Indonesia’s provinces of Central Java and the Special Region of Yogyakarta experienced a 6.3-magnitude earthquake on May 27, 2006. It killed about 5,700 people, injured as many as 60,000, and destroyed or damaged more than 350,000 homes and other infrastructure. Estimated damages exceeded $3 billion. Seven weeks later, a second earthquake of 7.7 magnitude struck the special region and the provinces of West and Central Java. This earthquake generated a tsunami affecting 250 kilometers of Java’s southern coast. The combined event killed more than 500 people and displaced more than 200,000.
The government of Indonesia committed approximately $600 million to fund the reconstruction and rehabilitation of approximately 255,000 homes in the earthquakeaffected areas. Several development partners also contributed funds for a significantly smaller reconstruction initiative. At the government’s request, the World Bank used these additional contributions to create a recipient-executed Java Reconstruction Fund (JRF). The World Bank used the JRF’s resources to create the Community-Based Settlement Rehabilitation and Reconstruction Project (CSRRP) for Central and West Java and Yogyakarta Special Region. The CSRRP’s objective was to assist in meeting the needs of eligible households for earthquake-resistant housing and community infrastructure in the affected areas. These objectives were to be achieved through a community-based approach in which beneficiaries would have a major role in decisionmaking about reconstruction of their homes and the construction of their communities’ infrastructure.
The CSRRP became effective in early April 2007. The project’s original completion date of June 2009 was extended several times, eventually to June 2015, because of additional financing and expansion of scope after eight volcanic eruptions of Mount Merapi in 2010. The eruptions damaged access to infrastructure and communities’ social and economic facilities and required the relocation of thousands of households. The project’s objectives were not revised with the additional financing.
The project’s relevance of objectives and relevance of design are both rated substantial. The project’s development objectives were relevant to the conditions that existed after the earthquakes in 2006 and were suitably adapted to the posteruption conditions in 2010. The project development objectives were also consistent with the government’s strategies and with the World Bank’s Country Assistance Strategy in effect when the project started, but they could have been more relevant had they included more explicit attention to reducing disaster vulnerability, for example, through institutional aspects.
The project’s reliance on a community-driven approach was well suited to the needs of the affected communities. It promoted local ownership, benefited from reliance on Javanese cultural practices, and successfully reduced opportunities for corruption through direct, conditional cash transfers to community groups.
Achievement of the project’s first objective, “to assist in meeting the needs of eligible households for earthquake-resistant housing,” is rated substantial. Grants provided to villagers covered the costs of constructing “core” or incremental houses (frame, foundation, and roof but no walls) of up to 36 square meters (approximately 387 square feet) that would be earthquake resistant. The CSRRP was successful in reconstructing 15,153 houses that the earthquake had destroyed or severely damaged in May 2006. Another 2,516 houses were constructed in new locations after the volcanic eruptions. People occupying these houses were voluntarily relocated with the understanding that they could not reoccupy their former dwellings but could continue to use the agricultural land they owned in the volcano’s highest-risk zones. The World Bank claimed the CSRRP was “one of the fastest housing reconstruction programs in the world.” Despite this success, many homeowners have potentially compromised their homes’ structural integrity and resilience to disasters through unregulated and unapproved enlargement, sometimes by adding a second floor.
Achievement of the project’s second objective, “to assist in meeting the needs for community infrastructure in the affected areas,” is rated modest. Support from the CSRRP led to the completion of 265 community settlement plans (CSPs) that identified proposed infrastructure projects and that were intended to “incorporate hazard risk management strategies, such as emergency preparedness planning, local hazard mapping, and awareness-raising.” CSPs were to be developed before the initiation of housing reconstruction. This occurred in a few instances, but most CSPs were completed after the reconstruction of earthquake-affected homes had finished. Implementation of the CSPs led to the financing of 378 kilometers of village roads, 367 kilometers of drains and irrigation canals, 6,574 water supply systems, 1,162 communal toilets, more than 490 evacuation facilities, and signage in 310 villages.
The project was less effective in developing communities’ capacity for emergency preparedness and district governments’ capacity to integrate CSPs into higher-level development plans. Other than indicating that the CSPs had been developed, the project did not include any measurable indicators of outcomes related to the CSPs’ enhancement of disaster risk reduction or increasing communities’ capacity for emergency preparedness. Neither the institutions included in the capacity building efforts nor the outcomes of these efforts are clear; they are not included in the World Bank’s semiannual progress reports, aide-mémoire, or the Implementation Completion and Results Report.
