Mozambique - Poverty Reduction Strategy Paper

Document of The World Bank FOR OFFICIAL USE ONLY Report No.: 22664 MOZ THE REPUBLIC OF MOZAMBIQUE Assessment of the Poverty Reduction Strategy Paper August 28, 2001 Prepared by Staffs of the International Development Association and the International Monetary Fund This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. j   FOR OFFICIAL USE ONLY THE INTERNATIONAL MONETARY FUND AND THE INTERNATIONAL DEVELOPMENT ASSOCIATION MOZAMBIQUE Joint Staff Assessment of the Poverty Reduction Strategy Paper Prepared by the Staffs of the International Monetary Fund and the International Development Association Approved by Jiirgen Reitmaier and Michael Hadjimichael (IMF) and Callisto E. Madavo and John Page (IDA) August 28, 2001 I. INTRODUCTION 1. The Government of Mozambique has been working on poverty reduction for over a decade. During the early 1990s, following thirty years of civil war, poverty reduction was part of the national reconciliation process. In 1995, an initial poverty reduction strategy was formulated and a poverty unit was set up in the Ministry of Finance and Planning, which steered the first National Poverty Assessment in 1996/7. In 1999, a policy framework for poverty eradication was developed in the form of 'Guidelines' and an 'Action Plan' for the 'Eradication of Absolute Poverty'. The latter document was incorporated into the interim-PRSP, and submitted to the Boards of IDA and the Fund in March 2000. 2. The full PRSP (PARPA 2000-05) is a significant improvement over the interim- PRSP (PARPA 2000-04). It has a fuller diagnosis of poverty, improved prioritization of sectoral programs, an expanded justification of macroeconomic policy, a new priority area on good governance, and a financing plan linked to the Medium Term Expenditure Framework. Above all, both PARPAs have been disseminated to donors, civil society groups and the private sector, and PARPA 2000-05 was modified in important aspects as a result. 3. Some weaknesses remain, particularly with regard to the identification of risks to the strategy and the costings of targeted improvements in social indicators. The latter mainly reflects institutional capacity constraints, which are being addressed in the near term (see paragraphs 28 and 29 below). The government views the PARPA as a 'rolling and dynamic programming tool,' and is committed to an ongoing process of development through consultations, dissemination, monitoring and review. This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  - 2 - II. COUNTRY OWNERSHIP AND PARTICIPATION 4. There is strong government ownership and commitment to the goal of poverty reduction, and the PARPA in particular. The document has been wholly produced by the government, with virtually no donor support. It was prepared by a policy unit in the Ministry of Finance and Planning, in coordination with the line ministries. It has been approved by the Cabinet as the overarching framework for government policy, and has been fully integrated into government planning, budgeting and reporting processes. 5. Mozambique has a centralized government tradition that tends to approach participation primarily as information dissemination and consultation on prepared documents. Consultative mechanisms have been developing rapidly at the sectoral level (detailed in Annex 1 of the PARPA), but neither government nor civil society organizations have much experience with partnership in policy analysis and design. However, within this context, there has been extensive consultation on this PARPA/PRSP, which is detailed in Annex 2. 6. During 2000, the I-PRSP (PARPA 2000-04) was published and disseminated in Maputo and in a number of provinces, along with provincial poverty profiles. Towards the end of the year, a number of consultations were held with specific stakeholder groups, including donors, the private sector, the media, civil society organizations, and community leaders to discuss early drafts of the full PRSP. Representatives of these groups told staffs of the Bank and Fund that they welcomed the consultations, although they would have appreciated more advance planning of the process. 7. The consultations provided confirmation that the PARPA priority areas were the right ones, and stressed the need to improve implementation and delivery and to address problems of governance and corruption. There was, in addition, some criticism of specific policies and some complaints of unethical behavior amongst some public servants in schools and health centers. This feedback is briefly described in the PARPA 2000-05, along with details of how the strategy was revised as a result. 8. The PARPA sets out plans for further deepening the consultative process, with greater involvement of parliamentarians and local government, building on participatory planning at district level. However, the institutional arrangements for steering this process, as well as the ongoing development and monitoring of the PARPA in general, have yet to be delineated. Publicly available details of such arrangements would clarify the decision-making process for stakeholders outside the Ministry of Finance and Planning, and would make the participatory process more effective in future. III. POVERTY DIAGNOSIS 9. Mozambique is fortunate to possess good quality poverty data in the form of a Household Consumption Survey (1996-97), a Demographic and Health Survey (1997) and a Population Census (1997), and extensive quantitative analyses of these data, which  - 3 - are briefly summarized in the PARPA.1 The quantitative analysis is complemented by qualitative information from an early Participatory Poverty Assessment (1995-96), and the provisional results of new Participatory Rural Assessments, carried out in January 2001. These data provide a good basis for an analysis of the characteristics and causes of poverty in Mozambique during the late 1990s. However, time series data are only available for some social indicators; as a result an analysis of trends in income poverty, including a fuller assessment of the impact of policies on poverty reduction will have to await the conclusion of a new household consumption survey currently underway. 10. The staffs agree with the broad conclusions of the poverty analysis, but note that there are some weaknesses, particularly regarding gender and vulnerability2. Gender is only referred to in the context of household headship, educational attainment and maternal and child health. There is no discussion of the causes of gender inequality and little in terms of policy interventions to address it. There are few gender-specific targets, and the PRSP fails to build on recent work by donors on gender inequality and on recent policy reforms aimed at strengthening women's legal rights. 11. In general, issues of vulnerability are not given a high priority in the PARPA. It does not play a large part in the poverty diagnosis, and social protection is not included as a priority sector. There is relatively little mention of natural disasters and food insecurity, which is surprising, given the recurrence of such disasters. HIV/AIDS is only analyzed in terms of its macro-level effects on demographic and economic growth; its contribution to new forms of vulnerability is not described. 12. The future poverty research agenda mentioned in the PARPA is appropriate; this includes alternative methodologies for the establishment of poverty lines, benefit incidence analysis of public services in education and health, and poverty mapping in order to identify poverty foci at the district level. However, the staffs would urge the government to give priority to analyzing poverty trends, once the new data become available, in order to understand the distributional impacts of recent growth and government policy. For example, the social indicators in table 2.6 show a mixed picture, with significant increases in school enrolments and vaccination coverage, but little movement in school retention and infant nutrition. These data suggest that a more critical analysis of the poverty impacts may be warranted. IV. THE POVERTY REDUCTION STRATEGY 13. The overall objective of the PARPA is to reduce poverty by about 30 percent over thirteen years, from 70 percent in 1997 to below 60 percent in 2005 and 50 percent by 2010. The basis for these figures is not given, although they appear to be drawn from computer simulations of the poverty impact of different growth scenarios, based on the household consumption survey of 1996/7 (table 3.2). The table suggests that this goal 1 Understanding Poverty and Wellbeing in Mozambique: the First National Assessment (1996-97), Ministry of Finance and Planning, Eduardo Mondlane University and IFPRI, Maputo 1998; and the National Human Development Report 1999, UNDP, Maputo 2000. 2 Vulnerability is very difficult to measure, especially (as in the case of Mozambique) in the absence of panel data.  would be conservative if the high growth rates projected by the PARPA were achieved. However, in view of the limited information on the actual distributional effects of recent growth, it is difficult to evaluate this objective. Future versions of the PARPA should elaborate on the linkages between growth and poverty reduction. 