42996 How Parliamentarians Can Help Papers Ensure Accountability for Spending on HIV/AIDS and Reproductive Health Amanda Glassman orking W WBI How Parliamentarians Can Help Ensure Accountability for Spending on HIV/AIDS and Reproductive Health Amanda Glassman World Bank Institute Washington, D.C. Copyright © 2007 The International Bank for Reconstruction and Development /The World Bank 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. The World Bank enjoys copyright under protocol 2 of the Universal Copyright Convention. This material may nonetheless be copied for research, educational or scholarly purposes only in the member countries of The World Bank. This paper has not undergone the review accorded to official World Bank publications. The findings, interpretations, and conclusions expressed herein are those of the author(s) and do not necessarily reflect the views of the International Bank for Reconstruction and Development / The World Bank and its affiliated organizations, or those of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. How Parliamentarians Can Help Ensure Accountability for Spending on HIV/AIDS and Reproductive Health Amanda Glassman 2007. 35 pages. Stock No. 37270 Contents Acknowledgements...................................................................................................................................... iv Acronyms ...................................................................................................................................................... v Executive Summary..................................................................................................................................... vi I. Introduction...................................................................................................................... 1 II. Budgetary process............................................................................................................ 2 III. Public policy for HIV/AIDS prevention and treatment................................................ 4 Table 1: Potential indicators for parliamentary oversight of HIV/AIDS programs................................... 6 IV. Available expenditure tracking tools............................................................................ 12 A. Domestic resource tracking tools.................................................................................. 12 Table 2: IDASA Report on HIV/AIDS Allocations in South Africa .......................................................... 13 B. Medium Term Expenditure Frameworks (MTEFs)................................................... 13 C. National Health Accounts (NHAs) and sub-accounts................................................. 14 D. Public Expenditure Reviews (PER) and Tracking Surveys (PETS).......................... 15 Table 3: Review of PETS Results for Africa in the 2000s (Lindelow et al 2006) ..................................... 16 E. Censuses and surveys..................................................................................................... 17 F. Donor accountability requirements.............................................................................. 17 V. What can parliaments do? ............................................................................................ 17 VI. Case Example: Rwanda................................................................................................ 19 Table 4 Indicators for Rwanda Case Example......................................................................................... 21 References ................................................................................................................................................... 24 Annex: Civil Society Organizations that Watchdog Budgets in Africa ................................................. 28 iii Acknowledgements* This paper was prepared by Amanda Glassman (consultant) under the general direction of Tom Merrick (WBIHD) and Isabel Rocha Pimenta (WBIHD). Valuable guidance and comments were provided by Frederick Stapenhurst (WBIPR), Rene Bonnel (HDNGA), Festus Wachira Kingori (WBIHD-Secondee from Kenya Ministry of Finance), John Armiger (WBIHD)and Tony Worthington (former MP, UK, and member of the board of Parliamentary Network on the World Bank). * Views expressed in this paper are those of the author only. iv Acronyms AIDS Acquired Immune Deficiency Syndrome ART Anti-Retroviral Therapy BIS Baseline Indicator Set for Procurement CSO Civil Society Organization DHS Demographic and Health Survey ECOWAS Economic Community of West African States GDP Gross Domestic Product GF Global Fund for AIDS, TB and Malaria HIPC Highly Indebted Poor Country Initiative HIV Human Immunodeficiency Virus IMF International Monetary Fund LSMS Living Standards and Measurement Survey MAP World Bank Multi-Country HIV/AIDS Program MP Member of Parliament MTEF Medium-Term Expenditure Framework NHA National Health Accounts OECD/DAC Organization for Economic Cooperation and Development/Development Assistance Committee OOP Out of Pocket Spending (Household Spending) PEPFAR President's Emergency Plan for AIDS Relief PER Public Expenditure Review PETS Public Expenditure Tracking Survey PLWHA People Living With HIV and AIDS PRSP Poverty Reduction Strategy Paper SES Socio-Economic Status THE Total Health Expenditure UNAIDS Joint United Nations Program on HIV/AIDS UNGASS United Nations General Assembly Special Session on HIV/AIDS USAID United States Agency for International Development v Executive Summary HIV/AIDS is an important problem, destroying the lives and livelihoods of millions of people. UNAIDS estimates that 40 million people are living with HIV; more than half are in sub-Saharan Africa. In response to this crisis, governments are increasing spending on HIV and AIDS policies and programs, primarily in response to a massive influx of external financing. Increased funding can skew priority-setting inside and outside the health sector and put enormous pressure on countries' capacity to effectively use the funds. Increased funding also creates a context where the risk of misuse of funds can be high, particularly as current program performance is essentially measured by the magnitude and speed of disbursement. Bribes, corrupt officials and inappropriate procurement undermine health care delivery in much the same way they do for police services, law courts and customs whose functions become compromised by the culture of poor governance and corruption. Parliaments are increasingly asserting their independence from the executive branch and actively performing their constitutionally-designated functions of overseeing the executive, representing citizens, making policies, and enacting laws. Legislative oversight is nowhere more important than over the budget. Oversight of budgets and expenditures by parliament can potentially improve effectiveness and reduce risks of misuse and inefficiency. In the case of HIV/AIDS, parliaments can ensure that HIV/AIDS funds flow through the budget and that the amount, distribution and utilization of funding are consistent with national policies and priorities as well as international best practices. This note provides an overview of the budgetary process with specific reference to allocations for HIV/AIDS prevention and treatment and highlights potential trouble-spots where the exercise of Parliamentary oversight could improve financial performance as well as improve outcomes. Parliamentarians can make more effective use of existing tool for oversight. Using these tools, MPs and parliamentary committees can scrutinize how allocated funds are spent, the reasons behind under-spending allocated funds, and determine whether the total amount allocated to HIV/AIDS within the health budget is appropriate. Specific Parliamentary actions are discussed, and a country case example (Rwanda) is presented as an illustration of how oversight tools can be applied. vi I. Introduction The problem. HIV/AIDS is an important problem, destroying the lives and livelihoods of millions of people. UNAIDS estimates that 40 million people are living with HIV; more than half are in sub-Saharan Africa. The epidemic devastates families through the loss of adults in their productive ages, particularly women, while foregone human capital formation affects economic growth. In response to this crisis, governments are increasing spending on HIV and AIDS policies and programs, primarily in response to a massive influx of external financing. The IMF reports that HIV/AIDS resources flows were US $5 billion in 2003 and US $8 billion in 2004 and that three-quarters of this financing is external. Despite these increased contributions, there is still a large gap between what is needed and what is provided. At the global level, UNAIDS (2005) estimates that up to $18.8 billion annually will be required by 2007. Further increases in funding will necessary as the numbers of infected persons increase and the duration of their lives and demands for medical care rise. Yet the magnitude and characteristics of available financing for HIV/AIDS prevention and control poses its own challenges. The boom in external funding for HIV/AIDS is unprecedented. From 2000-02 to 2002-04, the average level of external funding increased dramatically in Lesotho (1,100 percent), Swaziland (951 percent), Tanzania (394 percent) and Zambia (698 percent) (Lewis 2006). External resources for HIV/AIDS now dwarf public expenditure on health in the most affected countries, and, in some cases, have been at the expense of the remainder of the health budget. In 2003/4, for example, Ethiopia's external flows were equal to the government's health budget, but in Uganda and Zambia, AIDS funds exceeded all public spending on health by almost 185 percent (World Bank 2005). As Lewis (2006) asks, "how [can] countries...accommodate and wisely allocate new resources for HIV/AIDS while their overall health spending declines?" The channeling of the majority of external financing outside of the budget -- Foster (2005) has estimated up to 80 percent of the total -- also creates problems for sound health and fiscal policy. These pressures create a context where the risk of misuse of funds can be high, particularly as current program performance is essentially measured by the magnitude and speed of disbursement (Tayler and Dickinson 2006). Bribes, corrupt officials, and inappropriate procurement undermine health care delivery in much the same way they do for police services, law courts, and customs whose functions become compromised by the culture of poor governance and corruption. Specific characteristics of AIDS treatment create additional incentives for misuse; expensive ARV treatment has been counterfeited or resold, or subject to excessive mark ups or other corrupt or extortive practices. Compounding these challenges is the fact that the prevention and treatment of HIV/AIDS places exceptional demands on the health system at all levels. Testing, requirements for rigid adherence to the treatment protocol1, treatment of opportunistic infections, 1A 2004 World Bank study in India found that if drugs are not administered according to strict protocols, resistant strains of the virus can develop, jeopardizing treatment options for all (Over et al 2004). 2 Glassman management of ART side effects, palliative care, and psychosocial support require patient-centered care. Effective patient-centered care involves coordination among functions and facilities, near perfect logistics systems and well-staffed and stocked facilities, both at primary and hospital levels. As programs become more successful, AIDS patients live longer with ARV and require significantly more medical care over their lifetimes, therefore increasing the pressure on health facilities to meet the new demand. Further, there is continuing pressure to apply more sophisticated technologies and treatment when first-line ARV regimens fail (Harries et al 2006). Finally, the emphasis on resource mobilization and quick disbursement neglects institutional and governance issues. HIV and AIDS programs rely on the existing infrastructure of physical and human capital. Where delays occur because of lack of a critical input ­ such as staffing, infrastructure, management, warehousing, logistics and information on performance and output ­ productivity is reduced and impact falters. Without attention to effectiveness, well-intentioned spending may have no impact. In summary, increased resource mobilization for the prevention and control of HIV/AIDS is vital, but it is equally important to ensure that health systems function so that services are delivered and health professionals are accountable to the public, government and, where they are providing funds, donors. The role of parliament. Parliaments are increasingly asserting their independence from the executive branch and actively performing their constitutionally-designated functions of overseeing the executive, representing citizens, making policies, and enacting laws. Legislative oversight is nowhere more important than over the budget. However, there is much work to be done to improve budgetary oversight by parliaments. A recent survey of African legislators found that parliaments are not taking full advantage of their constitutionally mandated powers to address the HIV/AIDS crisis (National Democratic Institute for International Affairs 2004). Oversight of the budgetary process is often limited by lack of know-how, capacity and demand for disaggregated budgetary data on the part of legislators, but also by budgetary complexity and a lack of transparency and detail in executive branch budgetary reporting. Oversight of budgets and expenditures by parliament can potentially improve effectiveness and reduce risks of misuse and inefficiency. In the case of HIV/AIDS, parliaments can ensure that HIV/AIDS funds flow through the budget and that the amount, distribution and utilization of funding are consistent with national policies and priorities as well as international best practices. II. Budgetary process Budgets are regularly produced financial plans of government comprising revenues and expenditures of the state. If resources were limitless, there would be no need to budget. In reality, governments have to make choices about the allocation of scarce resources to meet competing needs in society. Budgeting is effective in facilitating this process when it forces awareness of overall fiscal constraints, enables the prioritization of spending in line with policy objectives, and supports the efficient implementation of policies. How Parliamentarians Can Help Ensure Accountability 3 Wehner (2004) provides a useful overview of budgetary process and the role of key players; the section draws on his work. Despite the critical role of the budget process, as of 2004 only 10 of 66 low-income aid recipient countries have a moderately functional budget system according to the World Bank's Country Policy and Institutional Assessments2, suggesting many countries have poor budgetary data and practices (World Bank 2006). The highest performing countries in Africa are Benin, Burkina Faso, Mali, Tanzania, and Uganda. The remainder of countries scored at or below 3. These institutional weaknesses imply that the oversight task is complicated by the near total absence of budgetary data in an utilizable format. Key actors in the budgetary process. The Finance Ministry or Treasury coordinates and leads the budget process. This role involves preparation of short and medium-term macroeconomic projections which forecast the resources from taxes, donations and other sources that will be available for spending, formulating fiscal policy, guiding the drafting of the budget, and monitoring budget implementation. Spending departments or ministries are responsible for expenditures within their jurisdiction, such as health, education, family welfare and so on. Spending departments will often try to extract as many resources as possible to meet their goals, whereas the role of the finance ministry is that of the guardian of the public purse. Democratic constitutions require taxation and public spending to be approved by parliament. Therefore, the role of the legislature is to scrutinize and authorize revenues and expenditures, and to ensure the budget is properly implemented. Independent supreme audit institutions such as auditor generals or audit courts carry out an audit of government accounts in order to determine whether government did in fact implement the budget as passed by the legislature. Some of them also consider whether this was done efficiently and effectively. There is considerable heterogeneity in the institutional design, capacity, resources, and types of audit conducted, but the goal is to support the executive branch managers, the legislature, and citizens in the oversight of spending. Stages of the budget process. The budget process is governed by a timeline that is separated into four stages: (i) drafting; (ii) legislative; (iii) implementation; and (iv) audit and evaluation. The figure at the top of the next page describes these stages and the institutional roles associated. 2 Country Policy and Institutional Assessments (CPIAs) are carried out annually by the World Bank. CPIAs rate the quality of policies and institutions, including those related to budgeting. Bank staff score individual countries along an absolute 1 (worst) ­ 6 (best) scale in 16 different governance categories. Several rounds of scoring are used to ensure consistency in scoring across countries. A score of 4 or higher indicates a moderately functional system. 4 Glassman Stages of the annual budget process Finance ministry or Budget tabled in the Funds apportioned to treasury issues legislature spending departments to Supreme audit institution guidelines to spending Consideration by implement activities assesses departmental departments or agencies parliamentary Finance ministry accounts and Spending departments committee(s) monitors spending performance submit draft budgets Parliament accepts, Request for legislative Audit reports published Negotiation and final amends or rejects the approval of adjustment and reviewed by decisions by executive budget budget if necessary parliament Drafting Legislative Implementation Audit Before beginning of relevant fiscal year Fiscal year starts and ends Following end of fiscal year Source: Wehner 2004. Budgeting is a process rather than an event and budget cycles are ongoing and interconnected. The role of parliament should not be restricted to budget approval and the review of audit findings. For instance, in some countries, parliamentary committees ask government to report on the process of drafting an upcoming budget yet to be tabled, and legislators might request certain documentation that is used in the drafting process. During budget execution, the legislature should have access to actual revenue and expenditure data on an ongoing basis in order to keep track of progress in implementation of the approved budget. This regular monitoring allows for earlier identification of problems in budget execution, before they result in significant deviations between the approved budget and actual revenues and spending. When parliamentarians follow the entire budget process as it unfolds they will be in a position to acquire relevant expertise and to keep track of emerging issues. Legislative effectiveness in budget scrutiny is enhanced by continuous oversight. III. Public policy for HIV/AIDS prevention and treatment While the search for a vaccine or cure for HIV/AIDS continues, the greatest hope for combating the disease is through the prevention of transmission. A well-known World Bank report (Over and Ainsworth 1997) found that public policies directed to correct for three types of market failures could be effective in the reduction of transmission rates. In keeping with the principles of public economics, the report suggested that governments finance or directly implement interventions that are essential to stopping the spread of HIV, but that private individuals or firms will not have sufficient incentive to finance on their own. In particular, the Bank recommended the provision of public goods such as information, the reduction of the negative externalities of behavior that spreads HIV such as unprotected intercourse and unsafe injecting practices, and the protection of the poor from HIV infection through, for example, subsidy of condoms and treatment of sexually transmitted infections. This strategy was thought to prevent the largest number of subsequent infections. How Parliamentarians Can Help Ensure Accountability 5 Yet while cost-effective prevention plays an important role in stemming the epidemic, supporting those already infected in living healthier, longer lives is also crucial to minimizing the economic impact of the epidemic, and the international consensus is that the two need to advance in parallel. Until recently, life-prolonging treatment was available only to a tiny fraction of HIV-positive people in Africa. However, the emergence of a simpler treatment regime, the dramatic drop in the cost of ART, an international consensus on a medical treatment protocol for resource-limited settings, and increased international funding for ART in low-income countries have made scaling up of programs a potential reality. However, the challenge of scaling up from current initiatives to the comprehensive prevention and treatment programs needed in Africa poses significant logistical and support problems. The challenges noted by the United Nations' Economic Commission for Africa's Commission on HIV/AIDS and Governance in Africa (2004) include: (i) low and declining numbers of health professionals; (ii) high dependence on external financing for ART; (iii) inadequate laboratory and patient care infrastructure; (iv) poor patient follow-up leading to low adherence and the risk of emergence of drug-resistant strains; and (v) drug shortages and delays in drug distribution. Each area represents a special need for oversight and getting the balance right among program priorities more of an art than a science. Role of public policy and expenditures in achieving better results. As noted earlier, the budget is a government's most powerful economic policy instrument, and therefore it's major opportunity to influence the HIV and AIDS epidemic. The information and analyses required to effectively oversee public financing for health, including HIV and AIDS programs, can be divided into five broad areas: (i) level of spending; (ii) allocation of spending; (ii) execution of spending; (iii) inputs purchased by spending; (iv) outputs produced by inputs; and (v) outcomes produced by program (see Table 1). This paper focuses on oversight of (i)-(iii) and refers the reader to alternative sources for (iv)-(v).3 Analysis of a country's HIV and AIDS programs will often require examining more than one budget, as well as addressing questions about which actors are shaping HIV and AIDS policy and which ones are delivering the services. External financing for HIV and AIDS programs, whether project-specific or general budget support, is important to consider. Health systems in many African countries rely heavily on these external funds. 