The World Bank declared that dissemination of the CSPs to stakeholders and to local and provincial governments would be critical, but the plans were not widely distributed. Moreover, the Indonesian government had anticipated that the CSPs would be “living documents” that would evolve over time, but this did not occur. The plans are now largely irrelevant.
The project’s efficiency is rated modest. The World Bank’s project appraisal document (PAD) did not include an economic or financial analysis of the project’s projected economic rate of return. Likewise, the World Bank’s end-of-project assessment did not provide an ex post calculation of the rate of return or provide a comparison of actual unit costs with similar projects. The World Bank underestimated the unit cost of the in situ reconstruction of homes after the earthquakes. The cost of the project’s management was far higher than anticipated, even when considering the project’s expanded scope due to the eruptions. Had there been no volcanic eruptions, the project would have closed in June 2011, two years later than initially projected and three years after completion of all the housing reconstruction.
The project’s overall outcome is rated moderately satisfactory. The project’s development objectives were relevant to the prevailing conditions after the earthquakes in 2006 and suitably adapted to the posteruption conditions in 2010. The CSRRP was also consistent with the government’s strategies at appraisal and with the World Bank’s Country Assistance Strategy when the project started. The project’s reliance on a community-driven approach was well suited for both situations. Thousands of earthquake-resistant homes were constructed, many subvillages were successfully relocated after the volcanic eruptions, and community infrastructure was rehabilitated or enhanced in many subvillages. The CSRRP was less successful in enhancing longerterm attention to disaster risk reduction, ensuring the longer-term utility of the CSPs, and in measuring or addressing efficiency.
The project’s risk to development outcomes is rated modest. The community infrastructure the project funded is likely to continue in use for many years. This prospect is enhanced by communities’ commitment to maintain the infrastructure. The project provided training on the operation and maintenance of the infrastructure and assisted in the creation of community groups responsible for the maintenance. Although the project funded earthquake-resistant homes, this resistance is potentially in jeopardy because of the large number of houses that have been enlarged without building permits or suitable inspections, especially those on or near Mount Merapi. These houses are on the volcano’s slopes and thus remain at risk because of their proximity to the ashes, noxious gasses, and lava flows associated with eruptions. Had these subvillages been in place during the eruptions in 2010, all their residents would have been evacuated and their homes placed at risk because of the eruptions.
Quality at entry is rated moderately satisfactory. The World Bank staff in the country office benefited from its experience in Aceh province, which had suffered from an earthquake and a tsunami in late 2004. They modified the Aceh approach to accommodate the circumstances in the three provinces and prepared a comprehensive PAD in just a few months. In addition, the World Bank took advantage of the presence of two existing World Bank–funded projects to begin reconstruction before funds from the CSRRP were available. Nonetheless, several parts of the design were overambitious in what could be done about risk reduction in a postdisaster situation.
The World Bank’s quality of supervision is rated moderately satisfactory, as is overall Bank performance. World Bank staff were conscientious in their supervision of the project, which was aided by the task team leader’s full-time presence in the country office and their lead role in preparing the PAD. The World Bank’s semiannual project reviews were consistently comprehensive and forthright in acknowledging issues that were slowing or impeding project implementation.
The quality of supervision was less satisfactory in other areas. The JRF was a recipientexecuted trust fund, which meant that the World Bank should not have had a role in its implementation. The World Bank’s frequent interventions were well intentioned, but they risked undermining the government’s ownership of the CSRRP. The project’s results framework was not updated after work began on the posteruption activities, and there was no related reporting of results achieved relative to targets.
The government’s performance over the life of the project is rated satisfactory. The project became effective in early April 2007, and funds were first released the next month. The national government was highly supportive of the project and provided additional funding to it. The provincial and district governments, especially in the special region, were particularly supportive.
The performance of the implementing agency, the Ministry of Public Works, is rated satisfactory. It facilitated the completion of the CSRRP’s preparation and was responsible for hiring people to oversee the project’s implementation. This hiring was not always timely. In other instances, the project’s teams found themselves understaffed and without the requisite expertise. These problems were not always addressed in a timely manner.