14. The public action strategy of the PARPA emphasizes economic growth, public sector investment in human capital and productive infrastructure, and institutional reform to improve the enabling environment for the private sector investment. To implement this strategy the document identifies six priority areas and eleven complementary areas.3 Good justifications are provided for the selection of the priority areas, which are broadly in line with the poverty diagnosis and were confirmed during the consultations with civil society and the private sector. The staffs consider the selection of priority areas to be appropriate, although protection against natural disasters could be given greater emphasis. 15. Substantial progress with policy prioritization has been made, with respect to the interim PRSP. This has been achieved in three ways: first, through the distinction between the priority and complementary programs; second, through the identification of key measures within each of the priority sectors; and third, through the setting of targets and milestones in the operational matrix. The staffs welcome the progress with prioritization and agree that the areas prioritized are the right ones, although in future, more information about policy choices and trade-offs would be welcome. It is also likely that further prioritization will be needed, and this could be achieved with more attention to regional priorities and the sequencing of reforms. Growth and macroeconomic policy 16. The PARPA stresses that economic growth must be both rapid and broad-based to benefit the poor, and the strategy is based on an average growth rate of 8 percent for the period 2001-2010. The sources of growth include production from identified large- scale, capital-intensive projects financed with private foreign capital ('megaprojects'); productivity and value-adding gains in agriculture and small manufacturing; and a general expansion in internal trade, transport and services. These processes would be complemented by public investment directed towards poverty-reduction objectives. Moreover, as indicated in the PARPA, Mozambique needs to continue to work towards developing an enabling business environment, for domestic as well as foreign investment. 17. Attainment of this ambitious growth objective-which would maintain the performance of the last five years-requires continued macroeconomic stability and a deepening of the reform program. The staffs agree that, as indicated in the PARPA, there are important uncertainties that will need to be carefully monitored and may need to be incorporated into annual revisions of the framework. In particular, the recent heavy flood damage-which lowered the growth rate to just over 2 percent in 2000-illustrates the 3 The priority areas are education; health; agriculture and rural development; basic infrastructure; good governance; and sound macroeconomic and financial management. The complementary areas are: employment and business development; social action; housing; mines; fisheries; tourism; industry; transport and communications; technology; the environment; and protection against natural disasters.  risk that periodic exogenous shocks may constrain the actual growth in output. In addition, while the likely impact of AIDS in Mozambique has yet to be fully understood and quantified, the high rates of HIV infection (about 16 percent of the adult population) in Mozambique may also lower the rate of growth. 18. The staffs agree that a number of private sector megaprojects, in the area of hydroelectric power and natural resource extraction, are likely to contribute very significantly to medium-term growth. Indeed, by focusing only on the already identified projects of this nature, the framework may understate the role to be played by large scale private investment, particularly in the outer years. However, the PARPA could more explicitly address the contribution of the megaprojects to poverty reduction. The nature of these projects, which are highly capital- and import-intensive and geographically concentrated in the south of the country (where the incidence of poverty is lower), suggests that their links to poverty reduction are indirect. Nevertheless, the staffs recognize that the success of these projects will send an important signal to attract other potential investors in Mozambique's development corridors. Other sources of growth could also be more fully elaborated in the PARPA, in particular, the reasons for the envisaged strong growth in the agricultural sector (see also paragraph 27 below). 19. The staffs consider the macroeconomic framework of the PARPA to be broadly sound.4 The staffs fully support the central role assigned to the maintenance of macroeconomic stability, including annual inflation of 5-7 percent, in fostering higher growth. As stressed in the PARPA, a prudent fiscal policy to limit recourse to domestic financing and support monetary restraint will be key to maintaining stability. The staffs support the Government's stance of applying a strong fiscal adjustment strategy both in the short term-sharply reducing the deficit and spending in 2002 after the temporary increases to meet emergency and reconstruction spending after the floods and bank restructuring cost-and in the medium term to reduce its current exceptionally high level of aid-dependency and ensure fiscal sustainability. To this end, the staffs welcome the government's plans, stated in the PARPA, to develop a medium-to-long-term strategy for external and internal debt, following the debt relief initiatives of recent years. 20. The staffs agree with the emphasis placed on raising revenues from their current low level of 12.4 percent of GDP to over 15 percent of GDP by 2005, through a combination of continued reforms of the tax system and improved administration. The risk, noted above, that growth may fall short of the objective implies a related risk that the framework may overestimate the medium-term budget envelope. In addition the staffs are of the view that it may be prudent to anticipate a somewhat sharper decline in external support for the budget as a percentage of GDP in the outer years, in line with the objective of reducing aid dependency. The strategy embodied in the medium term fiscal scenario also implies a rationalization of spending over the medium term. The staffs 4 The staffs concur with the underlying assumptions and policy stance of the macro framework, over the next few years which is reflected in the evolution of the fiscal and external balances. However, long-run economic growth may be somewhat lower than the 8 percent projected by the authorities. In addition, the staffs consider the wide annual movements in private savings-derived as a residual in the framework-to be the result of problems of classification of expenditure aggregates and not indicative of the underlying behavior of the private sector.  - 6 - welcome the priority to be given to expenditures for poverty reduction (see below) so that the objectives of the PARPA can be met while reducing aid dependency. However, the PARPA would benefit from a fuller discussion of how savings are to be achieved in other areas, by improving expenditure control and the effectiveness of the spending through, inter alia, the implementation of the new financial management law. 21. The staffs agree with the acknowledgment in the PARPA that minimizing the risks of financial crises is a major objective of financial sector development. The recent difficulties surrounding the resolution of a large insolvent bank (Banco Austral) have underlined the need for strong banking supervision. However, the staffs note that this issue did not come to a head until after the text of the PARPA was finalized. Governance 22. The staffs welcome the emphasis placed on good governance, which is one of the priority areas in the PARPA,5 in particular, the attention given to enhancing transparency, accountability and the capacity and effectiveness of public institutions. A number of important initiatives are underway in these areas, especially the new Public Financial Management Law and the Integrated Strategic Plan for the Justice Sector, both expected to be completed before the end of 2001. The staffs welcome the attention given to decentralization and local participatory planning as mechanisms for improving local service delivery and accountability. 23. The PARPA is the first widely disseminated official document of the government to recognize the need to address corruption, which was one of the issues raised during the consultation process. The authorities are clearly at the early stages of a response on this issue; the document does not provide an analysis of the nature of the problem and detailed concrete proposals to address it, nor does it set out clear objectives or indicators of progress in this area. Nevertheless, the staffs believe that the systemic reforms envisaged to improve governance and the legal system will be crucial to a sustained reduction in corruption and welcomes more recent measures taken to strengthen the ministry of justice and related agencies and to establish an Anti-Corruption Commission. Future versions of the PARPA should provide more detail on these measures and progress with their implementation. Concerted efforts and progress in tackling corruption will be essential in establishing the conditions for equitable growth supported by private investment, and donor assistance. Sectoral policies 24. The PARPA's improved treatment of macroeconomic and governance issues, complements its strong sectoral focus, which emphasizes human and rural development. As mentioned, this emphasis is appropriate, in the light of the low social indicators and high incidence of rural poverty. ' Governance subcomponents include decentralization, legal and judicial reform, strengthening institutions that interface with the private sector, protection of frontiers, rationalization of public finances and anti- corruption measures.  - 7- 25. Given the important role of education for pro-poor growth, and the low levels of post-primary education, the staffs agree with the strategy of expanding access to the system at all levels. However, the policy would benefit from greater attention to increasing quality, as indicated, for example, by improved retention rates. Improvements in teacher training and incentives would be an important part of this process. Issues of planning, budgeting and decentralized management should also receive greater attention. 26. In the health sector, the PARPA focuses on primary health care (PHC), combating the major epidemics, improving the network of basic health facilities, developing human resources, and improving planning and management for the sector. These measures will be implemented through the recently approved Health Sector Strategic Plan, and will improve health services for the poor and reduce the health risks to which they are disproportionately affected. Special attention is given to HIV/AIDS in the health sector, although the PARPA has relatively little to say about the broader cross-sectoral actions contained in the National Strategy against STDIHIV/AIDS. Future policy analysis would benefit from a broader approach to combating HIV/AIDS. 27. The promotion of rural development through the provision of basic infrastructure, agricultural extension, and assistance with credit and marketing are clearly important interventions in a country with such a large, impoverished rural population. The overall growth strategy includes annual average growth of 8 percent in agriculture, based on expansion in cash crops and increased production of food crops. However, the PARPA could be strengthened by a consideration of the key constraints to investment and growth in the sector, including uncertainties over land rights, high transportation costs and volatile international prices. Costing, budget and fimancing 28. In general, Mozambique's public expenditure management systems are weak, and budgetary classifications have not facilitated the identification and tracking of poverty expenditures. Cost data are not centralized and uniform, and not easy to obtain outside of multi-donor funded sector programs (SWAPS).6 As a result costing and budgeting processes need to be strengthened. These issues are now being addressed in the context of the participatory Public Expenditure Review currently underway and the new Public Financial Management Law. Although a new detailed classification system will be implemented with the 2002 budget, meaningful policy-enhancing results may take some time. 29. In this context, the proposals for the financing of the PARPA have mainly relied on the establishment of budget targets for the priority areas, based on a projected budget envelope for 2001-05. Nonetheless, the staffs' view is that the proposed public expenditure program is consistent with the priorities of the stratepy. This program is anchored in the Medium-Term Expenditure Framework (MTEF) , linking the broader 6 The budget ceilings set by the Ministry of Finance and Planning are based on funding proposals drawn up by the line ministries, which are apparently based on internal cost calculations. However, the staffs have limited information about the nature of these calculations, particularly those outside sectoral programs. 7 The Mozambican MTEF is officially known as Medium-Term Fiscal Scenario 2001-2005.  - 8 - vision of growth with the shorter-tern budgetary implications. The staffs welcome that the PARPA contemplates the inclusion of remaining undeclared own receipts8 ("off- budgets") in the budgetary process during 2002. Unit costs are provided for those sectors that have them (education, health, and infrastructure) as an illustration of future work to be undertaken, rather than as underpinnings for the expenditure program. 30. The PARPA budget indicates that the proportion of total expenditure allocated to the priority areas (education, health, infrastructure and agriculture and rural development) will rise from almost 60 percent in 1999 to 67 percent in 2003.9 As a percentage of GDP, expenditure in the priority areas is projected to rise from 13 percent in 1999 to 19 percent in 2003. 31. Total projected expenditure is disaggregated in terms of the priority sectors (Table 7.4) and further disaggregation within these sectors is presented in Table 7.5. It should be noted that the selected priority expenditure lines within sectors remain fairly constant over the 2001-05 period and there is little discussion of possible changes in allocations between sectors or subsectors. 32. The budgeting exercise presents a single scenario without any contingency plans in the case of a shortfall in revenues or financing. As mentioned, it is based on ambitious assumptions regarding the GDP growth rate, revenue effort, and concessional and grant financing. The staffs expect that some of these issues will be reviewed during the next medium-term programming exercise, for the period 2002-06. Targets, Indicators and Monitoring 33. The monitoring of the PARPA will be integrated into the regular system of quarterly and annual government reports to Parliament. A special annual poverty report will also be prepared, based on both quantitative and qualitative data. The PARPA does not specify the form of the poverty report, but ideally it should include monitoring at three levels: sectoral performance; execution of programmed expenditures and revenues; and changes in welfare as measured by poverty and social indicators. The main quantitative data sources will be (a) administrative data produced by the line ministries, and (b) annual household surveys of key welfare indicators (Core Welfare Indicators Questionnaires)'0. The key institutions involved in producing these data are competent and committed to establishing a sustainable monitoring system. 34. In general, government reporting to Parliament is based on the monitoring of inputs and activities more than on outcomes or results, and the operational matrix has been devised with this purpose in mind (although it does include some output measures). It was developed through discussions between the Ministry of Finance and Planning and the line ministries and is consistent with existing sectoral targets. The indicator table in the monitoring section represents an initial attempt to focus on a smaller number of key targets and indicators for each priority area, with a clearer distinction between 8 Funds collected by ministries that are not recorded in the budget. 9 Or, in dollar terms, from approximately US$ 542 million in 1999 to US$ 705 million in 2003. 10 QUIBB in Portuguese.  - 9 - intermediate and outcome indicators. However, some of these measures are provisional, as in some cases the precise quantities still need to be established and the relevant data sources defined. Targets and indicators are best specified in those sectors that have SWAPS in place and for the most part they are realistic (e.g. education, health). As reporting on the PARPA becomes more institutionalized, a further refinement of its indicators may be expected. In that process, the link to poverty outcomes should be strengthened. 35. The PARPA recognizes that civil society would play an important role in poverty monitoring, although it is not clear how this would be organized. This aspect will be of critical importance for both the credibility and legitimacy of the PARPA and that of the consultation process itself. Risks to the strategy 36. The PARPA does not analyze the risks to the strategy. These risks are mainly associated with Mozambique's ability to continue growing at the high rates observed in the second half of the 1990s-which is especially important given the central role of economic growth in the PARPA strategy. First, the negative impact of the severe floods in 2000 highlights the risks to the strategy from natural disasters, particularly in the agricultural sector. Second, the social and economic consequences of AIDS/HIV could be substantially higher than reflected in the PARPA. Third, although the projections include only already identified megaprojects, implementation of these may be vulnerable to unforeseen market developments or political risks. Given the large role of megaprojects in the growth strategy, the impact of growth on poverty reduction may also be less than anticipated. Fourth, the strategy rests upon a single scenario which does not indicate the likely response to less favorable developments, including possible shortfalls in revenue or grant financing. Finally, implementation of the strategy will depend crucially on efforts that are now underway to strengthen institutional capacity. All of these factors indicate the need for caution and, as the PARPA notes, the likely need for revisions to macroeconomic projections in the light of developments. V. CONCLUSIONS 37. Compared with the I-PRSP, the PARPA/PRSP represents a major step forward in Mozambique's fight against poverty. There is a stronger link between the poverty diagnosis and the policy agenda, improved integration of macroeconomic policies, institutional reforms and sectoral programs, better prioritization and clearer, monitorable targets. There is increased ownership through the consultation process. 