Joint Staff Advisory Note on the Poverty Reduction Strategy Paper

This Joint Staff Advisory Note focuses on the Poverty Reduction Strategy Paper (PRSP) for Djibouti. Djibouti’s National Initiative for Social Development (“Initiative Nationale pour le Développement Social”—INDS) provides a comprehensive vision for economic growth and poverty reduction. Notwithstanding some important achievements of the PRSP-I, overall progress in achieving its objectives has been limited. IMF staff commends the government for a well-designed and ambitious poverty reduction strategy. IMF staff considers that the INDS benefits from better prioritization and draws lessons from the shortcomings in the implementation of the PRSP-I.


This Joint Staff Advisory Note focuses on the Poverty Reduction Strategy Paper (PRSP) for Djibouti. Djibouti’s National Initiative for Social Development (“Initiative Nationale pour le Développement Social”—INDS) provides a comprehensive vision for economic growth and poverty reduction. Notwithstanding some important achievements of the PRSP-I, overall progress in achieving its objectives has been limited. IMF staff commends the government for a well-designed and ambitious poverty reduction strategy. IMF staff considers that the INDS benefits from better prioritization and draws lessons from the shortcomings in the implementation of the PRSP-I.

I. Overview

1. Djibouti's National Initiative for Social Development (“Initiative Nationale pour le Développement Social”—INDS) provides a comprehensive vision for economic growth and poverty reduction. The strategy was completed in January 2008, one year after the launch of the initatitive, and covers the period of 2008-11. The INDS document includes an assessment of the previous poverty reduction strategy (PRS-I), and outlines the second PRS. It maintains the thrust of the PRS-I, which was presented to the IMF and World Bank Executive Boards in May 2004. Similarly to the PRS-I, the INDS is based on four pillars, with added emphasis on spatial planning and environmental preservation: (a) accelerating growth while maintaining macroeconomic stability; (b) achieving universal access to basic services and development of human capital; (c) promoting a harmonious and balanced local development that preserves the environment; (d) and improving governance and capacity building.

2. Notwithstanding some important achievements of the PRSP-I, overall progress in achieving its objectives has been limited. The INDS indicates that many social improvements have been achieved since the PRSP-I: increased enrollment rates for primary education, reduced gender disparities in education, improved literacy among young adults, especially women, and significantly reduced infant mortality, achieved through a combination of concerted actions in macroeconomic management, human resource development, and integrative community development. However, the report recognizes that the implementation of the PRSP-I slowed, if not halted, after the first year. As a result, only 17 percent of the priority actions were fully implemented, about thirty percent were partially implemented and many objectives, particularly in the second pillar of developing human capital, were not achieved.

3. The staff commends the Government for a well-designed and ambitious poverty reduction strategy. The staffs consider that the INDS benefits from better prioritization and draws lessons from the shortcomings in the implementation of the PRSP-I. The INDS acknowledges administrative and resource constraints, weak country ownership of the PRSP-I, and a lack of effective monitoring and evaluation as the key reasons behind the poor implementation of the strategy. In the staffs' view these weaknesses have been well addressed in the INDS. The government is making a big effort, with support of the World Bank, to establish a Monitoring and Evaluation System for the INDS and has begun to tackle governance and public administration issues. Government ownership of the INDS process is stronger than in the PRSP-I. Lack of external financing could constrain the implementation of the INDS, but this is mitigated by emphasis on strengthening policies for shared growth.

However, the INDS could be strengthened by further elaborating on: (a) the poverty analysis which suffers from a paucity of accurate and reliable data; (b) the financing of the public spending program which potentially could challenge debt sustainability, (c) placing more emphasis on measures to promote more equitable economic growth; and (d) actions to make public administration reform operational.

II. Participatory Process

4. The authorities followed a participatory process to develop the INDS that involved critical elements of the society at large. While the original INDS document does not describe this participatory process, the authorities have provided additional details on the process in a letter to the IMF and the World Bank. In particular, they explained that the INDS was prepared in consultation with various stakeholders, including the sectoral ministries, parliamentary representatives, the Chamber of Commerce, trade unions, nongovermental organizations, and international donors. The content of the INDS was discussed with these stakeholders in several workshops that were open to the press. Moreover, the fourth pillar of the INDS document lays out the authorities' intent to strengthen the participation of civil society in the implementation of the strategy.