3Although this note is focused on budgetary oversight, it should be noted that there are many areas where parliamentarians might take an activist role with respect to HIV/AIDS legislation. Such issues include: discrimination against vulnerable groups in the workplace, schools and health care facilities; equal access to health care; voluntary testing and informed consent in HIV testing; confidentiality of medical information; partner notification of HIV status; regulation of blood safety standards; regulation of HIV- related goods and services; intentional exposure or transmission of HIV; HIV prevention and care services in prisons; gender equality with respect to property rights and marital relations; care and support of orphans; protection of rights of human participants in HIV/AIDS research; among others. 6 Glassman Table 1: Potential indicators for parliamentary oversight of HIV/AIDS programs Area Objective Indicators (potential sources of data described below) Level Assure that the level of spending· % GDP per capita allocated to health is sufficient to meet goals set, · % total government budget allocated to health while maintaining fiscal and (Abuja target) macroeconomic stability as · % health budget allocated to HIV/AIDS and RH defined in MTEF · Budgeted expenditure per PLWHA or per woman of reproductive age Allocation Assure that budget allocations · Correlation of HIV/AIDS budgetary allocations correspond to legislated and number of PLWHA (or STI cases or other priorities and distribution of proxy) by geographic region need as expressed through, for · HIV/AIDS budget by type of intervention example, HIV prevalence rates (prevention, treatment, etc.) or poverty rates · HIV/AIDS budget by function (human resources, capital, drugs and other inputs) · HIV/AIDS budget by type of provider (public, private, not for profit) · HIV/AIDS budget by type of beneficiary (poor, orphans, other vulnerable) Expenditures Assure that budget expenditures · % expended/budgeted are executed as budgeted; assure · Correlation of HIV/AIDS expenditure and that the poor are protected from number of PLWHA (or STI cases or other catastrophic expenditures proxy) by geographic region associated with HIV/AIDS · HIV/AIDS expenditure by type of intervention health care (prevention, treatment, etc.) · HIV/AIDS expenditure by function (human resources, capital, drugs and other inputs) · HIV/AIDS expenditure by type of provider (public, private, not for profit) · HIV/AIDS expenditure by type of beneficiary (poor, orphans, other vulnerable) · HIV/AIDS out of pocket spending per PLWHA by SES Inputs Assure that inputs purchased are · Baseline indicators on procurement done so efficiently and with no (www.oecd.org/dataoecd/12/14/34336126.pdf) misuse of funds. Outputs Assure that inputs have · UN General Assembly recommended indicators produced the desired outputs. (% trained, % receive services, etc.) (http://data.unaids.org/publications/irc- pub06/jc1126-constrcoreindic-ungass_en.pdf) · preventive service utilization frequency per capita · characteristics of service users (risk category, SES, gender, etc.) Outcomes Assure that outputs have · UN General Assembly recommended indicators produced the desired outcomes. (http://data.unaids.org/publications/irc- pub06/jc1126-constrcoreindic-ungass_en.pdf) The problem is that donor funds don't always flow through the budget, particularly in the case of project-specific activities, making this funding stream difficult to oversee. A How Parliamentarians Can Help Ensure Accountability 7 recent study of 14 countries receiving poverty reduction support credits found that only 20 percent of donor commitments were provided as general budget support. Fully 50 percent of assistance ran outside the budget, and 30 percent escaped government reporting altogether (Foster 2005). This phenomenon is troubling from a good governance perspective; in addition, aid that flows outside of budgets has been found to result in uneven access and in financing low priorities when compared to national development plan priorities. In spite of these issues, those working in HIV/AIDS have found that partnering with non-governmental agencies and the private sector is key to an effective response to the epidemic. However, it is entirely feasible to partner with these groups while still channeling resources through government budgets (although it does run into the Ministry of Finance concerns described in Box 1). Box 1: Guardians of the public purse The role of finance ministries as guardians of the public purse is frequently politically unpopular, particularly among ministries and stakeholder groups working to prevent and control the AIDS epidemic. Why? Simply, the problem is urgent yet funds are insufficient, volatile and unpredictable. The nature of the fiscal cycle requires that loan and grant financing fit within the medium-term resource envelope that is available to government, that is, revenues should at least equal expenditures in a given fiscal year. Even if budgets are balanced in the current year and grant financing is accommodated within a budget ceiling established for the health sector, a finance ministry may refuse external assistance for AIDS programs because a one-time grant implies scaled-up expenditure on human resources or drugs that the government has no means to independently finance and cannot be easily downsized in the future (Heller 2006). Further, even if aid commitments are long-term, the unpredictable timing of external financing can make additional human resource contracting impossible in contexts where fiscal space is limited (Lewis 2006). Finally, when spending departments or ministries under-spend with respect to budget and have no data on the dynamics and effects of their HIV and AIDS expenditures, finance ministries may be concerned that the funds are inefficiently utilized, regardless of the importance of the disease. It is also important to understand the role of private contributions, including out of pocket expenditures by patients or expenditures by private sector entities (insurance plans, religious groups, etc.). These amounts may be reflected in the budget in the form of fees collected or assumed in the level of funding for a particular activity. Capturing the impact of private contributions on the provision of health care is often difficult, but the fact that low-income populations can be very sensitive to out of pocket expenses may make it an important factor. Comprehensive national health accounts, with HIV and AIDS-specific sub-accounts, are the best means to monitor the level of committed expenditure. This tool is described in the following section of the paper. A limitation is that national health accounts are usually available 1-2 years after a budgetary year. 8 Glassman Levels of spending. As important as the objective of oversight on the level of spending is, there is no "right answer" to the question of what is the optimal level of public spending on health or HIV/AIDS programs. Public spending levels are determined by technical criteria (what is the level of financing required to bring the epidemic down to the lowest possible incidence and prevalence rates given the effectiveness of existing interventions and their prices?), international goals and commitments, interests of external financing agents, competing budgetary priorities, political interest groups and macroeconomic stability requirements, among others. Even from the technical point of view, however, few countries have accounted fully for existing financing and fewer still have established the optimal level of financing required to reduce the incidence and prevalence of HIV/AIDS. As a result, it is difficult to set a target range for HIV and AIDS program spending at the national level. Alternative analyses might look at whether a country is spending more or less than countries with similar characteristics, however, a problem with this approach is that it tends to focus almost exclusively on the inputs, that is, the amount of spending relative to income, and fails to consider the main goal of spending which is better health (Savedoff 2003). An assessment of the level of spending should also take macroeconomic and fiscal trends into account. Although tax revenue yields seem apparently irrelevant to HIV and AIDS programs, for instance, these revenues or lack thereof ultimately affect funding levels as even high-priority areas cannot remain immune from overall budget trends for long. In most cases, a useful short-term objective is that of convincing the executive and donors to account for the totality of public and external aid-financed HIV/AIDS projects as part of the government budget. While this practice is not as common as would be ideal, it is feasible. The Millennium Challenge Corporation, for example, requires that its projects provide regular budget reports to the Ministry of Finance even though the execution of the funds is controlled by external groups. Allocation of spending. Once the relevant activities in the relevant budgets have been identified, the next steps are likely to focus on whether the programs are designed in a way to provide the desired outcomes and whether the activities are adequately funded. One element relates to the geographic allocation of expenditure. Expenditure is considered to be pro-PLWHA (people living with HIV & AIDS) or pro-poor when it is distributed according to the distribution of need or risk. For example, the attached regional map depicts the estimated HIV seroprevalence levels in Southern Africa (http://gamapserver.who.int/mapLibrary/Files/Maps/SouthernAfrica.jpg). With country- specific epidemiological data, such as that displayed for South Africa and Swaziland, a simple analysis of the distribution of expenditure vis-à-vis the number of reported AIDS cases could allow for a gross assessment of the extent to which public spending of HIV/AIDS was indeed targeted according to need. However, it should be noted that epidemiological surveillance systems may not track well to the administrative divisions of a country, such that it may be difficult to judge how best to channel funds according to need. Further, certain types of expenditure may only occur in a location that has, for example, a hospital, which would tend to concentrate expenditure, but this characteristic in and of itself may not necessarily imply that this spending is inappropriate as patients may travel from high-risk settings to lower-risk settings with better infrastructure to provide the needed health services. How Parliamentarians Can Help Ensure Accountability 9 It is critical to note that although allocation indicators provide information that may suggest over- or under-emphasis on a given area, there is little research that would allow us to state normatively what is the appropriate level and composition of expenditure that would achieve impact on controlling the epidemic or improving reproductive health. UNAIDS has produced a listing of essential services and minimum coverage levels required to achieve an impact on the epidemic under different epidemiological scenarios (low level, concentrated or generalized epidemic) (UNAIDS 2005). For 2006, for example, an ideal allocation among HIV and AIDS program activities looked like this: 56 percent prevention, 20 percent treatment and care, 11 percent orphans and vulnerable children, 10 percent program costs and 3 percent human resources. In the future, specific country studies should be undertaken to provide input into this oversight; such an analysis should seek to determine the full cost of implementing desired programs and use these funding levels as a benchmark for assessing government budgets. Further, it is important to assess whether the allocation of spending has the potential to crowd out the prevention and treatment of other diseases and conditions. In principle, all patients suffering from illness, regardless of cause, should be equally eligible for public assistance, particularly if they are poor (Over 1998). For example, a patient suffering from terminal cancer should have the same right to public support as an HIV-infected patient. However, this is difficult to enforce in practice. Purchasing inputs. Budgets can't add value if there is little relation between the budget on paper and the way money is actually spent. As a result, oversight of input purchasing is critical, particularly in the health sector where expenditure misuse and corruption issues can be prominent (Lewis 2006; Tayler and Dickinson 2006). Human resources. From an oversight perspective, there are four main human resources issues that may impact the effectiveness of policies and programs to reduce HIV/AIDS and promote reproductive health. A first issue is simply that the lack of qualified human resources ­ attributable both to a shortage in absolute terms and to poor distribution with respect to need -- is considered the binding constraint to the scale up of the prevention and treatment of HIV/AIDS. Recent research suggests that scaling up treatment with ARV would require between 20 and 50 percent of the available health workforce in four African countries, though less than 10 percent in the other 10 countries surveyed (Smith 2004). In general, however, the need for health workers outstrips supply and has a direct impact on coverage rates. For example, the Joint Learning Initiative estimates that, on average, countries with fewer than 2.5 health care professionals per 1000 population failed to achieve an 80 percent coverage rate for deliveries by skilled birth attendants or for measles immunization (Chen et al 2004). A second area of concern is that increased financing for HIV/AIDS may result in the already short supply of provider time being disproportionately skewed towards this single disease to the detriment of other conditions. While this concern has been aired, it is clear that in high prevalence countries, this shift occurred years before additional external financing in the form of the Global Fund and PEPFAR appeared on the scene. For example, a 1996 study of Kenyatta National Hospital in Nairobi, Kenya (Floyd and Gilks 1996), found that between 1998/89 and 1992, the number of HIV-infected patients admitted daily more than doubled. The mortality rate of HIV-negative inpatients rose by 10 Glassman more than two-thirds, indicating that those with less serious conditions could not even get admitted to the hospital. The Global Fund and PEPFAR are only recently beginning to implement activities; as a result, the recent System-Wide Effects of the Fund research program based at Abt Associates was not able to document the size and nature of human capacity constraints on implementation of GF-supported activities. To date, the key human resource capacity constraint in dealing with GF resources "has occurred at the central level ... [related to]...the process of applying for, planning, and launching GF- supported programs." A third area relates to the general performance challenges in the health sector, such as absenteeism and low productivity. Absenteeism is a widespread phenomenon. A recent multi-country study based on multiple, unannounced facility visits reports absenteeism rates ranging from 23 to 40 percent in the health sector (World Bank 2004). Even when health personnel report to work, they may display notoriously low productivity as measured by the number of patients seen on a given day. These practices are thought to be associated with low pay levels, but also the lack of incentives and supervision that would enable better performance and accountability to health system users. In general, additional responsibilities associated with the implementation of scaled-up external financing for HIV/AIDS do not correspond to increases in grade or salary. An exception has been Malawi where decision makers were able to utilize GF monies to facilitate the implementation of a broader health sector program to increase all health worker salaries. Finally, clientelistic and corrupt hiring practices (buying and selling positions, bribes) have also been observed. A fourth concern relates to the absence of managerial and planning skills that are necessary for scaling up. While much training has been financed over the years, it has been concentrated on clinical issues, while the managerial and planning skills critical to successful implementation are rarely addressed. Based on its interim report on three African countries, the System-Wide Effects for the Global Fund project has recommended an assessment of human resource constraints to scaling up that could be developed into an overarching national-level strategy that would promote the implementation of proven strategies of health work retention and motivation. Procurement. Procurement is at the core of government's discretionary spending. Irregularities in the form of procedural violations, embezzlement and collusion are important to monitor and sanction publicly. An adequate legislative framework for fair and transparent procurement is an important pre-requisite to effective oversight. Timely procurement of drugs is a critical piece of any treatment program; experts attribute much of the failures observed in these systems to limited capacity to forecast demand, manage inventory and conduct procurement activities, in addition to erratic and unreliable donor financing (Levine 2006). Given the magnitude and rapid scale up of HIV/AIDS financing, it is also an area that is particularly vulnerable to misuse and corruption. In 2006, Transparency International dedicated a section of its annual corruption report to HIV and AIDS programs and the How Parliamentarians Can Help Ensure Accountability 11 Global Fund itself has pulled financing when discoveries of misuse occur.4 Some of the cases highlighted include: Kenya's National AIDS Control Council for irregularities in procurement procedures, donations that could not be accounted for and embezzlement. The Efficiency Monitoring Unit in the Office of the President documented that at least half of the money transferred to community-based organizations was lost through wanton theft or private use. Director and senior staff at the Zimbabwe National Network for People Living with HIV/AIDS were suspended after allegations of corruption. The network received more than US $1.8 million from the National AIDS Council between 2003 and 2004. Nigeria's ARV program attracted much criticism in 2003 when treatment centers began handing out expired drugs and rejecting patients, but it is not yet clear whether the prime cause was corruption or a weak drug procurement, supply and distribution system that was unable to respond to the demands that the rapid scaling-up of the program had placed upon them. A Ugandan commission has been formed to investigate alleged mismanagement of Global Fund grants and has heard assertions of inflated expenditures, inaccurate receipts and improper allocation of funds, amounting to what head of the commission James Ogoola recently called a "pile of filth" (PlusNews, 4/3). The Global Fund in August 2005 announced the suspension of five grants to Uganda worth $367 million after an audit of one of the grants found evidence of mismanagement by the Ugandan Ministry of Health's Project Management Unit, which was established to implement the grants. The government later that month appointed a four-member commission to investigate the allegations and brought in international accounting and auditing firm Ernst & Young to take over temporary management of the country's AIDS funding from PMU (Kaiser Daily HIV/AIDS Report, 9/1/05). The Global Fund in November 2005 resumed funding to Uganda after signing an agreement with the Ugandan Ministry of Finance, Planning and Economic Development that strengthened the oversight of programs receiving funds (Kaiser Daily HIV/AIDS Report, 11/11/05). This area has acquired greater visibility in recent years and has led to a profusion of efforts to monitor the quality of public procurement systems and enhance their operability. The simplest form is monitoring is the physical observation of procurement practices and outcomes. A second form of procurement monitoring focuses on transparency: the publication of procurement opportunities and outcomes. In many countries electronic procurement systems have tremendously increased the visibility of public contracting and allowed government and non-governmental bodies alike the opportunity to review the distribution of contract awards as well as the price the government pays for its goods, works, and services. A third form of procurement monitoring is assessing performance of public procurement systems using defined performance indicators. The Baseline Indicator Set for Procurement (BIS) has been developed as part of the OECD-DAC Working Party on Improving Aid Effectiveness and 4Given that programs that provide ARV must be able to assure the steady supply and administration of medicines (or face increases in drug-resistant strains requiring more costly second-line treatment), the implications of pulling funding from a program that finances ARV are important and may serve as a powerful disincentive for reporting abuses. 