38. There is room for further development of the poverty analysis, particularly including aspects of gender, HIV/AIDS and vulnerability. Regarding the policy programs, there is a need for further attention to prioritization and sequencing, and a greater emphasis on implementation (including aspects of quality and efficiency as well as access). Above all, there is a need to deepen and institutionalize the participatory process, so that a wider group has ownership of the PARPA. In this connection, the staffs  - 10- welcome the authorities' intention to enhance the consultative process with greater involvement of parliamentarians and local government. 39. In view of the considerable progress made since the interim PRSP, the strong ownership of the PRSP/PARPA, and the commitment to its further development, the staffs of the World Bank and the IMF consider that it presents a credible poverty reduction strategy that is a sound basis for IDA and Fund concessional assistance. The staffs recommend that the Executive Directors of the World Bank and the IMF reach the same conclusion.  REPUBLIC OF MOZAMBIQUE ACTION PLAN FOR THE REDUCTION OF ABSOLUTE POVERTY (2001-2005) (PARPA) (STRATEGY DOCUMENT FOR THE REDUCTION OF POVERTY AND PROMOTION OF ECONOMIC GROWTH) (Translated from original text in Portuguese) Final Version Approved by the Council of Ministers April 2001   ACTION PLAN FOR THE REDUCTION OF ABSOLUTE POVERTY (2001-2005) (STRATEGY DOCUMENT FOR THE REDUCTION OF POVERTY AND PROMOTION OF ECONOMIC GROWTH) Table of Contents I. INTRODUCTION: OBJECTIVE AND STRATEGIC VISION II. CHARACTERISTICS OF POVERTY 111. THE DEMOGRAPHIC, MACRO-ECONOMIC AND TERRITORIAL CONTEXT IV ACTION PLAN FOR THE REDUCTION OF ABSOLUTE POVERTY AND THE PROMOTION OF ECONOMIC GROWTH. INTRODUCTION A. FUNDAMENTAL AREAS OF ACTION EDUCATION HEALTH INFRASTRUCTURE (ROADS, ENERGY AND WATER) AGRICULTURE AND RURAL DEVELOPMENT GOOD GOVERNANCE, LEGALITY AND JUSTICE MACROECONOMIC AND FINANCIAL POLICIES B. OTHER AREAS OF ACTION EMPLOYMENT AND BUSINESS DEVELOPMENT SOCIAL ACTION HOUSING MINES FiSHERIES TOURISM PROCESSING INDUSTRY TRANSPORT AND COMMUNICATIONS TECHNOLOGY ENVIRONMENT REDUCTION OF VULNERABILITY TO NATURAL DISASTERS V. CONSULTATION PROCESS VI. MONITORING AND EVALUATION STRATEGY VII. BUDGETARY IMPLICATIONS ANNEXES OPERATIONAL MATRIX   I. INTRODUCTION 1. The Government of Mozambique, in order to provide continuity to its strategy to combat absolute poverty, hereby presents the Action Plan for the Reduction of Absolute Poverty (PARPA) 2001-2005. The PARPA 2001-2005 explains the strategic vision for reducing poverty, the main objectives, and the key actions to be pursued, all of which will guide the preparation of the State's medium-term and annual budgets, programmes, and policies. The PARPA 2001-2005 is also Mozambique's first Poverty Reduction Strategy Paper (PRSP). 2. This document is based on prior Government plans, including the Lines of Actionfor the Eradication ofAbsolute Poverty (1999), the PARPA 2000-2004 (Interim PRSP), and the Government Programme 2000-2004, as well as the sectoral and inter-sectoral plans, policies and strategies developed by organs of the State. 3. Production of the poverty reduction strategy has benefited from a process of consultations with relevant groups and segments of society outside the public administration. The sectoral strategies and programmes, the provincial poverty profiles, and the previous PARPA, all benefited from a broad and extended process of participation by civil society and external financiers. This document itself has also benefited in form and content from contributions resulting from the broad participation of civil society. Deeper discussions on the document as a whole, and on various parts, will continue, since planning is an iterative process of continuous improvement. 4. Given its medium-term focus, the PARPA is a rolling and dynamic programming instrument. This will enable the plan to incorporate new elements arising from changes in society and the economy. The key objective - the reduction of absolute poverty - will not be altered, but the tools, policies and targets may change as our knowledge of different variables improves. Therefore, the PARPA is an instrument defining policies and actions that will be periodically reviewed and perfected, involving an ongoing process of consultations. Central Objective 5. The central objective of the Government is a substantial reduction in the levels of absolute poverty in Mozambique through the adoption of measures to improve the capacities of, and the opportunities available to all Mozambicans, especially the poor. The specific objective is to reduce the incidence of absolute poverty from 70% in 1997 to less than 60% by 2005 and less than 50% by the end of this decade. Context 6. The fight against poverty has been going on since the first days of Independence, when high priority was given to expenditure on health and education, to improve human development. Significant investments were also made in rehabilitating basic infrastructure. Starting in 1987, the Government adopted a stabilisation and structural adjustment programme with the objective of 1  re-establishing production and improving incomes through deep reforms aimed at creating an economy based on private initiative and market forces. 7. There have been notable successes. Over the last five years Mozambique achieved annual economic growth rates of around 8% in real terms, in the context of economic and political stabilisation. Democracy and peace were strengthened, and the Mozambican people continued their struggle for progress. With stability and reforms, GDP per capita grew to US$ 230 in 2000, significantly improving incomes compared to five years earlier. 8. Nevertheless, these achievements have not resolved our grave social and economic problems. The country remains one of the poorest in the world, and poverty clearly remains as the key challenge facing the country. Our ability to address this challenge is still limited by a severe scarcity of resources resulting from a serious structural weakness of the economy. 9. Analysis of data from the Household Survey (IAF) of 1996/97 has provided a detailed profile of poverty in Mozambique. Nearly 70% of the population lives in absolute poverty, and there are notable urban-rural and regional imbalances. The IAF data also permitted an identification of the main determinants of poverty in Mozambique, namely: (i) slow growth of the economy until the beginning of the 1 990s; (ii) low levels of education of working age household members, particularly women; (iii) high dependency rates in households; (iv) low productivity in the family agriculture sector; (v) lack of employment opportunities within and outside of the agricultural sector; and (vi) poor infrastructure, especially in rural areas. 10. Besides suffering from acute material poverty, the poor in Mozambique also suffer from a high degree of vulnerability to natural disasters and economic shocks. This was made clear by the tragic floods, which afflicted the country in 2000 and 2001, as well as the privations caused in recent years by low prices in the international markets for the country's main agricultural products. 11. To deal with this situation of poverty, its determinants, and economic constraints, the government decided to develop a comprehensive and integrated poverty reduction strategy and establish a detailed process for preparation of the PARPA. Strategic Vision for Poverty Reduction in Mozambique 12. The poverty reduction strategy in Mozambique depends critically on the basic assumption of the maintenance of peace and socio-political stability. 13. The previous PARPA (2000-2004) emphasised measures that benefit the poor in the short term. The new PARPA (2001-2005) broadens the strategic vision by also recognising the crucial importance of medium and long-term measures to fight poverty through policies to sustain rapid and broad-based economic growth. 14. For a poor country such as Mozambique, rapid growth is an essential and powerful tool for poverty reduction in the medium and long-term. Without growth, the objective of increasing the capacities and expanding the opportunities 2  for the poor will continue to be severely constrained by the lack of public and private resources. Therefore, the strategy contains policies aimed at creating a favourable climate for stimulating investment and productivity, and achieving an average annual GDP growth rate of 8%. 