III. Poverty Analysis

5. Due to the lack of recent household surveys, the INDS document does not include a poverty analysis. However, the PRSP process has played a crucial role in paving the way for new surveys. EDAM-IS2,1 the latest survey (2002) and EDIM2 (2006) were conducted specifically to provide poverty and socio-economic indicators for the PRSP and for the attainment of the millennium development goals (MDGs). According to these surveys, the incidence of extreme poverty (defined in Djibouti at $1.8 per day) in 2002 was 42.2 percent and the incidence of relative poverty (defined at US$3 per day by the national poverty line) surpassed 74.4 percent. A poverty map published in 2007 indicates that there is significant regional heterogeneity in poverty outcomes. The poverty rate is much higher in rural than in urban areas where the majority of the population (about 80 percent) is located. The staffs welcome the authorities intention to conduct a new household survey (EDAM-3), following the population census.

6. The success of the PRSP-I was limited. According to EDIM 2006 data, progress was made in improving access to primary education (enrollment rates increased from 42.7 percent in 2000 to 66.2 percent in 2006), increasing the literacy rate among young females, and reducing infant mortality. Also, economic growth increased from 1.6 percent in 2004 to 4 percent in 2006. However, the INDS notes that the planned social programs were only partially implemented and hence, not all objectives of PRSP-I were achieved. As mentioned above, resource gaps, administrative capacity constraints and lack of an effective monitoring and evaluation system were the main reasons for the limited implementation of social programs.

7. Poverty monitoring has been hampered by lack of a population census and household surveys. The last census was carried out more than three decades ago and all population statistics are based on a demographic survey from 1991. This poses a considerable problem for the accuracy of almost all statistics in Djibouti, including information on the size and living standard of the population. Recognizing these shortcomings, the authorities are in the process of carrying out a census which will be followed by a survey of households and that of economic activities. Completion of these surveys is expected in 2009. This will allow the authorities to develop more specific programs based on adequate poverty diagnostics. Improvement of the statistic framework in general, and completion of the population census and the above surveys are also key elements in the second poverty reduction and growth facility (PRGF) arrangement requested by the authorities. The staffs acknowledge that building local capacity for data analysis and coordination is a priority.

IV. Macrorconomic Framework

8. The authorities' medium-term economic policy program—included in the INDS—aims at fostering sustainable and balanced economic growth and reducing poverty. The program envisages measures to maintain macroeconomic stability, improve competitiveness, reduce inflation, and strengthen the external position. Fiscal policy aims at balancing the budget in the medium term in order to create fiscal space for poverty reducing expenditures without compromising macroeconomic stability and reducing fiscal vulnerability. The authorities' inflation target can be reached through rigorous budget management and productivity enhancing structural measures, albeit with some delay due to the negative impact of the food and oil price shocks since the elaboration of the INDS. Current account deficits are likely to be higher than targeted given the large foreign and domestic investments projected and, at least in the short term, the impact of higher food and oil prices.

9. The INDS document indicates that economic activity and prospects have improved following large inflows of FDI in the port, construction, and related infrastructure, but additional efforts are needed to improve competitiveness. Growth is projected to increase gradually and remain high over the medium term, boosted by foreign direct investment (FDI), structural reforms to improve the business climate—including the introduction of new commerce and labor codes—and sound macroeconomic policies. The staffs concur with the authorities that the success of the INDS hinges on maintaining macroeconomic stability and promoting private investment. However, it is important to underscore the significance of a continuation of structural reforms for the achievement of the strong growth objectives as well as the realization of a pipeline of large investment projects as part of the government’s strategy to transform Djibouti into a regional trade and services hub. The challenge is to ensure that growth does not remain limited to an enclave around the port and to move ahead expeditiously with structural reforms aimed at enhancing competitiveness to narrow the current account deficit in the medium term. Decisive action is needed to reduce Djibouti’s high unit labor costs to a level in line with neighboring and competing countries, along with fiscal tightening (including through lower public sector labor costs) and structural reforms to reduce other production costs (particularly utilities). This issue should be further elaborated in future Annual Progress Reports (APRs), once the population census and subsequent surveys have been completed.