12 Glassman has been adopted and used in more than 10 countries in the first six months after it was finalized. The BIS to be used as a regular monitoring tool and is available at www.oecd.org/dataoecd/12/14/34336126.pdf. Audits. Audits are intended to be the primary mechanism to monitor the appropriate utilization of expenditure. However, where auditors are not independent or under funded, audits do not occur as planned. In Zimbabwe, for example, the government has imposed an "AIDS levy" since 2000 whereby employees contribute 3 percent of their gross salaries towards a fund administered by the National AIDS Council (NAC). It is estimated that the government collects about US $20 million per year through this mechanism, but no information about how the fund is used and who benefits from it has ever been made public. In March 2005, the health ministry ordered an audit of the NAC, but to date it has not yet been published. Outputs and outcomes. The monitoring and evaluation of outputs and outcomes may or may not be part of a government's budgetary process. In July 2005, the UNGASS provided technical guidelines to National AIDS Councils on output and outcome indicators to measure the performance of HIV and AIDS programs. While most of the indicators do not include targets and there are other limitations associated, the document represents the current international consensus on monitoring HIV/AIDS programs' outputs and outcomes. Most programs generally monitor standard primary outcomes, such as the total number of patients ever starting ARV, the number alive and receiving treatment, and the number who died due to the discontinuation of treatment or improper maintenance of medication. Secondary outcomes relating to functionality are less frequently monitored and include the patient's ability to walk home unaided, side effects of treatment, and drug adherence. Such data are available from HIV units within Ministries of Health or from coordinating councils. In rare cases, these data are published, as is the case in Malawi (Harries et al 2006). Parliamentary demand to link expenditures to outcomes of this type may make an important contribution to accountability and greater policy effectiveness. IV. Available expenditure tracking tools A. Domestic resource tracking tools Parliaments can make more effective use of existing provisions and mechanisms for oversight. Simple budgetary analyses can allow for a calculation of the financial performance indicators listed above. Using these figures, MPs and parliamentary committees can scrutinize how allocated funds are spent, the reasons behind under- spending allocated funds, and determine whether the total amount allocated to HIV/AIDS within the health budget is appropriate. All budgets are organized by ministries or departments, the so-called "administrative unit." While the health ministry is probably the most important for HIV and AIDS programs, it may also be necessary to look at other ministries such as Public Works for questions of infrastructure, Social Assistance or Family Welfare for services directed to orphans and vulnerable children, or statistical agencies for data collection. Within a ministry, budgets are usually broken down by bureaus or sub-groups of the agency, and How Parliamentarians Can Help Ensure Accountability 13 then by programs. Generally, it is the program-level that is most important for identifying key activities with respect to HIV/AIDS. Budgets are also organized by "functional" and "economic" classifications. Function means the expenditures are classified according to the purpose for which they are to be used. Most HIV/AIDS and reproductive health spending will be in the health function, but some relevant spending is likely to be found elsewhere. Economic classifications distinguish between funds that are used, for instance, to pay wages or to construct a clinic. An excellent resource for data and analysis of budgets are civil society organizations (CSO)5 that watchdog AIDS, budget, and governance issues. Some CSOs track how government actually spends funds, with the aim of identifying inconsistencies, leakages and bottlenecks in the flow of financial resources or other inputs. For example, the Institute for Democracy in South Africa (IDASA) monitors HIV/AIDS budget allocations and reports findings on a regular basis, as shown in Table 2 below, which is reproduced from one of IDASA's "Budget Brief" series (Ndlovu 2005). Table 2: IDASA Report on HIV/AIDS Allocations in South Africa B. Medium Term Expenditure Frameworks (MTEFs) If a country participates in the Highly Indebted Poor Country (HIPC) Initiative, a Poverty Reduction Strategy Paper (PRSP) is prepared. The PRSP describes a country's macroeconomic, structural and social policies and programs to promote growth and reduce poverty, as well as associated external financing needs. PRSPs are prepared by governments through a participatory process involving civil society and development partners, including the World Bank and the IMF. One of the key elements of a Poverty Reduction Strategy is an expenditure plan that outlines the resource envelop for budget allocations over several years. These plans have been labeled "Medium Term Expenditure Frameworks" (MTEFs). The purpose of an MTEF is to indicate the size of the financial resources needed during the medium term, usually between three to five years, in order to carry out existing policy. The MTEF concept differs from multiyear budgeting, which involves fixed appropriations for a certain number of fiscal years. Usually, only the first year of an MTEF is approved by the 5A list of CSOs that perform these functions in Africa is attached in Annex 1. 14 Glassman legislature as the annual budget, whereas the outer years are nonbinding projections of the future expenditures given existing policy. The firmer these projections become, the more they move to the centre stage of the budget process and form the basis for the annual negotiation of allocations, resulting in a system of `rolling budgets.' All OECD countries have medium term frameworks, and many developing countries are in the process of adopting them. The level of sectoral detail is MTEFs is typically limited to overall sectoral expenditures rather than to allocations within each sector. Parliamentarians concerned about HIV/AIDS and reproductive health need to pay close attention to the way in which these areas are addressed in PRSPs and to the allocation of resources for them, but will also need to look beyond the MTEF to sectoral expenditure plans and track actual spending patterns that evolve during implementation of the country's poverty reduction program. C. National Health Accounts (NHAs) and sub-accounts Implemented in 60 middle and low-income countries, NHAs are an internationally recognized framework for measuring total (public, private and donor) health expenditures in a given country. NHA methodology tracks the flow of funds through the health sector, from funding sources, through financing agents, to providers and functions. In addition to illustrating total national health care spending patterns, the NHA framework has been adapted to enable "subanalyses" that can be used to capture data on a specific disease by breaking down expenditures on related individual services and disease areas, such as HIV/AIDS interventions. In describing the flow of HIV/AIDS funds, NHA reveals not only how much is spent on HIV/AIDS but also the types of services the population uses ­ prevention, treatment of opportunistic infections, ARV treatment, etc. ­ and how those services are financed ­ by government funds, donor grants, or out-of-pocket payments by households. By disaggregating HIV/AIDS expenditures in this manner, the NHA subanalysis provides legislators, policymakers, program managers and donors with a clear picture of the magnitude and composition of spending. Further analysis has, in some countries, produced information on HIV/AIDS expenditures by socioeconomic status, geographic grouping, and/or gender, allowing for an assessment of the financial burden of the disease on different subpopulations. Note, however, that it is critical to complete a comprehensive NHA before embarking on sub- accounts. Most of the indicators described in Table 1 are sourced in a NHA exercise. The subaccounts have illustrated that the magnitude of public spending for HIV/AIDS is still quite low and while donor-financed expenditures are increasing and out-of-pocket household payments decreasing (good news for poverty reduction objectives), they fall far short of the funds required to meet HIV/AIDS prevention and control goals. HIV/AIDS subaccounts are available for Burkina Faso, Ghana, Kenya, Rwanda and Zambia. Even if HIV/AIDS subaccounts are unavailable, budget analysis may be done to illustrate the issues. Reproductive health subaccounts also exist in some countries. To allocate resources to the most cost-effective interventions for combating the HIV/AIDS pandemic, parliamentarians need to know what works and how much is being spent on specific interventions. Financial indicators are conspicuously absent from many How Parliamentarians Can Help Ensure Accountability 15 M&E frameworks. Each global initiative has developed its own monitoring and evaluation framework for reporting by grant recipients, yet few have added financial indicators to track the very investment results called for within initiative goals. Additionally, national AIDS strategies often set their own targets and countries are encouraged to report on core indicators for tracking progress towards targets set in the UNGASS Declaration of Commitment on HIV/AIDS-neither highlights the importance of tracking expenditures against achievements. Multiple monitoring and evaluation frameworks persist despite agreement, between various HIV/AIDS stakeholders, on "the three ones" principle: that each country should have one HIV/AIDS budget, one national coordinating committee and one M&E system (Cohen 2006). Estimating the gaps between resources promised, committed, and expended has been effective in advocating for increased mobilization of resources for HIV/AIDS, however, it is now necessary to demonstrate that the money is being used wisely. Current initiatives to objectively track resources focused on HIV/AIDS at the country level are supported by UNAIDS and bilateral donors such as USAID. As monitoring and evaluation systems continue to evolve toward a common framework, it is critical that financial indicators be included, tracked and reported. To learn more about efforts in tracking resources, the UNAIDS (www.unaids.org) and IDASA (www.idasa.org) web sites are useful, as is the Partners in Health Reform plus project at www.phrplus.org. D. Public Expenditure Reviews (PER) and Tracking Surveys (PETS) Public expenditure reviews are periodic World Bank staff reports that study a government's objectives and assess whether the government is using public financing to achieve their objectives efficiently, equitably, and sustainably. Key questions to be answered might include: (i) Is government's policy and implementation sound?; (ii) Are key public goods and services being adequately provided?; (iii) Are programs intended to reach the poor doing so effectively?; (iv) Are some programs merely crowding out private activities?; and (v) What are the institutional weaknesses of the budget system that need attention? The PER are also intended to support government capacity to translate PRSP commitments into real budget implementation, to advise government on options for reform of policy and management, among others. While few PER focus specifically on HIV/AIDS, many analyze public spending on health and its adequacy, equity, efficiency, contribution to risk protection (protected households from impoverishing expenditure on health), delivery mechanisms, sustainability and impact. Some PERs provide break outs of program-specific financing; the case of Vietnam is provided in a World Bank document that provides guidelines for the preparation of PER in the health sector. PETS are tracking tool s that aim to answer the question "Does public money spent on health actually reach frontline health facilities?" The studies track the flow of public resources through the different layers of the administrative hierarchy to individual service providers and develop quantitative estimates of fiscal leakage, that is, the failure of resources intended for frontline service providers to reach their intended destination. The WB and other organizations have conducted PETS in over two dozen countries. A recent review by Lindelow et al (2006) documents results in the health sector; Table 3 16 Glassman excerpts PETS findings from Africa. As can be seen, local capture, leakage and bureaucratic impediments result in 80 percent of non-salary expenditure never reaching frontline clinics in Ghana, 70 percent in Uganda and 40 percent in Tanzania. These governance challenges raise doubts about the value of simply increasing funding, or of the need for "just" more health workers. Table 3: Review of PETS Results for Africa in the 2000s (Lindelow et al 2006) Country Year Sample Leakage Other findings Ghana 2000 200 facilities; 40 Leakage of non-salary Greater leakage between districts recurrent expenditures center and district than estimated at 80% between district and facility. Mozambique 2002 90 facilities; 167 staff; Some evidence of Documented delays and 679 users leakage of drugs in bottlenecks in budget transfer from provinces execution and supply to districts, within the management; primary health care incomplete registering system of user fee revenues by facilities; absenteeism estimated at 19% Nigeria 2002 252 facilities; 30 local No firm estimate of 42% of staff experience govt.; 700 staff leakage. salary delays despite sufficient budget Rwanda 2000 351 facilities;40 districts Some evidence of Evidence of delays in leakage between regions budget execution and and districts. low execution rates (80% of non-wage funds released at year end) Senegal 2002 100 facilities; 10 Some evidence of Delays in fund transfers. districts; 37 local govts. leakage at regional and communal level in allocation of non-salary resources. Tanzania 2001 20 facilities; 5 districts Leakage estimated at Substantial delays at all 40% levels, especially non- wage expenditures Uganda 2000 155 facilities Leakage of specific drugs and supplies estimated at 70% in government and private non-profit facilities. Reinikka and Svensson (2003) have reviewed applications of PETS in the health and education sectors in ways that can help parliamentarians focus accountability interventions on specific trouble spots. In Uganda, for example, they found that in 1995, only 22 percent of the money allocated to schools by the central government actually reached those schools. The PETS process showed that most of the leakages were in non- wage allocations and that they occurred most frequently at the local government level ­ information that was used to target more focused and efficient accountability How Parliamentarians Can Help Ensure Accountability 17 interventions. As a result, when the amounts of central allocations were made public in local newspapers and monitored by parent-teacher organizations, the leakage rates dropped dramatically. E. Censuses and surveys In addition to these frequently available sources, censuses and household surveys provide key information for the indicators described in Table 1. A census provides denominators for both epidemiological and financial performance indicators. A simple measure such as HIV/AIDS spending as a percentage of GDP per capita is meaningless without somewhat accurate data on total population size and conclusions would vary enormously based on these numbers. Household sample surveys, such as the LSMS and the DHS, allow for a more nuanced analysis of expenditure data and provide the information necessary to analyze expenditure according to different socioeconomic characteristics of a household. The DHS infrequently includes expenditure modules but has recently included a seroprevalence survey in some countries. This information should be a critical input into policy dialogue and formulation. The Rwanda 2005 seroprevalence survey, for example, found levels among the productive age population that were substantially lower than those forecasted earlier in the decade, resulting in ­perhaps ­ an over dimensioning of the level of financing allocated to HIV/AIDS control efforts. The level of representativity of these sample surveys should be crucial for lawmakers in parliament; if surveys are representative at the level of representation of each lawmaker, for example at the state or province level, more evidence-based budgeting and oversight of HIV/AIDS programs can be conducted. Unfortunately, fiscal constraints have decreased the regularity of censuses and surveys in many countries, leading to a near total absence of the type of information required to generate basic budgetary performance indicators. F. Donor accountability requirements Another source of information regarding financial flows for HIV/AIDS is those related to donor accountability requirements. Even where a recipient country government does not have a comprehensive picture of expenditure, it is possible to piece together a picture from the reporting requirements of donors. For example, the Global Fund publishes a listing of all awarded grants and grantees along with disbursement progress to date; the GF also estimates the number of persons receiving ART under each grant based on grantee reporting. With a few exceptions, donor accountability has not yet evolved beyond expenditure use information and narrative evaluation of programs. V. What can parliaments do? The need for oversight is clear. Making oversight effective, however, may take time. On the one hand, budgetary systems are weak, lack the capacity to provide real time information to legislators and are insufficiently disaggregated in their standard formats to 18 Glassman allow for the estimation of the indicators suggested in this module. On the other hand, the heavy dependence on external aid for HIV and AIDS program financing in sub- Saharan Africa and the extent to which this expenditure occurs off-budget implies that even if budgetary systems worked well, parliaments would only oversee a fragment of overall spending on HIV/AIDS. In the short-term therefore, Parliaments may work effectively in three main areas: (i) improving the production and availability of data; (ii) increasing transparency at every stage of the process; and (ii) reducing misuse and corruption. Production and availability of data. The best way to incentivize the production and availability of data is to request the data from the executive branch and utilize it during legislative sessions and in budget negotiations. Draw on the sources mentioned in this report ­ NHA, PER, PETS, MTEF, donor accountability requirements ­ to pose questions and propose improvements. For example, MPs can lobby for budgets that include a specific line item for HIV/AIDS in the health budget, which will assist in the oversight task. Further, MPs can draw on civil society organizations to carry out key analyses ­ analyzing bills, motions and draft laws with an HIV/AIDS lens -- in order to make relevant amendments. Increasing transparency. Parliament should work with the executive to foment the production and publication of the financial and other data that is necessary to effectively conduct oversight. For example, governments and health authorities should publish regularly updated information on the Internet on health budgets and performance at the national, local and health delivery center levels. This can be required through a so-called "sunshine" law ­ fiscal transparency and accountability legislation. Further, government departments, hospitals, health insurance entities and other agencies handling health funds must be subject to independent audits. Governments and health authorities have the responsibility to ensure that information about tender processes, including offers to tender, terms and conditions, the evaluation process and final decisions, is publicly available on the Internet. Donors and international agencies must also be transparent about what they are giving, when and to whom, and should evaluate their programs in terms of health outcomes and not level or speed of disbursement. To review this information, Parliaments in high prevalence countries should strive to have an HIV/AIDS standing committee. Misuse and corruption. In addition to increasing transparency, independent audits and civil society watch dogging are the best sources for detecting misuse and corruption in the health sector, as the Global Fund examples indicate. Where private participation in service delivery exists, as it does in most countries, it is critical that legislation allow for private citizens to file lawsuits charging fraud in government programs. These How Parliamentarians Can Help Ensure Accountability 19 "whistleblowers" can receive financial incentives and compensation for settled cases. Some of the tools mentioned in this module, such as PETS, are also useful to objectively document misuse of funds. In addition to these initiatives, Parliamentarians may wish to join the Coalition of African Parliamentarians on HIV and AIDS (CAPAH), an independent network of parliaments dedicated to work together on HIV and AIDS efforts, in collaboration with existing institutions and regional bodies such as SADC-PF and ECOWAS, to improve the effectiveness of oversight. Other support is available through a donor-finance project Parliamentarians for Women's Health which assigns expert interns to MP offices to support development and monitoring of legislation and finances surveys and polling in constituent districts on the issues of HIV/AIDS and women's health so that legislators can better reflect the needs and realities of their constituents in budget debates. VI. Case Example: Rwanda Rwanda is a small, landlocked country with a population of approximately 8.2 million persons and per capita GDP of $230. Its population density is among the highest in the world. Rwanda has made substantial progress in stabilizing and rehabilitating its economy to pre-1994 levels. Leaving aside countries in Africa that received oil windfalls, Rwanda's economy has experienced relatively high levels of growth in the past decade. Total poverty has declined in the post-genocide period to 64 percent; however, inequality is increasing, particularly differentials between urban and rural areas where approximately 85 percent of the country lives. Although Rwanda has made a remarkable transition from reconstruction to development in the eleven years since the genocide, the legacy remains and the country is likely to depend on significant levels of external assistance for years to come, in spite of agreements on debt forgiveness. Health status. Child and maternal mortality have fallen since 1994; however, they have yet to return to pre-genocide levels that were already among the highest in sub-Saharan Africa. The major causes of mortality for children under five are largely avoidable. Reproductive health is a critical issue in Rwanda. The country has one of the highest recorded rates of maternal mortality in the world, as well as one of the lowest rates of contraceptive prevalence and a relatively high number of births per woman of reproductive age (UNDP 2004). While performance has improved over time, in 2002, only 15 percent of all births were attended by a skilled birth attendant. With adult HIV seroprevalence of 4 percent6, the AIDS epidemic represents a challenge to Rwanda's health system in particular and development prospects in general. Poverty, war, and the genocide have resulted in vulnerable populations of orphans, child-headed 6Government of Rwanda. Demographic and Health Survey 2002. Results reported on Ministry of Health web site. 20 Glassman households, victims of rape, and widows, on whom the negative impact of AIDS is particularly severe. The Government has committed to stabilizing the spread of HIV during the period 2002 to 2006. Donors have joined this commitment and, in 2003, Rwanda became a recipient of Global Fund and PEPFAR monies. For the 2006-2008 period, Rwanda is authorized to receive up to $100 million USD. However, these funds are still in the process of being transferred to executors on the ground. PRSP and HIV/AIDS. Rwanda is a HIPC country with a Poverty Reduction Strategy Paper in place since 2002. This document articulates a medium-term program (2002- 2007), coordinated with a MTEF. Government domestic spending has been increasing since 2002. Given increased availability of external financing, implementation will begin in earnest on a full range of programs, including new constitutionally mandated institutions, free basic education, government counterpart contributions for a comprehensive HIV/AIDS treatment program, supplementary pay for health sector workers, human resource development in tertiary education, labor intensive rural works, and export promotion, all of which are drawn from the PRSP. The 2004 budget also reflects 0.5 percentage points of GDP in demobilization outlays that have been carried over from 2003. As a result, government priority spending is budgeted to rise to 45 percent of domestic expenditure. A 2005 PRSP performance review published by government points to significant achievements in human development, particularly in the health sector. Although the leading development partner is the World Bank, who has provided the majority of technical assistance to the PRSP process, and the Bank finances a large HIV/AIDS reduction project (the MAP project), there are few PRSP performance indicators related to HIV/AIDS (last of 13 action areas in the health sector identified as a priority by government) and these indicators are related to communication campaigns and condom distribution (no quantitative targets included in policy matrix) and establishment of a number of HIV testing and counseling sites. This focus is at odds with the fact that in 2002, 30 percent of total health expenditure on health goes towards HIV/AIDS and the arrival of GF and PEPFAR in 2003 have increased this share significantly. In 2005, the World Bank estimated that commitments for AIDS financing in Rwanda are 115 percent of total government health spending. This may reflect a disconnect between PRSP- governed priority setting and actual budgetary and expenditure allocations. General health financing. Rwanda is one of the few countries in Africa with a National Health Accounts series that allows for an analysis of the sources and uses of health expenditure. A 2002 round indicates that about 50 percent of total financial resources in the sector are from international cooperation, 10 percent from government (primarily civil service salaries) and 40 percent from private sources (33% is direct out of pocket expenditure by households). External financing is concentrated in disease-specific interventions (Table 4). The contribution of the state to the functioning of the health sector is limited though increasing over time, about 8 percent of the national budget, equivalent to 2.50 USD per capita annually. In spite of increases, the sustainability of How Parliamentarians Can Help Ensure Accountability 21 external finance flows in health is a major issue. Compared to other countries in East and Southern Africa, Rwanda is well below the average regional measure for overall health expenditure as a percentage of GDP but above the regional average for share of public spending of total health expenditure. In terms of use of the budget, a PETS in 2000 found that 80 percent of non-salary expenditure in the health sector was disbursed at the end of the fiscal year, indicating significant challenges in budgetary execution and/or liquidity problems. It should also be noted that Government has an excellent track record on budget execution ­ at or above 85 percent of total allocations between 2002 and 2005 -- indicating little problem with absorptive capacity up to 2005 (Scorraille et al 2006). A follow up PETS in 2003 confirmed earlier findings of substantial delays in budgetary transfers from the central administration to district health offices and identified troubling issues related to public expenditure management by government, such as poor bookkeeping and lack of internal financial controls and auditing requirements (Fofack et al 2003). Table 4 Indicators for Rwanda Case Example Indicators 2000 2002 HIV seroprevalence rates (adults) 5.1% 5.1% Number of PLWHA 200,000 199,279 Total Health Expenditure (THE) US$77,992,817 US$70,101,480 THE for HIV/AIDS US$6,009,287 US$10,313,032 % of THE allocated to HIV/AIDS 8% 15% General out-of-pocket spending per inhabitant $2.51 $2.13 HIV/AIDS out-of-pocket spending per PLWHA $11.27 $7.59 Total HIV/AIDS spending as % GDP (in current prices) 0.3% 1% Expenditure on HIV/AIDS by financing source: - Public 8% 9% - Private 43% 7% - Donors 49% 75% Expenditure on HIV/AIDS by provider type: - Public providers 33% 16% - Private providers 9% 3% - Government-assisted not-for-profit providers 5% 3% - Private pharmacies 7% 3% - Provision and administration of public health 46% 66% - Administration 0% 9% Expenditure on HIV/AIDS by function: - Preventive and public health programs 46% 66% - Curative care 48% 23% - Administration 0% 9% - Pharmaceuticals purchased at independent 7% 3% pharmacies Source: Republic of Rwanda, Ministry of Health, 2005. HIV/AIDS financing. According to the HIV/AIDS accounts for 2002, donors are the major financing source of HIV/AIDS health care. More than one-third of donor financing is HIV/AIDS related and 75 percent of HIV/AIDS health care is financed through these external sources. Households contribute 16 percent, while government 22 Glassman represents 9 percent of the total. Seventy-six percent of donor spending on HIV/AIDS is channeled through NGOs and is off-budget, and this share has increased since 2000. In 2002, financing was concentrated overwhelmingly in prevention activities; the arrival of GF and PEPFAR in 2003 will likely have modified this scenario. Key attributes of HIV/AIDS spending prior to 2003 indicate a declining level of spending on HIV/AIDS, a significant financial burden on PLWHA (represented by out of pocket spending ­ OOP ­ above) and the paucity of donor funds. NHA analysts attribute much of the aid increases approved in 2005/6 ­ a 100% increase in real terms -- to the availability of NHA expenditure data documenting the problems faced. Health policy. Prior to recent reforms, central government budgetary allocations covered only the salaries of health workers and, as a result, public health facilities relied on household contributions and fees, along with sporadic contributions from donors and NGOs. A household survey in 2003 found that 95 percent of respondents who needed to see a health provider but did not do so cited the high costs of care, and among those that consulted a health provider, 80 percent were dissatisfied with the costs (Fofack et al 2003). In response to these challenges, a new Health Sector Policy was adopted by Cabinet in early 2005. While the long-term financial sustainability of the system is a main challenge, the issue of health financing is not discussed systemically. Financing is treated under a number of different objectives: (i) under human resources, the introduction of incentive structures for qualification, specialization and geographic redistribution; (ii) under drugs, the implementation of a system of cost recovery at health centers according to which drugs are bought at district pharmacies and resold to patients at the lowest possible price; and (iii) under financial accessibility to health services, preferentially for the poor, the increase of public share of financing, the promotion of mutuelles (systems of health care pre-payment) and health insurance, the definition of a pricing policy for services and drugs and the subsidy of essential services to vulnerable groups (not defined). Further, a Mutual Health Insurance Policy was approved in December 2004 which specifies that the mutuelles would be the vehicle to provide health services and subsidies for these services to the poor.7 In 2005, the government estimated that 25 percent of the population is now enrolled in mutuelles; current estimates are higher due to a recently implemented requirement to enroll in a mutuelle prior to government issuance of identity documents and marriage licenses. Rwanda has negotiated a special scheme with the Global Fund by which Fund monies will finance mutuelle enrollment for the extreme poor, orphans, vulnerable groups and PLWHA. Rwanda has also been the site of several pilots to improve the productivity, quality and responsiveness of health providers through monetary incentive payments (Soeters et al 2006). These pilots appear to have improved performance; however, the organizational 7However, mutuelles evaluations and the mutuelle policy itself recognize that the requirement to pre-pay out of pocket represents an important obstacle to the affiliation of the poor and that, without government intervention to subsidize the poor, will tend to cover the relatively better off segments of the population. How Parliamentarians Can Help Ensure Accountability 23 structure that accompanies this scheme has not yet been translated into the mutuelle structure. As a result, the future of these mechanisms is still unclear and the institutional structures required to scale up understudied. The poverty or risk targeting of subsidies has not been the subject of an explicit government policy. The World Bank MAP project targets the poorest geographical areas. A recent government initiative to provide free delivery care is a mechanism to target user fee waivers; user fee waivers are also reportedly applied to female headed households which are over-represented among the poor. Political economy. The government plans that donor financing for health be structured predominantly through a Sector Wide Approach with the goal of enhancing the management of external resources and expanding budgetary support so as to channel resources more directly towards the development of the health system. The share of donor financing going to budget support is not publicly available. In 2003, a detailed mapping of donor activities and resources was conducted to highlight gaps and duplication of efforts; this led to the formation of a number of consultative and coordinating bodies, including a health cluster. The cluster meets frequently under the leadership of the Secretary General of the Ministry of Health and Belgian Technical Cooperation is the secretariat. Six health cluster working groups were also formed on the issues of performance-based contracting, mutuelles, HIV/AIDS integration into health system, human resources, disease control and intervention mapping; these groups are providing technical advice and position papers to government. A review of PRSP progress in 2005 by World Bank and IMF staffs noted that the strategy preparation process encountered bottlenecks relating to poor coordination within the sector. These were partially overcome through collective workshops and strong support to planning from budget support donors. However, the report notes that there continues to be a disconnect between health system development planning and the activities of some unnamed "vertical" programs within the sector which surfaced during the strategic planning process. Other sources report anecdotal reports of AIDS financing "crowding out" other types of health services and "distorting" priority-setting within government; however, this phenomena has yet to be systematically analyzed.8 8A detailed study on this topic will be undertaken by the World Bank in 2006. It is an urgent requirement as the US Government-financed PEPFAR will begin disbursements in Rwanda in 2006 as well, resulting in a substantial increase of available funds for HIV and AIDS programs -- $100 million USD for the 2006- 2008 period. 24 Glassman References Barkan J, Ladipo Adamolekun and Yongmei Zhou, "Emerging Legislatures: Institutions of Horizontal Accountability" in Brian Levy and Sahr Kpundeh (eds.) Building State Capacity in Africa. Washington, DC: World Bank Institute, 2004. Chen L, Evans T, Anand S, Boufford JI, Brown H, Chowdhury M et al. Human resources for health: overcoming the crisis. The Lancet 2004, 364: 1984-1990. Cohen J. The New World of Global Health. Science 2006, 311:162-167. Dmytraczenko T, De S, Chanfreau C, Kidane L. The Value of Beneficiary Analyses: Who Benefits from Funds Targeted For HIV/AIDS? Bethesda, MD: Abt Associates and PHRplus, 2006. www.phrplus.com Floyd K, Gilks CF. Impact of and response to the HIV epidemic at Kenyatta National Hospital, Nairobi. Report 1 (April) Liverpool, England: Liverpool School of Medicine, 1996. Fofack, Hippolyte, Chukwuma Obidegwu, Robert Ngong. Public expenditure performance in Rwanda: evidence from a Public Expenditure Tracking Study in the health and education sectors. Africa Region Working Paper Series No. 45. Washington, DC: World Bank, 2003. www.worldbank.org/afr/wps/index.htm Foster, Mick. MDG Oriented Poverty Reduction Strategies: Lessons from Experience in Health. Washington, DC: World Bank, 2005. http://www.mickfoster.com/docs/FosterMDGStrategiesFinal.pdf Government of Rwanda, Ministry of Health. Rwanda National Health Accounts 2004. Available at www.phrplus.com Harries, Anthony D, Erik J Schouten, and Edwin Libamba. Scaling up antiretroviral treatment in resource-poor settings. The Lancet 2006; 367; 1870-1872. www.thelancet.com/journals/lancet/article/PIIS0140673606688090/fulltext Hecht R, Adeyi O, Semini I. Making AIDS part of the global development agenda. Finance & Development, Washington, DC: International Monetary Fund, 2002. Guthrie T, Hickey A (Eds.). Funding the Fight: Budgeting For HIV/AIDS in Developing Countries. Cape Town, South Africa: IDASA, 2004. Keith-Brown K. Investing for life: making the link between public spending and the reduction of maternal mortality. Mexico D.F.: Fundar, Centro de Análisis e Investigación, 2005. http://www.internationalbudget.org/ How Parliamentarians Can Help Ensure Accountability 25 Lewis M (a). A war chest for fighting HIV/AIDS. Finance & Development. Washington, DC: International Monetary Fund, December 2005. Lewis M (b). Tackling healthcare corruption and governance woes in developing countries. CGD Brief. Washington, DC: Center for Global Development, 2006. www.cgdev.org Levine, Ruth. New HIV/AIDS Buffer Stock Service for Africa. Global health policy blog, October 11, 2006. www.cgdev.org Lindelow, Magnus, Inna Kushnarova and Kai Kaiser. Measuring corruption in the health sector: what we can learn from public expenditure tracking and service delivery surveys in developing countries. Pages 29-36, Transparency International, Global Corruption Report 2006, Berlin: Transparency International, 2006. http://www.transparency.org/publications/gcr National Democratic Institute for International Affairs (NDI) and the Southern African Development Community Parliamentary Forum (SADCPF). Survey of legislative efforts to combat HIV/AIDS in the Southern Africa Development Community region. November 2004. http://www.ndi.org/ndi/library/1795_sa_hiv_110104.txt Ndlovu, Nhlanhla. HIV and AIDS Allocations: A First Look at Budget 2005. IDASA, Budget Brief No. 152, 2005. http://www.internationalbudget.org/resources/howto/HIV.pdf Over, Mead and Martha Ainsworth. Confronting AIDS: Public Priorities in a Global Epidemic. Washington, DC: Oxford University Press, 1997. www.worldbank.org/aids- econ/confront/ Over, Mead. Coping with the impact of AIDS. Finance & Development. Washington, DC: International Monetary Fund, March 1998. www.imf.org/external/pubs/ft/fandd/1998/03/pdf/over.pdf Over, Mead, Peter Heywood, Julian Gold, Indrani Gupta, Subhash Hira, and Elliot Marseille. HIV/AIDS treatment and prevention in India: modeling the cost and consequences. Washington, DC: