down-payment of 5% of construction costs and an employer's permitting direct deduction of the beneficiary's salary. Repayments are timely so far at the cooperative level, but reimbursements from individual beneficiaries to cooperatives are a source of concern: 52% of Phase I beneficiaries have arrears, 42% of which exceed five months. The Bank lacks cash-flow projections that would allow comparisons with actual collections. However, loan collection seems to be ahead of schedule because some beneficiaries have sold their house.

Benefits

Studies indicate that the housing stock deficit in Addis Ababa is close to 30,000. Effective demand, however, differs from need. The main cause for the housing shortage is the fact that over 80% of the population cannot afford to build their homes without subsidized building materials and credit. The project has helped the local population overcome these obstacles. The Project Implementation Office (PIO) national budget covers transport costs for the raw materials, electrical costs of block-making machines and engineering services. As a result, construction costs in the Lafto project are less than half the costs in other similar programmes. The interest rate is also substantially lower than for similar national programmes.

Impact on targeted population

Less than 20% of beneficiaries were drawn from Teklehaimanot, where low or unreliable incomes rendered many interested candidates ineligible for bank loans. The mission's socio-economic survey of project beneficiaries illustrates a shift in characteristics between Phase I and Phase II: 35% of Phase I households had monthly incomes below 250 Birr and 32% were from Teklehaimanot. In Phases II and III these respective ratios dropped to 1.5% and 10%. This unilateral decision by UDPO violated the project agreement and should not have been taken without the consent of the donor. Most beneficiaries are civil servants or working in State-owned companies (including 12% from the Municipality or UDPO). In 1984, the monthly income of 65% of Teklehaimanot households was below Birr 100, compared with a monthly wage typically ranging between Birr 200 and 500 in Lafto (between 380 and US$ 940 /yr). Even though the figures are separated by 10 years and an inflation as high as 100%, one cannot say the project has addressed the poorest segment of society, although it does benefit relatively low-income households. Regarding gender considerations, 35% of cooperative members are women, and projections indicate that there will be some 5,500 children under 16 living in Lafto when the project is completed.

Institutional arrangements

The project is nationally executed. Both the UDPO, which implemented the project until 1993, and the currently implementing PIO of Region 14 (Addis Ababa), were established for the WB Urban 1 project and are still managing WB projects, as is the CBB. The PIO inherited suitable staff from UDPO. The equipment that was transferred included two dump trucks, three cars and office equipment, all financed by UNCDF, plus one concrete mixer and one block-making press not originally procured for Lafto. None of the earth-moving equipment used in Lafto by UDPO was transferred to the PIO because they were procured with WB funding; this resulted in a loss of momentum in construction activities after the transfer. The 64 cooperatives are legal, democratic entities. They recently organized themselves into one strong body, the Overall Coordinating Committee, which was capable of efficiently lobbying the implementing agencies but is not yet represented at TPR meetings.

Policy environment

The project falls squarely within nationally-established priorities and strategies. Co-operativization, subsidized credit, and construction materials had been strongly encouraged under military rule and are still supported by the present Government. However, the national housing policy framework has changed in two ways since the project was approved:

• All interest rates for Government-funded programmes were increased to 10 to 15% while the rate for Lafto is still fixed at 4.5%; and
  
  
• The new land-lease policy reasserted State ownership of all land and raised the standards for commercial lease applications, with the effect of halting the allocation of 355 plots for commercial purpose in Lafto.

B. Assessment of Project Design

The project did not provide basic urban facilities such as market places, primary schools, or health posts. This may imply a hasty preparation and approval of the project, since these essential components of a new community had been successfully incorporated in the World Bank project and, logically, the UNCDF project should have gained from this experience.

Lastly, of the four satellite activities associated with this project that were planned at project inception but never funded (income generation, kebele development, building materials and rain water utilization), the evaluation mission found that while all are needed, the first two should be implemented immediately.

VI. Summary of Recommendations

The mission report listed numerous recommendations, the most important of which are the following:

General Recommendations

• Improve the coordination among project partners through regular meetings for more effective on site monitoring. Such coordination meetings should include beneficiary representatives, constructors, suppliers, PIO, UNCDF, and HSB representatives;
  
  
• Address the lack of basic training for cooperative leaders through seminars on basic accounting principles, project formulation and management, fund raising for new projects, and sustainability issues; and
  
  
• Projects proposals should be prepared and submitted to potential donors (ESRDF, NGO's) for satellite projects (proposals should include kindergartens, a key feature for working women).

Specific Recommendations for the Municipality

• Address the absence of social and economic facilities by providing temporary pits for garbage collection and incineration, and establish a health post to be temporarily managed by trained beneficiaries.

Specific Recommendations for the PIO

• Improve technical supervision. Immediate actions should include digging drainage trenches around buildings. Ensure that beneficiaries who install showers, sinks and flush toilets construct separate septic tanks;
  
  
• Improve poor road infrastructure by accelerating the tendering and awarding process for rural roads within the project area;
  
  
• Provide open spaces for a children's playground, and use some of the unoccupied houses as daycare centres;
  
  
• Organize regular site meetings including beneficiary representatives, contractors, suppliers, and implementing institutions, for improved coordination, scheduling, and monitoring;
  
  
• Organize training seminars for cooperative leaders on basic accounting principles, project formulation and management, fund raising and sustainability issues; and
  
  
• Negotiate with Reg. 14 Administration to ease the land-lease policy for this project, since it was approved before the new regulation, and allocate the 355 commercial plots.

Specific Recommendations for the CBB

• Improve the management of the UNCDF Trust Fund, in particular by providing cash-flow provisions and securing independent audits;
  
  
• Submit a proposal for future use of the UNCDF Trust Fund, which remains unclear. The mission recommends that it be used to finance either income-related "satellite projects" (income-generating activities and kebele development), or similar housing construction projects, an option preferred by all partners; and
  
  
• Protect the Fund from inflation. The current interest rate of 4½% could be increased to 6% to this purpose, with the consent of all parties.

Specific Recommendations for the Beneficiaries

Through the cooperatives and their Overall Coordinating Committee, the beneficiaries should:

• Continue lobbying parastatals for the provision of electricity and potable water; and

• Participate in all project meetings, including the TPRs.

VII. Policy Implications and Lessons Learned

The project's experience would suggest the following broad considerations for future housing schemes:

• To reach lower-income groups, loan conditions such as advance payments should be eased. Public information campaigns and a transparent beneficiary selection process should be mandatory.
  
  
• The design of an entirely new settlement scheme should always include economic and social facilities and services such as schools, health posts, employment opportunities, markets, etc. It is not enough to propose ancillary projects; follow-up action to develop concrete proposals and secure funding is essential.
  
  
• It is necessary to keep abreast of new policy changes such as the new land-lease regulations, and to introduce the required modifications and realignments in project design and strategy at an early stage.

VIII. Evaluation Team

The mid-term evaluation mission team consisted of Garedew Asfaw, Civil Engineer and Team Leader; Demisse Haile, Financial Specialist; and Haile Selassie Araya, Sociologist.