V. Structural and Sectoral Policies

10. The staffs welcome the emphasis on economic growth and macroeconomic stability in the INDS poverty reduction strategy. Continued economic growth is essential to guarantee the financing of the INDS and the creation of new jobs. Djibouti suffers from a very high rate of unemployment, estimated at 40 percent, and recent growth spurts were not associated with a proportionate increase in job opportunities. Further reliance on foreign direct investment, which mainly goes into capital-intensive and highly automated service activities, may further accelerate economic growth but is unlikely to generate more income for the poor population. The INDS therefore aims at promoting growth industries (tourism, mineral refinery, agriculture and fisheries) and strengthening the domestic business environment by improving the legal and regulatory framework while decreasing the costs of infrastructure service inputs such as telecommunication, energy and transportation. The government is also encouraging foreign investors to move into light manufacturing and other labor-intensive activities by creating a conducive environment for them in the existing and new free zone (e.g., power generators).

11. However, the INDS could benefit from more far-reaching initiatives to stimulate economic development and private sector growth. While the INDS presents a long list of actions to reduce the cost of infrastructure services and to create new jobs, the strategy could be strengthened by better addressing the country’s main economic development challenges: (a) excessively high energy costs which depress national competitiveness and private sector development, (b) severe scarcity of water and recurrent droughts, which constrain agricultural production, increases food insecurity and poses significant health risks to the poor population, and (c) weak capacity of the public administration for effective service delivery to the poor. Although many actions relate to these three challenges, they often do not seem far-reaching enough to induce structural change. Future APRs could usefully elaborate on reforms in the energy sector, including steps to privatize the public electricity company, and financial sector reform, drawing upon the findings of the upcoming FSAP. A more comprehensive approach to poverty reduction would also help. For example, the large urban development initiative proposed in the INDS document is exemplary and has benefited from strong planning and donor coordination. Consideration could also be given to lifting the structural divide between the growth poles (free zone and port) and the onshore economy which are governed by different incentive frameworks.

12. The imminent food crisis in Djibouti reveals gaps in the social protection strategy of the INDS. The INDS document does not clearly spell out how the authorities intent to mitigate the vulnerabilities of the poor to shocks. Djibouti’s economic structure is prone to price volatilities and external shocks as presently evidenced by the acute food crisis. The INDS mentions food aid and several medium- and long-term measures to reduce food insecurity. Yet, a clear vision on how to address vulnerability in urban areas, which are mostly excluded from existing social protection initiatives, is missing. In fact, the food security strategy of the government, developed in 2007 and updated in 2008, is not part of the INDS which could lead to conflicting priorities. For instance, the government is investing in agricultural production in neighboring countries in order to secure food supply while the INDS calls for more investments in domestic agriculture and fisheries to improve income generation opportunities for the poor. The staffs recommend including the government’s food security strategy in the INDS to ensure the coherence of government interventions in a crisis situation.

13. The staffs encourage the authorities to link social protection closely to other initiatives for social and human development. The INDS includes a framework for improving social protection. However, the proposed policy actions mainly consist of increasing coverage of assistance through a privately funded cash transfer system. There are some concerns about the implementation of this system with respect to coverage, sustainability, accountability, incentives, and the targeting mechanisms. In light of reoccurring droughts, the present trends in international food and energy markets as well as high poverty rates in Djibouti, a stronger effort is needed to develop well-targeted social protection programs. The staffs note the willingness of the authorities to enter into a dialogue with the World Bank on improving the social safety net.