World Bank, 2004. www.worldbank.org/servlet/WDS_IBank_Servlet?pcont=details&eid=000090341_20040 921112225 PHRPlus. Using the National Health Accounts Subanalysis to Track Resource Flows for HIV/AIDS. NHA Global Policy Brief. Bethesda, MD: Abt Associates, June 2005. http://www.phrplus.org/Pubs/sp16.pdf 26 Glassman PHRPlus Project. AIDSTREATCOST (ATC) model software to help determine costs of ARV programs, and user manual, developed by PHRplus. PHRPlus project. System-wide effects of the Global Fund. http://www.phrplus.org/swef.php Reinikka, Ritva and Jakob Svensson. Survey techniques to measure and explain corruption. Washington, DC: World Bank, 2006 http://web.worldbank.org/WBSITE/EXTERNAL/TOPICS/EXTPUBLICSECTORANDG OVERNANCE/EXTPUBLICFINANCE/0,,contentMDK:20235447~pagePK:148956~piP K:216618~theSitePK:1339564,00.html Savedoff, William. How much should countries spend on health? FER/EIP Discussion Paper Number 2. Geneva, Switzerland: World Health Organization, 2003. http://www.who.int/health_financing/en/how_much_should_dp_03_2.pdf Schneider P, Bhatt P. Linking Indicators from National Health Accounts and the NHA HIV/AIDS Subanalysis to Health Policy Goals. Bethesda, MD: PHRplus 2004 Scorraille, Guy de, Jacquie Nachtigal, Jean Munyenpenda. Revue des Depenses Publiques dans le Secteur de la Sante 2002-2005. Republique de Rwanda, Ministere de la Sante, March 2006. Smith O. Human resource requirements for scaling up antiretroviral therapy in low- resource countries (Appendix E). In: Curran J, Debas H, Arya M, Kelley P, Knobler S, Pray L, eds. Scaling up treatment for the global AIDS pandemic: challenges and opportunities. Washington, DC, National Academies Press (Board of Global health), 2004. Soeters, Robert, Jean Perrot, Etienne Sekaganda, Antonio Lozito, and Laurent Musango. Improving health services delivery through performance-based financing in Rwanda: what are the factors associated with success? Submitted to Health Policy and Planning, 2006. Stapenhurst, F. C. and R. Pelizzo (2002). `A Bigger Role for Legislatures.' Finance and Development 39(4): 46-48. http://www.imf.org/external/pubs/ft/fandd/2002/12/stapenhu.htm Tayler Liz and Clare Dickinson (2006). "The link between corruption and HIV/AIDS" in Transparency International, Global Corruption Report 2006, Berlin: Transparency International, 104-111. http://www.transparency.org/publications/gcr UNAIDS. Resource needs for an expanded response to AIDS in low and middle-income countries. UNAIDS, August 2005. How Parliamentarians Can Help Ensure Accountability 27 http://data.unaids.org/publications/irc-pub06/resourceneedsreport_en.pdf UNAIDS. UNGASS indicators. http://data.unaids.org/publications/irc-pub06/jc1126- constrcoreindic-ungass_en.pdf UNAIDS. National Spending For AIDS 2004. Geneva, Switzerland: UNAIDS 2004. Wehner, Joachim. Effective financial scrutiny: the role of parliament in public finance. Washington, DC: The World Bank, 2004. World Bank (1998). Public Expenditure Management Handbook. Washington, D.C., World Bank. http://www1.worldbank.org/publicsector/pe/english.htm World Bank. World Development Report: Making Services Work for Poor People. Washington, DC: Oxford University Press, 2004. http://publications.worldbank.org/ecommerce/catalog/product?item_id=1764996 World Bank. Global Monitoring Report 2006: Strengthening Mutual Accountability ­ Aid, Trade and Governance. Washington, DC: Oxford University Press and the World Bank, 2006. http://web.worldbank.org/WBSITE/EXTERNAL/EXTDEC/EXTGLOBALMONITOR/E XTGLOBALMONITOR2006/0,,menuPK:2186472~pagePK:64218926~piPK:64218953 ~theSitePK:2186432,00.html 28 Glassman Annex: Civil Society Organizations that Watchdog Budgets in Africa (source: http://www.internationalbudget.org/groups/) Cameroon: www.eitdr.org www.cameroonbudgets.org www.e-citizens.net www.vs-aprm.net Ghana: http://www.isodec.org.gh Malawi: Malawi Economic Justice Network mejn@sdnp.org.mw Civil Society Coalition for Quality Basic Education (CSCQBE) cscqbe@sdnp.org.mw Namibia: Institute for Public Policy Research www.ippr.org.na Namibian NGO Forum (NANGOF) nangof@iafrica.com.na Namibia Chamber of Commerce and Industries (NCCI) http://goss.rho.net/ncci/indexp.htm Nigeria: Integrity www.theconvention.org Social & Economic Rights Action Center (SERAC) serac@linkserve.com.ng; serac@hyperia.com; seracnig@aol.com South Africa Institute for Democracy in South Africa (IDASA)'s Budget Information Service (BIS) shun@idasact.org.za How Parliamentarians Can Help Ensure Accountability 29 SA National NGO Coalition (SANGOCO) http://www.sangoco.org.za Public Service Accountability Monitor (PSAM) http://www.psam.org.za http://www.myrights.org.za Tanzania Hakikazi Catalyst www.hakikazi.org Tanzania Gender Networking Programme (TGNP) http://www.tgnp.org Action Aid admin@actionaid.org How Parliamentarians Can Help Ensure Accountability for Spending on HIV/AIDS and Reproductive Health Amanda Glassman World Bank Institute Washington, D.C. Copyright © 2007 The International Bank for Reconstruction and Development /The World Bank 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. The World Bank enjoys copyright under protocol 2 of the Universal Copyright Convention. This material may nonetheless be copied for research, educational or scholarly purposes only in the member countries of The World Bank. This paper has not undergone the review accorded to official World Bank publications. The findings, interpretations, and conclusions expressed herein are those of the author(s) and do not necessarily reflect the views of the International Bank for Reconstruction and Development / The World Bank and its affiliated organizations, or those of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. How Parliamentarians Can Help Ensure Accountability for Spending on HIV/AIDS and Reproductive Health Amanda Glassman 2007. 35 pages. Stock No. 37270 Contents Acknowledgements...................................................................................................................................... iv Acronyms ...................................................................................................................................................... v Executive Summary..................................................................................................................................... vi I. Introduction...................................................................................................................... 1 II. Budgetary process............................................................................................................ 2 III. Public policy for HIV/AIDS prevention and treatment................................................ 4 Table 1: Potential indicators for parliamentary oversight of HIV/AIDS programs................................... 6 IV. Available expenditure tracking tools............................................................................ 12 A. Domestic resource tracking tools.................................................................................. 12 Table 2: IDASA Report on HIV/AIDS Allocations in South Africa .......................................................... 13 B. Medium Term Expenditure Frameworks (MTEFs)................................................... 13 C. National Health Accounts (NHAs) and sub-accounts................................................. 14 D. Public Expenditure Reviews (PER) and Tracking Surveys (PETS).......................... 15 Table 3: Review of PETS Results for Africa in the 2000s (Lindelow et al 2006) ..................................... 16 E. Censuses and surveys..................................................................................................... 17 F. Donor accountability requirements.............................................................................. 17 V. What can parliaments do? ............................................................................................ 17 VI. Case Example: Rwanda................................................................................................ 19 Table 4 Indicators for Rwanda Case Example......................................................................................... 21 References ................................................................................................................................................... 24 Annex: Civil Society Organizations that Watchdog Budgets in Africa ................................................. 28 iii Acknowledgements* This paper was prepared by Amanda Glassman (consultant) under the general direction of Tom Merrick (WBIHD) and Isabel Rocha Pimenta (WBIHD). Valuable guidance and comments were provided by Frederick Stapenhurst (WBIPR), Rene Bonnel (HDNGA), Festus Wachira Kingori (WBIHD-Secondee from Kenya Ministry of Finance), John Armiger (WBIHD)and Tony Worthington (former MP, UK, and member of the board of Parliamentary Network on the World Bank). * Views expressed in this paper are those of the author only. iv Acronyms AIDS Acquired Immune Deficiency Syndrome ART Anti-Retroviral Therapy BIS Baseline Indicator Set for Procurement CSO Civil Society Organization DHS Demographic and Health Survey ECOWAS Economic Community of West African States GDP Gross Domestic Product GF Global Fund for AIDS, TB and Malaria HIPC Highly Indebted Poor Country Initiative HIV Human Immunodeficiency Virus IMF International Monetary Fund LSMS Living Standards and Measurement Survey MAP World Bank Multi-Country HIV/AIDS Program MP Member of Parliament MTEF Medium-Term Expenditure Framework NHA National Health Accounts OECD/DAC Organization for Economic Cooperation and Development/Development Assistance Committee OOP Out of Pocket Spending (Household Spending) PEPFAR President's Emergency Plan for AIDS Relief PER Public Expenditure Review PETS Public Expenditure Tracking Survey PLWHA People Living With HIV and AIDS PRSP Poverty Reduction Strategy Paper SES Socio-Economic Status THE Total Health Expenditure UNAIDS Joint United Nations Program on HIV/AIDS UNGASS United Nations General Assembly Special Session on HIV/AIDS USAID United States Agency for International Development v Executive Summary HIV/AIDS is an important problem, destroying the lives and livelihoods of millions of people. UNAIDS estimates that 40 million people are living with HIV; more than half are in sub-Saharan Africa. In response to this crisis, governments are increasing spending on HIV and AIDS policies and programs, primarily in response to a massive influx of external financing. Increased funding can skew priority-setting inside and outside the health sector and put enormous pressure on countries' capacity to effectively use the funds. Increased funding also creates a context where the risk of misuse of funds can be high, particularly as current program performance is essentially measured by the magnitude and speed of disbursement. Bribes, corrupt officials and inappropriate procurement undermine health care delivery in much the same way they do for police services, law courts and customs whose functions become compromised by the culture of poor governance and corruption. Parliaments are increasingly asserting their independence from the executive branch and actively performing their constitutionally-designated functions of overseeing the executive, representing citizens, making policies, and enacting laws. Legislative oversight is nowhere more important than over the budget. Oversight of budgets and expenditures by parliament can potentially improve effectiveness and reduce risks of misuse and inefficiency. In the case of HIV/AIDS, parliaments can ensure that HIV/AIDS funds flow through the budget and that the amount, distribution and utilization of funding are consistent with national policies and priorities as well as international best practices. This note provides an overview of the budgetary process with specific reference to allocations for HIV/AIDS prevention and treatment and highlights potential trouble-spots where the exercise of Parliamentary oversight could improve financial performance as well as improve outcomes. Parliamentarians can make more effective use of existing tool for oversight. Using these tools, MPs and parliamentary committees can scrutinize how allocated funds are spent, the reasons behind under-spending allocated funds, and determine whether the total amount allocated to HIV/AIDS within the health budget is appropriate. Specific Parliamentary actions are discussed, and a country case example (Rwanda) is presented as an illustration of how oversight tools can be applied. vi I. Introduction The problem. HIV/AIDS is an important problem, destroying the lives and livelihoods of millions of people. UNAIDS estimates that 40 million people are living with HIV; more than half are in sub-Saharan Africa. The epidemic devastates families through the loss of adults in their productive ages, particularly women, while foregone human capital formation affects economic growth. In response to this crisis, governments are increasing spending on HIV and AIDS policies and programs, primarily in response to a massive influx of external financing. The IMF reports that HIV/AIDS resources flows were US $5 billion in 2003 and US $8 billion in 2004 and that three-quarters of this financing is external. Despite these increased contributions, there is still a large gap between what is needed and what is provided. At the global level, UNAIDS (2005) estimates that up to $18.8 billion annually will be required by 2007. Further increases in funding will necessary as the numbers of infected persons increase and the duration of their lives and demands for medical care rise. Yet the magnitude and characteristics of available financing for HIV/AIDS prevention and control poses its own challenges. The boom in external funding for HIV/AIDS is unprecedented. From 2000-02 to 2002-04, the average level of external funding increased dramatically in Lesotho (1,100 percent), Swaziland (951 percent), Tanzania (394 percent) and Zambia (698 percent) (Lewis 2006). External resources for HIV/AIDS now dwarf public expenditure on health in the most affected countries, and, in some cases, have been at the expense of the remainder of the health budget. In 2003/4, for example, Ethiopia's external flows were equal to the government's health budget, but in Uganda and Zambia, AIDS funds exceeded all public spending on health by almost 185 percent (World Bank 2005). As Lewis (2006) asks, "how [can] countries...accommodate and wisely allocate new resources for HIV/AIDS while their overall health spending declines?" The channeling of the majority of external financing outside of the budget -- Foster (2005) has estimated up to 80 percent of the total -- also creates problems for sound health and fiscal policy. These pressures create a context where the risk of misuse of funds can be high, particularly as current program performance is essentially measured by the magnitude and speed of disbursement (Tayler and Dickinson 2006). Bribes, corrupt officials, and inappropriate procurement undermine health care delivery in much the same way they do for police services, law courts, and customs whose functions become compromised by the culture of poor governance and corruption. Specific characteristics of AIDS treatment create additional incentives for misuse; expensive ARV treatment has been counterfeited or resold, or subject to excessive mark ups or other corrupt or extortive practices. Compounding these challenges is the fact that the prevention and treatment of HIV/AIDS places exceptional demands on the health system at all levels. Testing, requirements for rigid adherence to the treatment protocol1, treatment of opportunistic infections, 1A 2004 World Bank study in India found that if drugs are not administered according to strict protocols, resistant strains of the virus can develop, jeopardizing treatment options for all (Over et al 2004). 2 Glassman management of ART side effects, palliative care, and psychosocial support require patient-centered care. Effective patient-centered care involves coordination among functions and facilities, near perfect logistics systems and well-staffed and stocked facilities, both at primary and hospital levels. As programs become more successful, AIDS patients live longer with ARV and require significantly more medical care over their lifetimes, therefore increasing the pressure on health facilities to meet the new demand. Further, there is continuing pressure to apply more sophisticated technologies and treatment when first-line ARV regimens fail (Harries et al 2006). Finally, the emphasis on resource mobilization and quick disbursement neglects institutional and governance issues. HIV and AIDS programs rely on the existing infrastructure of physical and human capital. Where delays occur because of lack of a critical input ­ such as staffing, infrastructure, management, warehousing, logistics and information on performance and output ­ productivity is reduced and impact falters. Without attention to effectiveness, well-intentioned spending may have no impact. In summary, increased resource mobilization for the prevention and control of HIV/AIDS is vital, but it is equally important to ensure that health systems function so that services are delivered and health professionals are accountable to the public, government and, where they are providing funds, donors. The role of parliament. Parliaments are increasingly asserting their independence from the executive branch and actively performing their constitutionally-designated functions of overseeing the executive, representing citizens, making policies, and enacting laws. Legislative oversight is nowhere more important than over the budget. However, there is much work to be done to improve budgetary oversight by parliaments. A recent survey of African legislators found that parliaments are not taking full advantage of their constitutionally mandated powers to address the HIV/AIDS crisis (National Democratic Institute for International Affairs 2004). Oversight of the budgetary process is often limited by lack of know-how, capacity and demand for disaggregated budgetary data on the part of legislators, but also by budgetary complexity and a lack of transparency and detail in executive branch budgetary reporting. Oversight of budgets and expenditures by parliament can potentially improve effectiveness and reduce risks of misuse and inefficiency. In the case of HIV/AIDS, parliaments can ensure that HIV/AIDS funds flow through the budget and that the amount, distribution and utilization of funding are consistent with national policies and priorities as well as international best practices. II. Budgetary process Budgets are regularly produced financial plans of government comprising revenues and expenditures of the state. If resources were limitless, there would be no need to budget. In reality, governments have to make choices about the allocation of scarce resources to meet competing needs in society. Budgeting is effective in facilitating this process when it forces awareness of overall fiscal constraints, enables the prioritization of spending in line with policy objectives, and supports the efficient implementation of policies. How Parliamentarians Can Help Ensure Accountability 3 Wehner (2004) provides a useful overview of budgetary process and the role of key players; the section draws on his work. Despite the critical role of the budget process, as of 2004 only 10 of 66 low-income aid recipient countries have a moderately functional budget system according to the World Bank's Country Policy and Institutional Assessments2, suggesting many countries have poor budgetary data and practices (World Bank 2006). The highest performing countries in Africa are Benin, Burkina Faso, Mali, Tanzania, and Uganda. The remainder of countries scored at or below 3. These institutional weaknesses imply that the oversight task is complicated by the near total absence of budgetary data in an utilizable format. Key actors in the budgetary process. The Finance Ministry or Treasury coordinates and leads the budget process. This role involves preparation of short and medium-term macroeconomic projections which forecast the resources from taxes, donations and other sources that will be available for spending, formulating fiscal policy, guiding the drafting of the budget, and monitoring budget implementation. Spending departments or ministries are responsible for expenditures within their jurisdiction, such as health, education, family welfare and so on. Spending departments will often try to extract as many resources as possible to meet their goals, whereas the role of the finance ministry is that of the guardian of the public purse. Democratic constitutions require taxation and public spending to be approved by parliament. Therefore, the role of the legislature is to scrutinize and authorize revenues and expenditures, and to ensure the budget is properly implemented. Independent supreme audit institutions such as auditor generals or audit courts carry out an audit of government accounts in order to determine whether government did in fact implement the budget as passed by the legislature. Some of them also consider whether this was done efficiently and effectively. There is considerable heterogeneity in the institutional design, capacity, resources, and types of audit conducted, but the goal is to support the executive branch managers, the legislature, and citizens in the oversight of spending. Stages of the budget process. The budget process is governed by a timeline that is separated into four stages: (i) drafting; (ii) legislative; (iii) implementation; and (iv) audit and evaluation. The figure at the top of the next page describes these stages and the institutional roles associated. 