15. The strategy also includes policies and programmes to ensure that growth is inclusive, so that the poor will benefit integrally. This, in the final analysis, will occur through greater access to assets (including improvements in human capacity) and the more efficient use of such assets by individuals, families and other institutions, especially in rural areas. A pro-poor growth strategy also requires a policy climate which stimulates the private sector to accelerate job creation and increase income generating opportunities through self-employment. In addition, components of the programme will be implemented taking into account the need for a better regional balance, with special attention given to regions with the greatest concentration of poor people. 16. The dynamics of human development and broad-based growth are interdependent. Thus, a strategy combining programmes that benefit the poor in the short-term, with policies that deliver benefits in the medium and long-term by promoting rapid and balanced growth, provides the best chance for creating a "virtuous cycle" of accelerated and sustainable progress in the fight against poverty. 17. In concrete terms, the poverty reduction strategy in Mozambique is based on six priorities aimed promoting human development and creating a favourable environment for rapid, inclusive and broad-based growth. The "fundamental areas of action" are: (i) education, (ii) health; (iii) agriculture and rural development; (iv) basic infrastructure; (v) good governance; and (vi) macro-economic and financial management. These areas for action are considered as "fundamental" because they are absolutely essential for reducing poverty and stimulating growth, and also because their impact is both deep and broad. The selection of priorities was based on the diagnosis of the poverty determinants in Mozambique, as well as studies on the issue of poverty reduction (in light of international experience), and consultations with civil society and the private sector. 18. Education is a basic human right. Its fundamental role for poverty reduction is universally recognised. Access to education contributes directly to human development by improving capacities and opportunities for the poor, promoting greater social, regional and gender equity. Without a doubt, knowledge is an indispensable means for improving the living conditions of Man. Education is also essential for rapid growth, as it expands the quantity and quality of human capital available for productive activities, and the ability of the nation to absorb new technologies. The main objectives in the area of education include achieving universal primary education, while rapidly expanding secondary education, informal education, and technical-vocational training. The programme also includes a commitment to combat HIV/AIDS through schools. Given the serious scarcity of technical and management capacity, which is an impediment to economic growth, the programme takes into account the necessity of expanding and improving the system of higher education. Thus, a programme for each level 3  of education is required, though the largest share of resources will be allocated to primary education. 19. The health sector also plays a fundamental role in directly improving the well- being of the poor, while at the same time contributing to rapid economic growth by improving the quality of human capital. The main objectives in the field of health include an expansion of, and improvement in, the coverage of primary health care through special programmes geared towards target groups such as women and children, a campaign to reverse the current growth of the HIV/AIDS epidemic, and greater efforts in the fight against endemic diseases such as malaria, diarrhea, tuberculosis and leprosy. The strategy also includes initiatives in related areas linked to health, such as the provision of water (infrastructure) and food security (agriculture). 20. More than 70% of the population is concentrated in rural areas, and an even higher proportion depends on agriculture for survival. Agriculture contributes around 30% of GDP, with most production coming from the family sector, which covers more than 3 million families. For this reason, agriculture and rural development is a priority in the strategy for poverty reduction and broad-based growth. The principal objective of rural development is to increase income- generating opportunities, especially for the family sector. The generation of income depends on agronomic advances to raise productivity, but also, and fundamentally, on access to markets. Rural development will benefit from actions to develop human capital and infrastructure, amongst others. It is also promoted through increased rural production, especially in agriculture. The expansion in production of the agricultural sector will be carried out with the support of rural extension programmes based on specific crops and technologies, as well as improvements in the financial system, whose role will analysed. The strategy for rural and agricultural development will also focus on food security policy, which is fundamental to reducing poverty and risks to the poor. 21. A fundamental role of the State in stimulating a market economy and expanding opportunities for the poor lies in the development of basic infrastructure. Improvements in the road network will permit better access to markets and a reduction in costs, and will facilitate communication and mobility, especially for those who live in rural areas and depend on agriculture. In parallel, the provision of water and energy is fundamental to the development of human capital and the expansion of national output. Priority in the rehabilitation and construction of basic infrastructure will be given to those areas of the country with the largest populations and highest levels of poverty. 22. Good Governance is a fundamental condition for success of the poverty reduction strategy. The quality of State institutions is very important for the provision of public services benefiting the poor. It is also a critical factor for achieving rapid and sustained growth. The current programme includes policies for promoting good governance in various forms, including: decentralisation and devolution of public administration, to bring government closer to the people; reform of public institutions so they can better respond to the needs of the people, starting with the simplification of bureaucratic procedures to eliminate "red tape"; programmes aimed at strengthening the capacity and efficiency of the legal and 4  judicial system, improving public safety, protecting the rights and freedoms of citizens, ensuring contract enforcement and facilitating the resolution of disputes; and developing a programme to reduce and contain corruption at all levels. 23. The last fundamental area of action is macro-economic and financial management. Experience internationally, and in Mozambique, shows that macro- economic management and financial development are basic requirements for creating an environment to stimulate rapid growth and poverty reduction. Principal priorities in this area include: (i) Fiscal, monetary and exchange rate policies to maintain low inflation and increase competitiveness of the economy. (ii) Policies to mobilise additional budgetary resources equitably and efficiently. (iii) Policies to improve the management of public expenditure. (iv) Policies to secure and expand financial markets, including a careful analysis of opportunities to expand financial services to rural areas and extend them to small and medium enterprises. (v) Policies to promote international trade. (vi) Policies to strengthen the management of domestic and foreign debt. 24. The strategy and action plan also cover, in less detail, complementary activities to those viewed as fundamental areas of action. The "other areas of action" include: selected social programmes (targeted social welfare programmes)', housing; sectoral policies and programmes that contribute to income generation and job opportunities (business development, fisheries, mining, industry, tourism); programmes to reduce vulnerability to natural disasters; and policies that support sustainable growth (transport and communications, technology, environmental management). 25. It may be noted that the poverty reduction strategy focuses on the principal determinants of economic growth - investment and productivity - through the creation of a favourable environment. In a market-driven development strategy, the main role of government in promoting investment and increasing productivity is covered by the priority areas cited above, in particular through investment in human capital, the development of infrastructure, programmes to improve the quality of public institutions, and policies for efficient macro-economic and financial management. Many of the complementary actions will also improve the investment climate and promote efficiency in specified areas. The critical issue of job creation is also dealt with through the priority programmes, in conjunction with measures to be taken in other complementary areas. 