14. The INDS document focuses on improving youth employment. Labor market participation remains very low, particularly among the youth: among those aged between 15 and 24, more than 90 percent are inactive and enrolment rates for this age cohort are also very low. The INDS mostly builds on the employment programs already included in the PRSP-I, except that these efforts are now supported by two special institutions, the ‘Observatoire de l’emploi et des qualification’ and the ‘Agence nationale de l’emploi, de la formation et de l’insertion professionnelle (ANEFIP)’. The INDS could benefit from greater specificity on how the authorities plan to use these agencies to increase the limited success of these programs during the PRSP-I. The staffs recommend carrying out a rigorous assessment of past efforts and learning from international experience before scaling up these initiatives. More emphasis could also be placed on the creation of a long-term vision and the sequencing of labor market interventions.

15. Although the INDS already envisages substantial improvements in governance and personnel management, further elaboration of these measures would be useful. While the INDS envisages judicial reform and improvements in governance and transparency, further progress is needed in advancing administrative reforms in the public sector to improve service delivery. By further emphasizing public service delivery and improving efficiency in the use of public financial resources the government will strengthen its ability to meet the ambitious INDS goals. This is especially needed in human resource and financial management, procurement, and decentralization. Such improvements are likely to provide impetus all other reforms, and improving education and health outcomes.

16. Personnel management reform could focus on the implementation of existing statutes. Although most statutes conform to the best international practice, they have not always been applied. Furthermore, improving coordination between the Ministry of Finance and the Ministry of Employment on matters of recruitment, promotion, and retirement of public employees will facilitate better wage bill management. Significant effort might also be warranted to strengthening the capacity of government to plan its requirements in terms of personnel and target these resources towards the activities where they are most needed. Indeed, the authorities are committed to improve governance and public administration. The President of the Republic has indicated, in a letter addressed to the World Bank President, that improving governance is one of the highest government priorities. Expanding work in this area is expected to take place over the next several months, with support from the Bank and other development partners.

VI. Public Expenditure Program and Financing

17. The public investment program (PIP) underlying the INDS raise some issues pertaining to fiscal sustainability. To achieve the ambitious targets in these areas, the staff projects a large public investment program (PIP) of about $700 million (equivalent to 80 percent of 2007 GDP) from 2008 to 2011. About one-fifth of total expenditure is expected to be domestically financed through budgetary resources and the rest (about $600 million) would be externally financed. The authorities have so far secured external financing for about $400 million—of which about one-half in the form of grants. The financing needs would be even higher if all projects in the INDS are implemented, regardless of priority. The implementation of the economic development strategy and the PIP could be seriously constrained by a lack of resources. The authorities recognize that a large increase in external aid will be required to cover the fiscal and external financing gaps, both projected to reach about 2 percent of GDP in the period 2008-11.3

18. The staffs recommend to implement an investment and debt management strategy consistent with reducing the high risks of debt distress identified in the recent debt sustainability analysis (DSA). The DSA highlighted the vulnerability of debt sustainability to lower than projected growth and less favorable than assumed financing terms. The temporary suspension of disbursements caused by occasional and relatively small accumulation of new external arrears underscores the need to improve debt management to boost donor confidence and ensure the availability of financing on highly concessional terms. The authorities intend to rely on grants and highly concessional loans to finance their large investment program. However, some form of future debt relief may be needed to strengthen external debt sustainability. Future APRs could cover the authorities’ plans and progress in this area.

19. The sectoral allocation of the PIP associated with the implementation of the INDS deserves further attention. It is unclear if the priority actions were subject to an evaluation by various sectoral Ministries in order to determine the costs of the investments and to facilitate the acquisition of required financing. In this context, it would also be important that work on multi-year public expenditure frameworks be initiated as soon as possible. The work relating to a medium-term expenditure framework related to the priority sectors of the INDS will allow improving the programming of the expenditure as well as the monitoring of the investment. Development of adequate evaluation systems for public investment project implementation would be particularly important to improve the efficiency of projects and improve governance.

20. The PIP gives priority to education and health as evidenced by their high share in expenditure (16.7 percent and 17.4 percent, respectively). Expenditures on water and sanitation, which were relatively low in the previous PIP, is now also receiving a high share of 11.2 percent in total PIP spending. The staff welcomes this change in priorities and shares with the authorities the view that the quality and access to water and sanitation have an important bearing on poverty reduction.