2 Country Policy and Institutional Assessments (CPIAs) are carried out annually by the World Bank. CPIAs rate the quality of policies and institutions, including those related to budgeting. Bank staff score individual countries along an absolute 1 (worst) ­ 6 (best) scale in 16 different governance categories. Several rounds of scoring are used to ensure consistency in scoring across countries. A score of 4 or higher indicates a moderately functional system. 4 Glassman Stages of the annual budget process Finance ministry or Budget tabled in the Funds apportioned to treasury issues legislature spending departments to Supreme audit institution guidelines to spending Consideration by implement activities assesses departmental departments or agencies parliamentary Finance ministry accounts and Spending departments committee(s) monitors spending performance submit draft budgets Parliament accepts, Request for legislative Audit reports published Negotiation and final amends or rejects the approval of adjustment and reviewed by decisions by executive budget budget if necessary parliament Drafting Legislative Implementation Audit Before beginning of relevant fiscal year Fiscal year starts and ends Following end of fiscal year Source: Wehner 2004. Budgeting is a process rather than an event and budget cycles are ongoing and interconnected. The role of parliament should not be restricted to budget approval and the review of audit findings. For instance, in some countries, parliamentary committees ask government to report on the process of drafting an upcoming budget yet to be tabled, and legislators might request certain documentation that is used in the drafting process. During budget execution, the legislature should have access to actual revenue and expenditure data on an ongoing basis in order to keep track of progress in implementation of the approved budget. This regular monitoring allows for earlier identification of problems in budget execution, before they result in significant deviations between the approved budget and actual revenues and spending. When parliamentarians follow the entire budget process as it unfolds they will be in a position to acquire relevant expertise and to keep track of emerging issues. Legislative effectiveness in budget scrutiny is enhanced by continuous oversight. III. Public policy for HIV/AIDS prevention and treatment While the search for a vaccine or cure for HIV/AIDS continues, the greatest hope for combating the disease is through the prevention of transmission. A well-known World Bank report (Over and Ainsworth 1997) found that public policies directed to correct for three types of market failures could be effective in the reduction of transmission rates. In keeping with the principles of public economics, the report suggested that governments finance or directly implement interventions that are essential to stopping the spread of HIV, but that private individuals or firms will not have sufficient incentive to finance on their own. In particular, the Bank recommended the provision of public goods such as information, the reduction of the negative externalities of behavior that spreads HIV such as unprotected intercourse and unsafe injecting practices, and the protection of the poor from HIV infection through, for example, subsidy of condoms and treatment of sexually transmitted infections. This strategy was thought to prevent the largest number of subsequent infections. How Parliamentarians Can Help Ensure Accountability 5 Yet while cost-effective prevention plays an important role in stemming the epidemic, supporting those already infected in living healthier, longer lives is also crucial to minimizing the economic impact of the epidemic, and the international consensus is that the two need to advance in parallel. Until recently, life-prolonging treatment was available only to a tiny fraction of HIV-positive people in Africa. However, the emergence of a simpler treatment regime, the dramatic drop in the cost of ART, an international consensus on a medical treatment protocol for resource-limited settings, and increased international funding for ART in low-income countries have made scaling up of programs a potential reality. However, the challenge of scaling up from current initiatives to the comprehensive prevention and treatment programs needed in Africa poses significant logistical and support problems. The challenges noted by the United Nations' Economic Commission for Africa's Commission on HIV/AIDS and Governance in Africa (2004) include: (i) low and declining numbers of health professionals; (ii) high dependence on external financing for ART; (iii) inadequate laboratory and patient care infrastructure; (iv) poor patient follow-up leading to low adherence and the risk of emergence of drug-resistant strains; and (v) drug shortages and delays in drug distribution. Each area represents a special need for oversight and getting the balance right among program priorities more of an art than a science. Role of public policy and expenditures in achieving better results. As noted earlier, the budget is a government's most powerful economic policy instrument, and therefore it's major opportunity to influence the HIV and AIDS epidemic. The information and analyses required to effectively oversee public financing for health, including HIV and AIDS programs, can be divided into five broad areas: (i) level of spending; (ii) allocation of spending; (ii) execution of spending; (iii) inputs purchased by spending; (iv) outputs produced by inputs; and (v) outcomes produced by program (see Table 1). This paper focuses on oversight of (i)-(iii) and refers the reader to alternative sources for (iv)-(v).3 Analysis of a country's HIV and AIDS programs will often require examining more than one budget, as well as addressing questions about which actors are shaping HIV and AIDS policy and which ones are delivering the services. External financing for HIV and AIDS programs, whether project-specific or general budget support, is important to consider. Health systems in many African countries rely heavily on these external funds. 3Although this note is focused on budgetary oversight, it should be noted that there are many areas where parliamentarians might take an activist role with respect to HIV/AIDS legislation. Such issues include: discrimination against vulnerable groups in the workplace, schools and health care facilities; equal access to health care; voluntary testing and informed consent in HIV testing; confidentiality of medical information; partner notification of HIV status; regulation of blood safety standards; regulation of HIV- related goods and services; intentional exposure or transmission of HIV; HIV prevention and care services in prisons; gender equality with respect to property rights and marital relations; care and support of orphans; protection of rights of human participants in HIV/AIDS research; among others. 6 Glassman Table 1: Potential indicators for parliamentary oversight of HIV/AIDS programs Area Objective Indicators (potential sources of data described below) Level Assure that the level of spending· % GDP per capita allocated to health is sufficient to meet goals set, · % total government budget allocated to health while maintaining fiscal and (Abuja target) macroeconomic stability as · % health budget allocated to HIV/AIDS and RH defined in MTEF · Budgeted expenditure per PLWHA or per woman of reproductive age Allocation Assure that budget allocations · Correlation of HIV/AIDS budgetary allocations correspond to legislated and number of PLWHA (or STI cases or other priorities and distribution of proxy) by geographic region need as expressed through, for · HIV/AIDS budget by type of intervention example, HIV prevalence rates (prevention, treatment, etc.) or poverty rates · HIV/AIDS budget by function (human resources, capital, drugs and other inputs) · HIV/AIDS budget by type of provider (public, private, not for profit) · HIV/AIDS budget by type of beneficiary (poor, orphans, other vulnerable) Expenditures Assure that budget expenditures · % expended/budgeted are executed as budgeted; assure · Correlation of HIV/AIDS expenditure and that the poor are protected from number of PLWHA (or STI cases or other catastrophic expenditures proxy) by geographic region associated with HIV/AIDS · HIV/AIDS expenditure by type of intervention health care (prevention, treatment, etc.) · HIV/AIDS expenditure by function (human resources, capital, drugs and other inputs) · HIV/AIDS expenditure by type of provider (public, private, not for profit) · HIV/AIDS expenditure by type of beneficiary (poor, orphans, other vulnerable) · HIV/AIDS out of pocket spending per PLWHA by SES Inputs Assure that inputs purchased are · Baseline indicators on procurement done so efficiently and with no (www.oecd.org/dataoecd/12/14/34336126.pdf) misuse of funds. Outputs Assure that inputs have · UN General Assembly recommended indicators produced the desired outputs. (% trained, % receive services, etc.) (http://data.unaids.org/publications/irc- pub06/jc1126-constrcoreindic-ungass_en.pdf) · preventive service utilization frequency per capita · characteristics of service users (risk category, SES, gender, etc.) Outcomes Assure that outputs have · UN General Assembly recommended indicators produced the desired outcomes. (http://data.unaids.org/publications/irc- pub06/jc1126-constrcoreindic-ungass_en.pdf) The problem is that donor funds don't always flow through the budget, particularly in the case of project-specific activities, making this funding stream difficult to oversee. A How Parliamentarians Can Help Ensure Accountability 7 recent study of 14 countries receiving poverty reduction support credits found that only 20 percent of donor commitments were provided as general budget support. Fully 50 percent of assistance ran outside the budget, and 30 percent escaped government reporting altogether (Foster 2005). This phenomenon is troubling from a good governance perspective; in addition, aid that flows outside of budgets has been found to result in uneven access and in financing low priorities when compared to national development plan priorities. In spite of these issues, those working in HIV/AIDS have found that partnering with non-governmental agencies and the private sector is key to an effective response to the epidemic. However, it is entirely feasible to partner with these groups while still channeling resources through government budgets (although it does run into the Ministry of Finance concerns described in Box 1). Box 1: Guardians of the public purse The role of finance ministries as guardians of the public purse is frequently politically unpopular, particularly among ministries and stakeholder groups working to prevent and control the AIDS epidemic. Why? Simply, the problem is urgent yet funds are insufficient, volatile and unpredictable. The nature of the fiscal cycle requires that loan and grant financing fit within the medium-term resource envelope that is available to government, that is, revenues should at least equal expenditures in a given fiscal year. Even if budgets are balanced in the current year and grant financing is accommodated within a budget ceiling established for the health sector, a finance ministry may refuse external assistance for AIDS programs because a one-time grant implies scaled-up expenditure on human resources or drugs that the government has no means to independently finance and cannot be easily downsized in the future (Heller 2006). Further, even if aid commitments are long-term, the unpredictable timing of external financing can make additional human resource contracting impossible in contexts where fiscal space is limited (Lewis 2006). Finally, when spending departments or ministries under-spend with respect to budget and have no data on the dynamics and effects of their HIV and AIDS expenditures, finance ministries may be concerned that the funds are inefficiently utilized, regardless of the importance of the disease. It is also important to understand the role of private contributions, including out of pocket expenditures by patients or expenditures by private sector entities (insurance plans, religious groups, etc.). These amounts may be reflected in the budget in the form of fees collected or assumed in the level of funding for a particular activity. Capturing the impact of private contributions on the provision of health care is often difficult, but the fact that low-income populations can be very sensitive to out of pocket expenses may make it an important factor. Comprehensive national health accounts, with HIV and AIDS-specific sub-accounts, are the best means to monitor the level of committed expenditure. This tool is described in the following section of the paper. A limitation is that national health accounts are usually available 1-2 years after a budgetary year. 8 Glassman Levels of spending. As important as the objective of oversight on the level of spending is, there is no "right answer" to the question of what is the optimal level of public spending on health or HIV/AIDS programs. Public spending levels are determined by technical criteria (what is the level of financing required to bring the epidemic down to the lowest possible incidence and prevalence rates given the effectiveness of existing interventions and their prices?), international goals and commitments, interests of external financing agents, competing budgetary priorities, political interest groups and macroeconomic stability requirements, among others. Even from the technical point of view, however, few countries have accounted fully for existing financing and fewer still have established the optimal level of financing required to reduce the incidence and prevalence of HIV/AIDS. As a result, it is difficult to set a target range for HIV and AIDS program spending at the national level. Alternative analyses might look at whether a country is spending more or less than countries with similar characteristics, however, a problem with this approach is that it tends to focus almost exclusively on the inputs, that is, the amount of spending relative to income, and fails to consider the main goal of spending which is better health (Savedoff 2003). An assessment of the level of spending should also take macroeconomic and fiscal trends into account. Although tax revenue yields seem apparently irrelevant to HIV and AIDS programs, for instance, these revenues or lack thereof ultimately affect funding levels as even high-priority areas cannot remain immune from overall budget trends for long. In most cases, a useful short-term objective is that of convincing the executive and donors to account for the totality of public and external aid-financed HIV/AIDS projects as part of the government budget. While this practice is not as common as would be ideal, it is feasible. The Millennium Challenge Corporation, for example, requires that its projects provide regular budget reports to the Ministry of Finance even though the execution of the funds is controlled by external groups. Allocation of spending. Once the relevant activities in the relevant budgets have been identified, the next steps are likely to focus on whether the programs are designed in a way to provide the desired outcomes and whether the activities are adequately funded. One element relates to the geographic allocation of expenditure. Expenditure is considered to be pro-PLWHA (people living with HIV & AIDS) or pro-poor when it is distributed according to the distribution of need or risk. For example, the attached regional map depicts the estimated HIV seroprevalence levels in Southern Africa (http://gamapserver.who.int/mapLibrary/Files/Maps/SouthernAfrica.jpg). With country- specific epidemiological data, such as that displayed for South Africa and Swaziland, a simple analysis of the distribution of expenditure vis-à-vis the number of reported AIDS cases could allow for a gross assessment of the extent to which public spending of HIV/AIDS was indeed targeted according to need. However, it should be noted that epidemiological surveillance systems may not track well to the administrative divisions of a country, such that it may be difficult to judge how best to channel funds according to need. Further, certain types of expenditure may only occur in a location that has, for example, a hospital, which would tend to concentrate expenditure, but this characteristic in and of itself may not necessarily imply that this spending is inappropriate as patients may travel from high-risk settings to lower-risk settings with better infrastructure to provide the needed health services. How Parliamentarians Can Help Ensure Accountability 9 It is critical to note that although allocation indicators provide information that may suggest over- or under-emphasis on a given area, there is little research that would allow us to state normatively what is the appropriate level and composition of expenditure that would achieve impact on controlling the epidemic or improving reproductive health. UNAIDS has produced a listing of essential services and minimum coverage levels required to achieve an impact on the epidemic under different epidemiological scenarios (low level, concentrated or generalized epidemic) (UNAIDS 2005). For 2006, for example, an ideal allocation among HIV and AIDS program activities looked like this: 56 percent prevention, 20 percent treatment and care, 11 percent orphans and vulnerable children, 10 percent program costs and 3 percent human resources. In the future, specific country studies should be undertaken to provide input into this oversight; such an analysis should seek to determine the full cost of implementing desired programs and use these funding levels as a benchmark for assessing government budgets. Further, it is important to assess whether the allocation of spending has the potential to crowd out the prevention and treatment of other diseases and conditions. In principle, all patients suffering from illness, regardless of cause, should be equally eligible for public assistance, particularly if they are poor (Over 1998). For example, a patient suffering from terminal cancer should have the same right to public support as an HIV-infected patient. However, this is difficult to enforce in practice. Purchasing inputs. Budgets can't add value if there is little relation between the budget on paper and the way money is actually spent. As a result, oversight of input purchasing is critical, particularly in the health sector where expenditure misuse and corruption issues can be prominent (Lewis 2006; Tayler and Dickinson 2006). Human resources. From an oversight perspective, there are four main human resources issues that may impact the effectiveness of policies and programs to reduce HIV/AIDS and promote reproductive health. A first issue is simply that the lack of qualified human resources ­ attributable both to a shortage in absolute terms and to poor distribution with respect to need -- is considered the binding constraint to the scale up of the prevention and treatment of HIV/AIDS. Recent research suggests that scaling up treatment with ARV would require between 20 and 50 percent of the available health workforce in four African countries, though less than 10 percent in the other 10 countries surveyed (Smith 2004). In general, however, the need for health workers outstrips supply and has a direct impact on coverage rates. For example, the Joint Learning Initiative estimates that, on average, countries with fewer than 2.5 health care professionals per 1000 population failed to achieve an 80 percent coverage rate for deliveries by skilled birth attendants or for measles immunization (Chen et al 2004). A second area of concern is that increased financing for HIV/AIDS may result in the already short supply of provider time being disproportionately skewed towards this single disease to the detriment of other conditions. While this concern has been aired, it is clear that in high prevalence countries, this shift occurred years before additional external financing in the form of the Global Fund and PEPFAR appeared on the scene. For example, a 1996 study of Kenyatta National Hospital in Nairobi, Kenya (Floyd and Gilks 1996), found that between 1998/89 and 1992, the number of HIV-infected patients admitted daily more than doubled. The mortality rate of HIV-negative inpatients rose by 10 Glassman more than two-thirds, indicating