26. The strategic vision emphasises the reduction of absolute poverty defined in terms of material needs and lack of capacity and opportunities. At the same time, the strategy contains important features covering other basic dimensions of poverty, namely vulnerability and empowerment 27. The need to reduce vulnerability is addressed in various ways, including: all of the education and health programmes; the programme for agriculture and rural In principle, the social action plan should be a "fundamental area" of the poverty reduction strategy. In the current PARPA, however, it is classified as an "other action area" because resources available are still very limited and the number of beneficiaries remains relatively small in relation to the size of the problem. This situation should change as additional resources became available as a result of economic growth. 5  development; the programme to improve access by the poor to basic infrastructure and improve the quality of public institutions, especially in regards to the legal and judicial systems and public safety; the management of macro-economic and financial policies, to minimise the risk of economic shocks; a profound commitment to preserving peace; and subsidiary activities in the areas of social action, management of natural disasters, and environmental protection. 28. The emphasis on education and health also has a direct bearing on empowerment of the poor. Other features of the programme that contribute to empowerment include the development of farmer associations, the decentralisation and devolution of Public Administration, a commitment to increase transparency and deepen the process of participation in Public Administration, and a fundamental commitment to consolidate democracy and respect for human rights. Budgetary Implications 29. The medium-term fiscal scenario (CFMP) is a tool for identifying priorities and allocating budgetary resources to critical programmes to achieve the objective of poverty reduction. The CFMP has a prioritisation of activities reflecting detailed work carried out by the sectors. Budgetary expenditure focuses, firstly, on education and health, roads, water, rural development and agriculture, strengthening of the legal and judicial systems and public safety, and on capacity- building for public administration. Expenditure will focus on regions with the greatest population and highest levels of poverty. Consultations 30. Consultations carried out after the elaboration of the l' draft of the PARPA contributed significantly to the final version, in the following ways: (a) Confirming the choice of the fundamental areas of action. (b) Emphasising the importance of Good Govemance, in particular the issues of deconcentration and decentralisation to allow for participatory planning at the local level, in the districts. In participatory district planning lies the hope of a more effective consultations, involving a decisive process geared towards concrete action. This process will be more effective in involving the most disadvantaged strata of society in the determination of priorities to be pursued. (c) Highlighting the need for the State to be more effective in delivering on the options taken and the programmes, plans and commitments adopted. (d) Highlighting the importance of the fight against corruption and the need to change attitudes of the public institutions and their agents in relation to the citizens and their institutions. 31. In relation to the process of consultation referred to above, the following are worthy of note: 6  (a) A degree of fatigue was noted about the frequency of consultations on various issues. As a corollary, the consultations made clear that priority must be given to the implementation and delivery of action by the State. (b) Objections were raised about public policies in areas such as agricultural marketing, the promotion of agro-industries, the nurturing of a national entrepreneurial class, and the development of financial services (to reach the rural areas, the poorest and the nascent national entrepreneurial class). There was an appeal for greater State intervention in these fields. The issues raised deserve the attention of the State in the context of seeking to ensure the adequacy of the policies being pursued. Institutional framework of the PARPA 32. The PARPA is a medium-term programming instrument of the public planning system. The institutional framework is given in the tables on pages 8 and 9 that follow. 7  Public Planning System: Articulation of Instruments Government Programme Medium Term Programming Annual Operational Plans Integration of Objective! and Goals Action Plan for the Reduction of Economic and Social Plan Absolute Poverty (PRSP - PARPA) Co-ordination of the (PES) A L monitoring process Priority Resource . / Allocation Harmonisat on of sector /_._*__ AllocationStrateg and policy Integration of sector I F Objectives and goals Consistency of Budget Sector and Provincial with Policy objectives Strategic Plans and Actions Integration of sector Resource framework and Macro and Sector strategy and policy Inter-sectoral allocation Ser F~~~~~~~~~~~~ L_ Medium Termn Fiscal Framnework /, __. | Investment lirnits (internal Expenditure Limits State Budget (OE) and extemal resources) TrennialPublic Investment l _ l Programme l nvestinent Budget 8  INSTRUMENTS OF THE PUBLIC PLANNING TIMEFRAME PREPARATION APPROVAL SYSTEM MEDIUM-TERM PLANNING SECTORAL/PROVINCIAL FIVE/THREE YEARLY Sectoral ministries and Council of Ministers STRATEGIC PLANS coordinated by the MPF ACTION PLAN FOR THE FIVE YEARLY Sectoral ministries and Council of Ministers REDUCTION OF ABSOLUTE coordinated by the MPF POVERTY MEDIUM-TERM FISCAL FIVE YEARLY Sectoral ministries and Economic Council, Council of SCENARIO coordinated by the MPF Ministers THREE YEAR PUBLIC THREE YEARLY Sectoral ministries and Council of Ministers, Assembly INVESTMENT PROGRAMME coordinated by the MPF of the Republic OPERATIONAL PLANS ECONOMIC AND SOCIAL ANNUAL Sectoral ministries and Council of Ministers, Assembly PLAN coordinated by the MPF of the Republic STATE BUDGET ANNUAL Sectoral ministries and Council of Ministers, Assembly __________________________ coordinated by the MPF of the Republic 9  II. CHARACTERISTICS OF POVERTY: CONCEPTS AND PERCEPTIONS 33. What is poverty? 33.1 The reduction and elimination of poverty is the principal objective of development policies in the medium and long term. However, in discussing poverty, different ideas often arise as to its meaning. This is because poverty is a complex, multi-dimensional phenomenon with diverse characteristics. 33.2 As one can surmise, there has been a long debate about how to define poverty, and how to measure it. In the literature, poverty was defined for many years in terms of a lack of income (in money or in kind) necessary to ensure access to a set of "basic needs". With the passage of time the concept of poverty has been redefined to mean not only the lack of income, but also the lack of access to health, education and other services. In recent decades the definition of poverty has expanded to include aspects such as powerlessness, isolation, vulnerability, and social exclusion, amongst others. 34 Poverty as a multi-dimensional phenomenon 34.1 In the study on the poverty situation in Mozambique (Understanding Poverty and Well-being in Mozambique, First National Assessment: 1996-97), and in subsequent official documents (PARPA 2000-2004), poverty was defined as the inability of individuals to ensure for themselves and their dependants a set of basic minimum conditions for their subsistence and well-being in accordance with the norms of society. This same studies used consumption per capita (the total consumption of the family household divided by the number of members) as the basic measure of individual well-being for the following reasons: * Consumption is the most appropriate indicator or measure of actual wellbeing (while income is a measure of potential wellbeing given that it may or may not be used for consumption); * Consumption is a more precise and stable indicator since it is less subject to fluctuations over time.2 Therefore, individuals are classified as poor or not poor in terms of a poverty line defined in terms of per capita consumption.3 2For more details on this question see the baseline study Understanding Poverty and Well-being in Mozambique: First National Assessment (1996-97) (MPF, 1998). 3 A poverty line is arrived at by defining a particular level of income or value of consumption, such that those who fall below the line are considered poor. Households with incomes or consumption above the line are considered not poor. A monetary value is therefore assigned to this poverty line. 10  Box 1. What is poverty? Definition in Mozambigue: "inability of individuals to ensure for themselves and their dependants a set of basic minimum conditions necessary for their subsistence and well-being in accordance with the norms of society. "4 Other definitions: "the lack of income necessary to satisfy basic food needs or minimum calorie requirements" (Absolute or Extreme Poverty in terms of income); "Lack of sufficient income to satisfy the basic essential food and non-food requirements given the average income of the country" (Relative Poverty); "Lack of basic human capacities, such as illiteracy, malnutrition, low life expectancy, poor matemal health, prevalence of preventable diseases, together with indirect measures such as access to the necessary goods, services and infrastructures necessary to achieve basic human capacities - sanitation, clean drinking water, education, communications, energy, etc.