VII. Implementation Risk and Monitoring

21. The staffs concur with the view that improving monitoring and evaluation is essential for the success of the INDS. The staffs therefore welcome the developments of the necessary institutional framework for the implementation of an M&E system since the launch of the initiative. However, improving the capacity of the Directorate for Statistics and Demographic Studies (DISED) will therefore be paramount for the sustainability and effectiveness of the M&E system. Another key concern will also be the differences in monitoring capacity across the various government agencies implementing the INDS. It could be useful to address these capacity constraints when designing the M&E framework and strategy.

22. The staffs concur with the poverty focus of the sector targets and their alignment with the MDGs. Depth, detail and accuracy of the indicators could be improved by re-calibrating and updating them once the census and the planned surveys are carried out. The staffs also encourage the government to set specific annual indicators and to link them to the annual budgeting exercise to ensure resource availability for the achievement of the targets within the established timeframe. It is also important that the selected indicators across sectors adequately reflect the different socioeconomic dimensions of poverty spelled out in the INDS.

23. The authorities’ emphasis on improving the statistical system is also shared by the staffs. Sound and regularly updated data is needed to monitor economic and social progress in the country. In this context, the staffs welcome the authorities’ strategy for the improvement of the national statistics system outlined in the INDS as well as the recent progress made in establishing a famine early warning system. This progress shows that capacity to produce quality data exists and can be enhanced with donor-supported technical assistance.

VIII. Conclusions and Issues for Discussion

24. The authorities recognize that the available data does not provide a solid basis for precise poverty analysis. However, the authorities have communicated their clear intent to implement the census and a household survey in 2008 and improve the poverty analysis, Similarly, the authorities are elaborating a strong monitoring and evaluation strategy with sectoral and local government involvement, including regular reporting of the status of implementation.

25. The staffs believe that the INDS could benefit from a broader perspective on poverty alleviation. In particular, further elaboration on structural measures to promote economic development and domestic private sector growth is needed. Also, the INDS could be substantially strengthened by expanding the coverage and re-considering the forms of social assistance to the poor, in light of Djibouti’s vulnerabilities to external price volatilities and droughts. Finally, it could be beneficial to evaluate active labor market policies and investment in employment programs and place greater emphasis on public administration reform.

26. While the foundations of the INDS are solid, there are risks involved in attaining its development objectives. First, institutional weaknesses may delay the implementation of the INDS. Second, in addition to the authorities’ efforts in broadening the domestic revenue base and reducing nonsocial current expenditure, the staffs consider that the achievement of the ambitious objectives of the INDS hinges on the availability of grants, highly concessional external resources, and a favorable treatment of external debt to avoid threatening debt sustainability. Any shortfalls in these areas would seriously compromise Djibouti’s proposed development strategy, which is highly dependent on scaled-up foreign financing. In this regard, the organization of a donors’ conference in the last quarter of 2008 is particularly welcome. Continued failure to obtain adequate poverty diagnostics due to lack of M&E instruments is an additional key risk for the strategy.

27. In consideration of the INDS and associated JSAN, Executive Directors’ views are sought on whether they agree with the main areas identified by the staffs as priorities for strengthening the strategy and its implementation, and areas identified as key implementation risks.


Enquête Djiboutienne Auprès des Ménages-Indicateurs Sociaux II.


Enquête Djiboutienne à Indiators Multiple


The fiscal financing gap is expected to be filled by budget support grants and loans as well as debt rescheduling. The external financing gap is defined as the amount of additional financing (including exceptional financing under the program) that would be needed to ensure that the gross foreign assets coverage of the currency board remains above 105 percent. These gaps are estimated on the assumption that all external project financing necessary to finance the PIP would be available. A shortfall in project financing compared to projections would result in wider fiscal and external financing gaps.

Djibouti: Joint Staff Advisory Note on the Poverty Reduction Strategy Paper
Author: International Monetary Fund