that those with less serious conditions could not even get admitted to the hospital. The Global Fund and PEPFAR are only recently beginning to implement activities; as a result, the recent System-Wide Effects of the Fund research program based at Abt Associates was not able to document the size and nature of human capacity constraints on implementation of GF-supported activities. To date, the key human resource capacity constraint in dealing with GF resources "has occurred at the central level ... [related to]...the process of applying for, planning, and launching GF- supported programs." A third area relates to the general performance challenges in the health sector, such as absenteeism and low productivity. Absenteeism is a widespread phenomenon. A recent multi-country study based on multiple, unannounced facility visits reports absenteeism rates ranging from 23 to 40 percent in the health sector (World Bank 2004). Even when health personnel report to work, they may display notoriously low productivity as measured by the number of patients seen on a given day. These practices are thought to be associated with low pay levels, but also the lack of incentives and supervision that would enable better performance and accountability to health system users. In general, additional responsibilities associated with the implementation of scaled-up external financing for HIV/AIDS do not correspond to increases in grade or salary. An exception has been Malawi where decision makers were able to utilize GF monies to facilitate the implementation of a broader health sector program to increase all health worker salaries. Finally, clientelistic and corrupt hiring practices (buying and selling positions, bribes) have also been observed. A fourth concern relates to the absence of managerial and planning skills that are necessary for scaling up. While much training has been financed over the years, it has been concentrated on clinical issues, while the managerial and planning skills critical to successful implementation are rarely addressed. Based on its interim report on three African countries, the System-Wide Effects for the Global Fund project has recommended an assessment of human resource constraints to scaling up that could be developed into an overarching national-level strategy that would promote the implementation of proven strategies of health work retention and motivation. Procurement. Procurement is at the core of government's discretionary spending. Irregularities in the form of procedural violations, embezzlement and collusion are important to monitor and sanction publicly. An adequate legislative framework for fair and transparent procurement is an important pre-requisite to effective oversight. Timely procurement of drugs is a critical piece of any treatment program; experts attribute much of the failures observed in these systems to limited capacity to forecast demand, manage inventory and conduct procurement activities, in addition to erratic and unreliable donor financing (Levine 2006). Given the magnitude and rapid scale up of HIV/AIDS financing, it is also an area that is particularly vulnerable to misuse and corruption. In 2006, Transparency International dedicated a section of its annual corruption report to HIV and AIDS programs and the How Parliamentarians Can Help Ensure Accountability 11 Global Fund itself has pulled financing when discoveries of misuse occur.4 Some of the cases highlighted include: Kenya's National AIDS Control Council for irregularities in procurement procedures, donations that could not be accounted for and embezzlement. The Efficiency Monitoring Unit in the Office of the President documented that at least half of the money transferred to community-based organizations was lost through wanton theft or private use. Director and senior staff at the Zimbabwe National Network for People Living with HIV/AIDS were suspended after allegations of corruption. The network received more than US $1.8 million from the National AIDS Council between 2003 and 2004. Nigeria's ARV program attracted much criticism in 2003 when treatment centers began handing out expired drugs and rejecting patients, but it is not yet clear whether the prime cause was corruption or a weak drug procurement, supply and distribution system that was unable to respond to the demands that the rapid scaling-up of the program had placed upon them. A Ugandan commission has been formed to investigate alleged mismanagement of Global Fund grants and has heard assertions of inflated expenditures, inaccurate receipts and improper allocation of funds, amounting to what head of the commission James Ogoola recently called a "pile of filth" (PlusNews, 4/3). The Global Fund in August 2005 announced the suspension of five grants to Uganda worth $367 million after an audit of one of the grants found evidence of mismanagement by the Ugandan Ministry of Health's Project Management Unit, which was established to implement the grants. The government later that month appointed a four-member commission to investigate the allegations and brought in international accounting and auditing firm Ernst & Young to take over temporary management of the country's AIDS funding from PMU (Kaiser Daily HIV/AIDS Report, 9/1/05). The Global Fund in November 2005 resumed funding to Uganda after signing an agreement with the Ugandan Ministry of Finance, Planning and Economic Development that strengthened the oversight of programs receiving funds (Kaiser Daily HIV/AIDS Report, 11/11/05). This area has acquired greater visibility in recent years and has led to a profusion of efforts to monitor the quality of public procurement systems and enhance their operability. The simplest form is monitoring is the physical observation of procurement practices and outcomes. A second form of procurement monitoring focuses on transparency: the publication of procurement opportunities and outcomes. In many countries electronic procurement systems have tremendously increased the visibility of public contracting and allowed government and non-governmental bodies alike the opportunity to review the distribution of contract awards as well as the price the government pays for its goods, works, and services. A third form of procurement monitoring is assessing performance of public procurement systems using defined performance indicators. The Baseline Indicator Set for Procurement (BIS) has been developed as part of the OECD-DAC Working Party on Improving Aid Effectiveness and 4Given that programs that provide ARV must be able to assure the steady supply and administration of medicines (or face increases in drug-resistant strains requiring more costly second-line treatment), the implications of pulling funding from a program that finances ARV are important and may serve as a powerful disincentive for reporting abuses. 12 Glassman has been adopted and used in more than 10 countries in the first six months after it was finalized. The BIS to be used as a regular monitoring tool and is available at www.oecd.org/dataoecd/12/14/34336126.pdf. Audits. Audits are intended to be the primary mechanism to monitor the appropriate utilization of expenditure. However, where auditors are not independent or under funded, audits do not occur as planned. In Zimbabwe, for example, the government has imposed an "AIDS levy" since 2000 whereby employees contribute 3 percent of their gross salaries towards a fund administered by the National AIDS Council (NAC). It is estimated that the government collects about US $20 million per year through this mechanism, but no information about how the fund is used and who benefits from it has ever been made public. In March 2005, the health ministry ordered an audit of the NAC, but to date it has not yet been published. Outputs and outcomes. The monitoring and evaluation of outputs and outcomes may or may not be part of a government's budgetary process. In July 2005, the UNGASS provided technical guidelines to National AIDS Councils on output and outcome indicators to measure the performance of HIV and AIDS programs. While most of the indicators do not include targets and there are other limitations associated, the document represents the current international consensus on monitoring HIV/AIDS programs' outputs and outcomes. Most programs generally monitor standard primary outcomes, such as the total number of patients ever starting ARV, the number alive and receiving treatment, and the number who died due to the discontinuation of treatment or improper maintenance of medication. Secondary outcomes relating to functionality are less frequently monitored and include the patient's ability to walk home unaided, side effects of treatment, and drug adherence. Such data are available from HIV units within Ministries of Health or from coordinating councils. In rare cases, these data are published, as is the case in Malawi (Harries et al 2006). Parliamentary demand to link expenditures to outcomes of this type may make an important contribution to accountability and greater policy effectiveness. IV. Available expenditure tracking tools A. Domestic resource tracking tools Parliaments can make more effective use of existing provisions and mechanisms for oversight. Simple budgetary analyses can allow for a calculation of the financial performance indicators listed above. Using these figures, MPs and parliamentary committees can scrutinize how allocated funds are spent, the reasons behind under- spending allocated funds, and determine whether the total amount allocated to HIV/AIDS within the health budget is appropriate. All budgets are organized by ministries or departments, the so-called "administrative unit." While the health ministry is probably the most important for HIV and AIDS programs, it may also be necessary to look at other ministries such as Public Works for questions of infrastructure, Social Assistance or Family Welfare for services directed to orphans and vulnerable children, or statistical agencies for data collection. Within a ministry, budgets are usually broken down by bureaus or sub-groups of the agency, and How Parliamentarians Can Help Ensure Accountability 13 then by programs. Generally, it is the program-level that is most important for identifying key activities with respect to HIV/AIDS. Budgets are also organized by "functional" and "economic" classifications. Function means the expenditures are classified according to the purpose for which they are to be used. Most HIV/AIDS and reproductive health spending will be in the health function, but some relevant spending is likely to be found elsewhere. Economic classifications distinguish between funds that are used, for instance, to pay wages or to construct a clinic. An excellent resource for data and analysis of budgets are civil society organizations (CSO)5 that watchdog AIDS, budget, and governance issues. Some CSOs track how government actually spends funds, with the aim of identifying inconsistencies, leakages and bottlenecks in the flow of financial resources or other inputs. For example, the Institute for Democracy in South Africa (IDASA) monitors HIV/AIDS budget allocations and reports findings on a regular basis, as shown in Table 2 below, which is reproduced from one of IDASA's "Budget Brief" series (Ndlovu 2005). Table 2: IDASA Report on HIV/AIDS Allocations in South Africa B. Medium Term Expenditure Frameworks (MTEFs) If a country participates in the Highly Indebted Poor Country (HIPC) Initiative, a Poverty Reduction Strategy Paper (PRSP) is prepared. The PRSP describes a country's macroeconomic, structural and social policies and programs to promote growth and reduce poverty, as well as associated external financing needs. PRSPs are prepared by governments through a participatory process involving civil society and development partners, including the World Bank and the IMF. One of the key elements of a Poverty Reduction Strategy is an expenditure plan that outlines the resource envelop for budget allocations over several years. These plans have been labeled "Medium Term Expenditure Frameworks" (MTEFs). The purpose of an MTEF is to indicate the size of the financial resources needed during the medium term, usually between three to five years, in order to carry out existing policy. The MTEF concept differs from multiyear budgeting, which involves fixed appropriations for a certain number of fiscal years. Usually, only the first year of an MTEF is approved by the 5A list of CSOs that perform these functions in Africa is attached in Annex 1. 14 Glassman legislature as the annual budget, whereas the outer years are nonbinding projections of the future expenditures given existing policy. The firmer these projections become, the more they move to the centre stage of the budget process and form the basis for the annual negotiation of allocations, resulting in a system of `rolling budgets.' All OECD countries have medium term frameworks, and many developing countries are in the process of adopting them. The level of sectoral detail is MTEFs is typically limited to overall sectoral expenditures rather than to allocations within each sector. Parliamentarians concerned about HIV/AIDS and reproductive health need to pay close attention to the way in which these areas are addressed in PRSPs and to the allocation of resources for them, but will also need to look beyond the MTEF to sectoral expenditure plans and track actual spending patterns that evolve during implementation of the country's poverty reduction program. C. National Health Accounts (NHAs) and sub-accounts Implemented in 60 middle and low-income countries, NHAs are an internationally recognized framework for measuring total (public, private and donor) health expenditures in a given country. NHA methodology tracks the flow of funds through the health sector, from funding sources, through financing agents, to providers and functions. In addition to illustrating total national health care spending patterns, the NHA framework has been adapted to enable "subanalyses" that can be used to capture data on a specific disease by breaking down expenditures on related individual services and disease areas, such as HIV/AIDS interventions. In describing the flow of HIV/AIDS funds, NHA reveals not only how much is spent on HIV/AIDS but also the types of services the population uses ­ prevention, treatment of opportunistic infections, ARV treatment, etc. ­ and how those services are financed ­ by government funds, donor grants, or out-of-pocket payments by households. By disaggregating HIV/AIDS expenditures in this manner, the NHA subanalysis provides legislators, policymakers, program managers and donors with a clear picture of the magnitude and composition of spending. Further analysis has, in some countries, produced information on HIV/AIDS expenditures by socioeconomic status, geographic grouping, and/or gender, allowing for an assessment of the financial burden of the disease on different subpopulations. Note, however, that it is critical to complete a comprehensive NHA before embarking on sub- accounts. Most of the indicators described in Table 1 are sourced in a NHA exercise. The subaccounts have illustrated that the magnitude of public spending for HIV/AIDS is still quite low and while donor-financed expenditures are increasing and out-of-pocket household payments decreasing (good news for poverty reduction objectives), they fall far short of the funds required to meet HIV/AIDS prevention and control goals. HIV/AIDS subaccounts are available for Burkina Faso, Ghana, Kenya, Rwanda and Zambia. Even if HIV/AIDS subaccounts are unavailable, budget analysis may be done to illustrate the issues. Reproductive health subaccounts also exist in some countries. To allocate resources to the most cost-effective interventions for combating the HIV/AIDS pandemic, parliamentarians need to know what works and how much is being spent on specific interventions. Financial indicators are conspicuously absent from many How Parliamentarians Can Help Ensure Accountability 15 M&E frameworks. Each global initiative has developed its own monitoring and evaluation framework for reporting by grant recipients, yet few have added financial indicators to track the very investment results called for within initiative goals. Additionally, national AIDS strategies often set their own targets and countries are encouraged to report on core indicators for tracking progress towards targets set in the UNGASS Declaration of Commitment on HIV/AIDS-neither highlights the importance of tracking expenditures against achievements. Multiple monitoring and evaluation frameworks persist despite agreement, between various HIV/AIDS stakeholders, on "the three ones" principle: that each country should have one HIV/AIDS budget, one national coordinating committee and one M&E system (Cohen 2006). Estimating the gaps between resources promised, committed, and expended has been effective in advocating for increased mobilization of resources for HIV/AIDS, however, it is now necessary to demonstrate that the money is being used wisely. Current initiatives to objectively track resources focused on HIV/AIDS at the country level are supported by UNAIDS and bilateral donors such as USAID. As monitoring and evaluation systems continue to evolve toward a common framework, it is critical that financial indicators be included, tracked and reported. To learn more about efforts in tracking resources, the UNAIDS (www.unaids.org) and IDASA (www.idasa.org) web sites are useful, as is the Partners in Health Reform plus project at www.phrplus.org. D. Public Expenditure Reviews (PER) and Tracking Surveys (PETS) Public expenditure reviews are periodic World Bank staff reports that study a government's objectives and assess whether the government is using public financing to achieve their objectives efficiently, equitably, and sustainably. Key questions to be answered might include: (i) Is government's policy and implementation sound?; (ii) Are key public goods and services being adequately provided?; (iii) Are programs intended to reach the poor doing so effectively?; (iv) Are some programs merely crowding out private activities?; and (v) What are the institutional weaknesses of the budget system that need attention? The PER are also intended to support government capacity to translate PRSP commitments into real budget implementation, to advise government on options for reform of policy and management, among others. While few PER focus specifically on HIV/AIDS, many analyze public spending on health and its adequacy, equity, efficiency, contribution to risk protection (protected households from impoverishing expenditure on health), delivery mechanisms, sustainability and impact. Some PERs provide break outs of program-specific financing; the case of Vietnam is provided in a World Bank document that provides guidelines for the preparation of PER in the health sector. PETS are tracking tool s that aim to answer the question "Does public money spent on health actually reach frontline health facilities?" The studies track the flow of public resources through the different layers of the administrative hierarchy to individual service providers and develop quantitative estimates of fiscal leakage, that is, the failure of resources intended for frontline service providers to reach their intended destination. The WB and other organizations have conducted PETS in over two dozen countries. A recent review by Lindelow et al (2006) documents results in the health sector; Table 3 16 Glassman excerpts PETS findings from Africa. As can be seen, local capture, leakage and bureaucratic impediments result in 80 percent of non-salary expenditure never reaching frontline clinics in Ghana, 70 percent in Uganda and 40 percent in Tanzania. These governance challenges raise doubts about the value of simply increasing funding, or of the need for "just" more health workers. Table 3: Review of PETS Results for Africa in the 2000s (Lindelow et al 2006) Country Year Sample Leakage Other findings Ghana 2000 200 facilities; 40 Leakage of non-salary Greater leakage between districts recurrent expenditures center and district than estimated at 80% between district and facility. Mozambique 2002 90 facilities; 167 staff; Some evidence of Documented delays and 679 users leakage of drugs in bottlenecks in budget transfer from provinces execution and supply to districts, within the management; primary health care incomplete registering system of user fee revenues by facilities; absenteeism estimated at 19% Nigeria 2002 252 facilities; 30 local No firm estimate of 42% of staff experience govt.; 700 staff leakage. salary delays despite sufficient budget Rwanda 2000 351 facilities;40 districts Some evidence of Evidence of delays in leakage between regions budget execution and and districts. low execution rates (80% of non-wage funds released at year end) Senegal 2002 100 facilities; 10 Some evidence of Delays in fund transfers. districts; 37 local govts. leakage at regional and communal level in allocation of non-salary resources. Tanzania 2001 20 facilities; 5 districts Leakage estimated at Substantial delays at all 