-" (human poverty). 34.2 Despite consumption per capita being used as a measure of well-being, one must recognise the importance of indicators that are not based on consumption, such as rates of illiteracy, mortality rates, and percentage of population with access to clean drinking water, amongst others which are normally taken into account in broader definitions of poverty (see Box 1). These indicators, which together are used to determine the Human Poverty Index in accordance with UNDP methodology, are essential for highlighting various dimensions of poverty complete the poverty profiles. Studying the correlation between these and other variables contributes to a clearer picture of the aspects most relevant to the case of Mozambique. 35 Qualitative perceptions 35.1 Poverty surveys often focus on defining poverty lines and profiling on the basis of data collected through conventional surveys, with an emphasis on quantitative aspects and averages derived from a sample of households. However, it is also very important to determine how these profiles correspond to what the people understand as constituting poverty, and their perceptions, obviously subjective but based on direct experience, of what it means to be poor. Other qualitative and participatory survey methods take into account these dimensions of poverty that are frequently overlooked. Participatory Poverty Diagnoses and Participatory Rural Appraisals, based on quick and low-cost surveys, are normally used to carry out this kind of research. 35.2 In Mozambique a Participatory Poverty Appraisal (APP) was undertaken, based on a series of participatory qualitative studies carried out in 18 districts, which preceded the National Poverty Assessment based on the Household Survey (IAF) of 96/97. These studies, organised by the Population Studies Centre at Eduardo 4 Official definition adopted from the poverty analysis study. 11  Mondlane University (CEP-UEM) in collaboration with the Ministry of Planning and Finance and other institutions, were carried out between 1995 and 1996. In January 2001, a series of participatory diagnoses of poverty, organised by the MPF in collaboration with the UEM, were carried out. 21 districts in 7 provinces (Cabo Delgado, Nampula, Zambezia, Sofala, Tete, Inhambane, and Maputo province). Only the preliminary results of these diagnoses are currently available. The final results will be analysed and taken into account in the development of future profiles and poverty reduction plans. At the moment, it is possible to state that no significant differences were found between the first and the latest evaluations. Some results of the participatory studies follow. 35.3 The development of a poverty profile was complicated by the existence of a widespread hegemonic discourse that "all were poor". However, in terms of perceptions of poverty, variations were found depending on those being surveyed. The factors mentioned in determining whether a person was considered poor or not may be divided into three groups: material goods; sources of income; and social relations/civil status. Distinctions were also drawn between the poverty of individuals, families, villages and regions/districts. Communities frequently referred to three groups (the poor, the middle group and the rich), though the material possessions characterising each group varied significantly between different communities. Hence, the perceptions of stratification and poverty were related to the surrounding socio-economic and cultural context. In terms of the frequency of the factors mentioned, for most people the poor are those who are unable to work because of physical handicaps, age or illness or civil status (widows); in other cases poverty is seen to be a case of working very hard but suffering from low income due, for example, to adverse climatic conditions and lack of means, or to be associated with one or another extreme symptom such as famine. In addition, there were views of poverty based on family origin, as a kind of predestination. But there is also a more materialistic understanding of poverty, seen as the lack of essential goods such as sufficient food or clothing, or not having children, or not having an alternative source of income. In the most recent diagnoses, the aspects most frequently mentioned as contributing to social status (poverty/wealth) were, amongst others, the following (in descending order): ownership of livestock; employment (working for others); ownership of a bicycle; ownership of a car; ownership of clothing and its quality; ownership of money, access to credit and goods to sell; ownership of a shop or stand; access to foodstuffs (quantity and quality). On the other hand the target groups, when discussing the poverty of communities (and not of individuals), also pointed out the lack of basic infrastructure, (roads, markets, transport), low agricultural productivity and the lack of welfare organisations as both symptoms and causes of poverty. 35.4 With regard to causes, in the 1996 assessments there was frequent reference to the war as a determining factor in explaining the current conditions of the poorest segments of the population. "Ignorance" and adverse climatic conditions were also considered to be causes of poverty. In the most recent diagnoses, less emphasis is given to the war, and more reference is made to the lack of support (social support, support from the Government and other institutions), the lack of employment opportunities, limited access to financial services, problems with marketing, or inability (due to physical or mental factors) to work, as the principal causes of poverty. The phenomenon of displacement (a result of the war) and the closing of 12  factories were also highlighted. The latter was particularly raised in areas where employment in factories was until recently the major source of cash income. In both sets of diagnoses other explanations were put forward, such as divine will, which meant that a person was predestined to be either poor or rich. At the same time there appeared to be concentric circles of poverty and misery. In fact, the closer one is to the administrative headquarters there is more infrastructure, a greater government presence, more food aid and the further away one is from the administrative headquarters the less infrastructure, shops, less government presence, in sum greater abandonment. 35.5 Finally, in terms of priorities, differences in established priorities may be linked to the age of those interviewed, their gender, wealth, and position in the power structure. The poorest tend to give priority, in terms of development activities or problems that need to be dealt with as a matter of priority, to the following in descending order of importance: transport, roads, prices, marketing of products, access to land and water and the introduction of motorised flour mills. These are, in sum, activities that increase the returns from their agricultural production, improve their terms of trade or their chances of participating in the market, and reduce the need for labour, which is always a limiting factor. The most recent diagnoses emphasise the need for basic social services, in particular the lack of accessible health posts, the lack of means to transport the seriously ill (ambulances), and the lack of personnel. There were accusations of corrupt practices amongst school teachers in some areas. The importance of water supply was constantly referred to in the districts visited. In many cases this problem is a result of the non-functioning of existing infrastructure due to the lack of maintenance of the equipment (such as pumps) and the lack of fuel. 36 Why and how to measure povet 36.1 The measurement of poverty is one of the first steps, which should guide the definition and implementation of policies, and allow comparisons to be made between areas and over time. Comparisons between poverty levels can be made between two different moments in time (for example, compare 1995 and 1985) or the effects of poverty on two sub-groups of the population (for example rural and urban) or between two different situations in terms of policies adopted (for example, with or without subsidies). 36.2 In the National Assessment, basic minimum conditions were identified in terms of an absolute poverty line, constructed as the sum of a food poverty line based on nutritional standards of approximately 2,150 calories per person per day, plus a modest amount of non-food expenditure based on consumption by households that suffer from food insecurity. In monetary terms, the national poverty line was fixed at 5,433.00 MT per person per day, after weighting the various provincial poverty lines and adjusting them to reflect variations in the cost of living.5 In addition, the concept of destitution (or abject poverty) was considered, characterising those who are unable to satisfy their daily calorie requirements (representing 60% of the base poverty line) (MPF 1998). 5 For example, a person living in Maputo city needs to spend 2.5 times more on daily consumption (for the basket of goods chosen for the poverty line) than a person who lives in a rural area in Nampula. 13  36.3 Three measures were used in the assessment studies on absolute poverty in Mozambique: * Headcount index * Poverty gap index * Squared poverty gap index The headcount index measures the proportion of the population defined as poor, i.e. those whose consumption (or any other indicator used to measure living standards) is below the poverty line. The poverty gap index is defined in terms of the average distance below the poverty line expressed as a proportion of that line, where the average is based on the total population, and counting the non-poor as having a poverty differential equal to zero. With this second index it is possible to measure changes in the income of the poor (increases or decreases) even when the headcount index remains unchanged. The third measure, the squared poverty gap index is defined as the average of the square of the proportional poverty deficits, and is sensitive to inequalities between the poor and to changes in these inequalities. THE PREVALENCE OF POVERTY 37. Results of the analysis of data from the IAF of 96-97 indicate that poverty levels in Mozambique remain extremely high. As one can see in Table 2.1, the average monthly per capita consumption was estimated at 160,780.00 MT, or around US$ 170 per annum using the prevailing exchange rate at the time of the survey. The rate of absolute poverty stood at 69.4%, which meant that more than two-thirds of the Mozambican population was living below the poverty line. There was a greater incidence of poverty in rural areas (71.2%) where 80%6 of the population lives, as opposed to 62.0% in urban areas. 38. As one can see from Table 2.2, given the concept of poverty used, the incidence of poverty at the provincial level is most marked in Sofala (87.92), Inhambane (82.60) and Tete (82.27). The prevalence of destitution (abject poverty)7 follows the same pattem, being highest in Sofala (65.19), Tete (53.60) and Inhambane (53.73). Maputo city shows the lowest levels of both poverty and abject poverty, while the least poor province in terms of per capita consumption is Cabo Delgado. 39. Table 2.3 presents non-consumption welfare indicators that are important in themselves. These indicators show that three out of every four Mozambican women are illiterate, that the vast majority of the population does not have access to clean drinking water, and that infant and maternal mortality rates remain high. There is substantial inequality between the welfare levels in urban and rural areas, given that the welfare indicators are significantly lower in the latter. 40. Table 2.3 also presents comparisons between the prevalence of absolute poverty based on average consumption per capita, and the prevalence of human poverty based on other indicators of wellbeing (life expectancy, education, sanitation, clean 6 This percentage is based on the IAF figures. According to the Population Census, the percentage of the rural population is 71.42%. 7 The abject poverty line (or destitution) is defined as 60% of the poverty line, and reflects a minimum level of subsistence. 14  water, etc). In fact, some provinces show large differences in terms of their position in relation to the consumption poverty index and the human poverty index (see Box 1 for definitions). For example, Cabo Delgado is the only province where the human poverty index is higher that the consumption poverty index, while Sofala, Inhambane, Tete and Maputo provinces show the highest relative differences between the index based on consumption and the human poverty index (between 25-35 percentage points higher in terms of consumption). Logically, these differences may be partly explained by the existence of, and access to, social infrastructure (education and health) and in the proportion of the urban population. THE POVERTY PROFILE 41. Decision-makers and planners need trustworthy and up to date informnation on poverty - for example, the number of households living in poverty, their composition in terms of age and sex, their geographic location and their employment situation, amongst other aspects. 42. The poverty profile provides a quick synthesis of the situation to politicians and those responsible for planning, as well as other interested persons. It has proved to be a useful instrument for the development of effective policies and programmes. Therefore the profile of characteristics identifying the poor includes location (rural or urban), occupation (farmers or workers), gender (households headed by men or women), and employment situation (employed, unemployed, casual workers). 43. National Profile The profile presents the scale of poverty and its distribution amongst various socio- economic groups, and provides information on the characteristics and heterogeneous nature of the poor.8 In general there are important differences between urban and rural areas, with the latter being poorer and less endowed with basic infrastructure. In brief, the profile highlights the following principal points (see Table 2.4): 43.1 Demographic characteristics. Poor individuals tend to live in larger households that the non-poor. The poor have more children that the non-poor and start having children earlier. Poor households have approximately twice as many dependants as the non-poor, and so their dependency rates are significantly higher than non-poor households. In regards to the feminisation of poverty, data from the IAF does not support this hypothesis a priori, given that there is not a higher proportion of female-headed households amongst the poor when compared to non- poor households.9 Only in the case of widows and divorcees in urban areas is there a significant difference, which supports the gender hypothesis. Nevertheless, multiple regression analysis does tend to indicate that the link between the gender of the household head and consumption per capita (controlling for other variables) This analysis is based on bivariate correlations between poverty and each of the variables taken into consideration. 9 In fact, estimates of the incidence of poverty by group, according to the gender of the head of household, indicate that households headed by men have a poverty prevalence rate of 69.9%, while those headed by women have a slight lower rate (66.8%), which is principally due to their relative presence in rural areas (72.1% for male-headed and 66.9% for female-headed). 15  is negative in the case of female-headed households. Gender disparities, above all in rural areas, are a notable feature in many countries. This implies the need to deepen the analysis of the poverty profiles in Mozambique through a gender perspective, hopefully producing information that will improve policies for fighting against poverty.'I 43.2 Education. The relationship between education and poverty is significant. However, differences between the poor and non-poor are often less significant than between sexes and between place of residence. Women and rural areas tend to be worse off. This means that a poor child in an urban area has a better chance of attending school than a non-poor child in a rural area; and in rural areas, a poor boy has a stronger probability of attending school than a non-poor girl. There is also a strong relationship between the level of education of the household head and the state of poverty of the household: households headed by those with a higher level of education tend to be less poor. This relationship is particularly strong in urban areas and in female-headed households. 43.3 Health and nutrition. In rural areas, access to health services is approximately the same for the poor as the non-poor. However, in urban areas the non-poor have better access to formal health services than the poor. The proportion of children between 6-11 months who have not been vaccinated is significantly higher in rural than in urban areas. Chronic malnutrition amongst children below the age of 5 is also higher in rural areas. 43.4 Agriculture and land ownership.'1 Almost all rural households have access to at least a plot of land to farm (machamba), compared to less than half of urban households. The poor and non-poor have approximately the same amount of land per household, but the non-poor tend to use more equipment (inputs) and to have more irrigated land than the poor. Nevertheless, the use of equipment and inputs is very low and this is reflected in low levels of agricultural productivity in the country. Land is not, therefore, a limiting factor for poor peasants, but rather their capacity (and therefore means of production) to work the land which they have, and achieve acceptable levels of productivity. Maize and cassava are the most common crops for both the poor and non-poor, while cash crops are of relatively marginal importance: the non-poor tend to cultivate more cotton and cashews, but nevertheless only low percentages of the non-poor do so (6.1% and 26% of the non- poor, respectively). This is reflected in the low levels of marketed output with, in general, less than 10% of households in each category (poor and non-poor) selling surpluses of maize, cassava or cotton. 43.5 Employment. While in urban areas the non-poor tends to work more for wages than the poor, in rural areas there is difference in this respect. In these areas it is not so much employment itself, but other factors such as the amount of wages or the number of dependants, that are the main determinants of poverty. In rural areas '