40% levels, especially non- wage expenditures Uganda 2000 155 facilities Leakage of specific drugs and supplies estimated at 70% in government and private non-profit facilities. Reinikka and Svensson (2003) have reviewed applications of PETS in the health and education sectors in ways that can help parliamentarians focus accountability interventions on specific trouble spots. In Uganda, for example, they found that in 1995, only 22 percent of the money allocated to schools by the central government actually reached those schools. The PETS process showed that most of the leakages were in non- wage allocations and that they occurred most frequently at the local government level ­ information that was used to target more focused and efficient accountability How Parliamentarians Can Help Ensure Accountability 17 interventions. As a result, when the amounts of central allocations were made public in local newspapers and monitored by parent-teacher organizations, the leakage rates dropped dramatically. E. Censuses and surveys In addition to these frequently available sources, censuses and household surveys provide key information for the indicators described in Table 1. A census provides denominators for both epidemiological and financial performance indicators. A simple measure such as HIV/AIDS spending as a percentage of GDP per capita is meaningless without somewhat accurate data on total population size and conclusions would vary enormously based on these numbers. Household sample surveys, such as the LSMS and the DHS, allow for a more nuanced analysis of expenditure data and provide the information necessary to analyze expenditure according to different socioeconomic characteristics of a household. The DHS infrequently includes expenditure modules but has recently included a seroprevalence survey in some countries. This information should be a critical input into policy dialogue and formulation. The Rwanda 2005 seroprevalence survey, for example, found levels among the productive age population that were substantially lower than those forecasted earlier in the decade, resulting in ­perhaps ­ an over dimensioning of the level of financing allocated to HIV/AIDS control efforts. The level of representativity of these sample surveys should be crucial for lawmakers in parliament; if surveys are representative at the level of representation of each lawmaker, for example at the state or province level, more evidence-based budgeting and oversight of HIV/AIDS programs can be conducted. Unfortunately, fiscal constraints have decreased the regularity of censuses and surveys in many countries, leading to a near total absence of the type of information required to generate basic budgetary performance indicators. F. Donor accountability requirements Another source of information regarding financial flows for HIV/AIDS is those related to donor accountability requirements. Even where a recipient country government does not have a comprehensive picture of expenditure, it is possible to piece together a picture from the reporting requirements of donors. For example, the Global Fund publishes a listing of all awarded grants and grantees along with disbursement progress to date; the GF also estimates the number of persons receiving ART under each grant based on grantee reporting. With a few exceptions, donor accountability has not yet evolved beyond expenditure use information and narrative evaluation of programs. V. What can parliaments do? The need for oversight is clear. Making oversight effective, however, may take time. On the one hand, budgetary systems are weak, lack the capacity to provide real time information to legislators and are insufficiently disaggregated in their standard formats to 18 Glassman allow for the estimation of the indicators suggested in this module. On the other hand, the heavy dependence on external aid for HIV and AIDS program financing in sub- Saharan Africa and the extent to which this expenditure occurs off-budget implies that even if budgetary systems worked well, parliaments would only oversee a fragment of overall spending on HIV/AIDS. In the short-term therefore, Parliaments may work effectively in three main areas: (i) improving the production and availability of data; (ii) increasing transparency at every stage of the process; and (ii) reducing misuse and corruption. Production and availability of data. The best way to incentivize the production and availability of data is to request the data from the executive branch and utilize it during legislative sessions and in budget negotiations. Draw on the sources mentioned in this report ­ NHA, PER, PETS, MTEF, donor accountability requirements ­ to pose questions and propose improvements. For example, MPs can lobby for budgets that include a specific line item for HIV/AIDS in the health budget, which will assist in the oversight task. Further, MPs can draw on civil society organizations to carry out key analyses ­ analyzing bills, motions and draft laws with an HIV/AIDS lens -- in order to make relevant amendments. Increasing transparency. Parliament should work with the executive to foment the production and publication of the financial and other data that is necessary to effectively conduct oversight. For example, governments and health authorities should publish regularly updated information on the Internet on health budgets and performance at the national, local and health delivery center levels. This can be required through a so-called "sunshine" law ­ fiscal transparency and accountability legislation. Further, government departments, hospitals, health insurance entities and other agencies handling health funds must be subject to independent audits. Governments and health authorities have the responsibility to ensure that information about tender processes, including offers to tender, terms and conditions, the evaluation process and final decisions, is publicly available on the Internet. Donors and international agencies must also be transparent about what they are giving, when and to whom, and should evaluate their programs in terms of health outcomes and not level or speed of disbursement. To review this information, Parliaments in high prevalence countries should strive to have an HIV/AIDS standing committee. Misuse and corruption. In addition to increasing transparency, independent audits and civil society watch dogging are the best sources for detecting misuse and corruption in the health sector, as the Global Fund examples indicate. Where private participation in service delivery exists, as it does in most countries, it is critical that legislation allow for private citizens to file lawsuits charging fraud in government programs. These How Parliamentarians Can Help Ensure Accountability 19 "whistleblowers" can receive financial incentives and compensation for settled cases. Some of the tools mentioned in this module, such as PETS, are also useful to objectively document misuse of funds. In addition to these initiatives, Parliamentarians may wish to join the Coalition of African Parliamentarians on HIV and AIDS (CAPAH), an independent network of parliaments dedicated to work together on HIV and AIDS efforts, in collaboration with existing institutions and regional bodies such as SADC-PF and ECOWAS, to improve the effectiveness of oversight. Other support is available through a donor-finance project Parliamentarians for Women's Health which assigns expert interns to MP offices to support development and monitoring of legislation and finances surveys and polling in constituent districts on the issues of HIV/AIDS and women's health so that legislators can better reflect the needs and realities of their constituents in budget debates. VI. Case Example: Rwanda Rwanda is a small, landlocked country with a population of approximately 8.2 million persons and per capita GDP of $230. Its population density is among the highest in the world. Rwanda has made substantial progress in stabilizing and rehabilitating its economy to pre-1994 levels. Leaving aside countries in Africa that received oil windfalls, Rwanda's economy has experienced relatively high levels of growth in the past decade. Total poverty has declined in the post-genocide period to 64 percent; however, inequality is increasing, particularly differentials between urban and rural areas where approximately 85 percent of the country lives. Although Rwanda has made a remarkable transition from reconstruction to development in the eleven years since the genocide, the legacy remains and the country is likely to depend on significant levels of external assistance for years to come, in spite of agreements on debt forgiveness. Health status. Child and maternal mortality have fallen since 1994; however, they have yet to return to pre-genocide levels that were already among the highest in sub-Saharan Africa. The major causes of mortality for children under five are largely avoidable. Reproductive health is a critical issue in Rwanda. The country has one of the highest recorded rates of maternal mortality in the world, as well as one of the lowest rates of contraceptive prevalence and a relatively high number of births per woman of reproductive age (UNDP 2004). While performance has improved over time, in 2002, only 15 percent of all births were attended by a skilled birth attendant. With adult HIV seroprevalence of 4 percent6, the AIDS epidemic represents a challenge to Rwanda's health system in particular and development prospects in general. Poverty, war, and the genocide have resulted in vulnerable populations of orphans, child-headed 6Government of Rwanda. Demographic and Health Survey 2002. Results reported on Ministry of Health web site. 20 Glassman households, victims of rape, and widows, on whom the negative impact of AIDS is particularly severe. The Government has committed to stabilizing the spread of HIV during the period 2002 to 2006. Donors have joined this commitment and, in 2003, Rwanda became a recipient of Global Fund and PEPFAR monies. For the 2006-2008 period, Rwanda is authorized to receive up to $100 million USD. However, these funds are still in the process of being transferred to executors on the ground. PRSP and HIV/AIDS. Rwanda is a HIPC country with a Poverty Reduction Strategy Paper in place since 2002. This document articulates a medium-term program (2002- 2007), coordinated with a MTEF. Government domestic spending has been increasing since 2002. Given increased availability of external financing, implementation will begin in earnest on a full range of programs, including new constitutionally mandated institutions, free basic education, government counterpart contributions for a comprehensive HIV/AIDS treatment program, supplementary pay for health sector workers, human resource development in tertiary education, labor intensive rural works, and export promotion, all of which are drawn from the PRSP. The 2004 budget also reflects 0.5 percentage points of GDP in demobilization outlays that have been carried over from 2003. As a result, government priority spending is budgeted to rise to 45 percent of domestic expenditure. A 2005 PRSP performance review published by government points to significant achievements in human development, particularly in the health sector. Although the leading development partner is the World Bank, who has provided the majority of technical assistance to the PRSP process, and the Bank finances a large HIV/AIDS reduction project (the MAP project), there are few PRSP performance indicators related to HIV/AIDS (last of 13 action areas in the health sector identified as a priority by government) and these indicators are related to communication campaigns and condom distribution (no quantitative targets included in policy matrix) and establishment of a number of HIV testing and counseling sites. This focus is at odds with the fact that in 2002, 30 percent of total health expenditure on health goes towards HIV/AIDS and the arrival of GF and PEPFAR in 2003 have increased this share significantly. In 2005, the World Bank estimated that commitments for AIDS financing in Rwanda are 115 percent of total government health spending. This may reflect a disconnect between PRSP- governed priority setting and actual budgetary and expenditure allocations. General health financing. Rwanda is one of the few countries in Africa with a National Health Accounts series that allows for an analysis of the sources and uses of health expenditure. A 2002 round indicates that about 50 percent of total financial resources in the sector are from international cooperation, 10 percent from government (primarily civil service salaries) and 40 percent from private sources (33% is direct out of pocket expenditure by households). External financing is concentrated in disease-specific interventions (Table 4). The contribution of the state to the functioning of the health sector is limited though increasing over time, about 8 percent of the national budget, equivalent to 2.50 USD per capita annually. In spite of increases, the sustainability of How Parliamentarians Can Help Ensure Accountability 21 external finance flows in health is a major issue. Compared to other countries in East and Southern Africa, Rwanda is well below the average regional measure for overall health expenditure as a percentage of GDP but above the regional average for share of public spending of total health expenditure. In terms of use of the budget, a PETS in 2000 found that 80 percent of non-salary expenditure in the health sector was disbursed at the end of the fiscal year, indicating significant challenges in budgetary execution and/or liquidity problems. It should also be noted that Government has an excellent track record on budget execution ­ at or above 85 percent of total allocations between 2002 and 2005 -- indicating little problem with absorptive capacity up to 2005 (Scorraille et al 2006). A follow up PETS in 2003 confirmed earlier findings of substantial delays in budgetary transfers from the central administration to district health offices and identified troubling issues related to public expenditure management by government, such as poor bookkeeping and lack of internal financial controls and auditing requirements (Fofack et al 2003). Table 4 Indicators for Rwanda Case Example Indicators 2000 2002 HIV seroprevalence rates (adults) 5.1% 5.1% Number of PLWHA 200,000 199,279 Total Health Expenditure (THE) US$77,992,817 US$70,101,480 THE for HIV/AIDS US$6,009,287 US$10,313,032 % of THE allocated to HIV/AIDS 8% 15% General out-of-pocket spending per inhabitant $2.51 $2.13 HIV/AIDS out-of-pocket spending per PLWHA $11.27 $7.59 Total HIV/AIDS spending as % GDP (in current prices) 0.3% 1% Expenditure on HIV/AIDS by financing source: - Public 8% 9% - Private 43% 7% - Donors 49% 75% Expenditure on HIV/AIDS by provider type: - Public providers 33% 16% - Private providers 9% 3% - Government-assisted not-for-profit providers 5% 3% - Private pharmacies 7% 3% - Provision and administration of public health 46% 66% - Administration 0% 9% Expenditure on HIV/AIDS by function: - Preventive and public health programs 46% 66% - Curative care 48% 23% - Administration 0% 9% - Pharmaceuticals purchased at independent 7% 3% pharmacies Source: Republic of Rwanda, Ministry of Health, 2005. HIV/AIDS financing. According to the HIV/AIDS accounts for 2002, donors are the major financing source of HIV/AIDS health care. More than one-third of donor financing is HIV/AIDS related and 75 percent of HIV/AIDS health care is financed through these external sources. Households contribute 16 percent, while government 22 Glassman represents 9 percent of the total. Seventy-six percent of donor spending on HIV/AIDS is channeled through NGOs and is off-budget, and this share has increased since 2000. In 2002, financing was concentrated overwhelmingly in prevention activities; the arrival of GF and PEPFAR in 2003 will likely have modified this scenario. Key attributes of HIV/AIDS spending prior to 2003 indicate a declining level of spending on HIV/AIDS, a significant financial burden on PLWHA (represented by out of pocket spending ­ OOP ­ above) and the paucity of donor funds. NHA analysts attribute much of the aid increases approved in 2005/6 ­ a 100% increase in real terms -- to the availability of NHA expenditure data documenting the problems faced. Health policy. Prior to recent reforms, central government budgetary allocations covered only the salaries of health workers and, as a result, public health facilities relied on household contributions and fees, along with sporadic contributions from donors and NGOs. A household survey in 2003 found that 95 percent of respondents who needed to see a health provider but did not do so cited the high costs of care, and among those that consulted a health provider, 80 percent were dissatisfied with the costs (Fofack et al 2003). In response to these challenges, a new Health Sector Policy was adopted by Cabinet in early 2005. While the long-term financial sustainability of the system is a main challenge, the issue of health financing is not discussed systemically. Financing is treated under a number of different objectives: (i) under human resources, the introduction of incentive structures for qualification, specialization and geographic redistribution; (ii) under drugs, the implementation of a system of cost recovery at health centers according to which drugs are bought at district pharmacies and resold to patients at the lowest possible price; and (iii) under financial accessibility to health services, preferentially for the poor, the increase of public share of financing, the promotion of mutuelles (systems of health care pre-payment) and health insurance, the definition of a pricing policy for services and drugs and the subsidy of essential services to vulnerable groups (not defined). Further, a Mutual Health Insurance Policy was approved in December 2004 which specifies that the mutuelles would be the vehicle to provide health services and subsidies for these services to the poor.7 In 2005, the government estimated that 25 percent of the population is now enrolled in mutuelles; current estimates are higher due to a recently implemented requirement to enroll in a mutuelle prior to government issuance of identity documents and marriage licenses. Rwanda has negotiated a special scheme with the Global Fund by which Fund monies will finance mutuelle enrollment for the extreme poor, orphans, vulnerable groups and PLWHA. Rwanda has also been the site of several pilots to improve the productivity, quality and responsiveness of health providers through monetary incentive payments (Soeters et al 2006). These pilots appear to have improved performance; however, the organizational 7However, mutuelles evaluations and the mutuelle policy itself recognize that the requirement to pre-pay out of pocket represents an important obstacle to the affiliation of the poor and that, without government intervention to subsidize the poor, will tend to cover the relatively better off segments of the population. How Parliamentarians Can Help Ensure Accountability 23 structure that accompanies this scheme has not yet been translated into the mutuelle structure. As a result, the future of these mechanisms is still unclear and the institutional structures required to scale up understudied. The poverty or risk targeting of subsidies has not been the subject of an explicit government policy. The World Bank MAP project targets the poorest geographical areas. A recent government initiative to provide free delivery care is a mechanism to target user fee waivers; user fee waivers are also reportedly applied to female headed households which are over-represented among the poor. Political economy. The government plans that donor financing for health be structured predominantly through a Sector Wide Approach with the goal of enhancing the management of external resources and expanding budgetary support so as to channel resources more directly towards the development of the health system. The share of donor financing going to budget support is not publicly available. In 2003, a detailed mapping of donor activities and resources was conducted to highlight gaps and duplication of efforts; this led to the formation of a number of consultative and coordinating bodies, including a health cluster. The cluster meets frequently under the leadership of the Secretary General of the Ministry of Health and Belgian Technical Cooperation is the secretariat. Six health cluster working groups were also formed on the issues of performance-based contracting, mutuelles, HIV/AIDS integration into health system, human resources, disease control and intervention mapping; these groups are providing technical advice and position papers to government. A review of PRSP progress in 2005 by World Bank and IMF staffs noted that the strategy preparation process encountered bottlenecks relating to poor coordination within the sector. These were partially overcome through collective workshops and strong support to planning from budget support donors. However, the report notes that there continues to be a disconnect between health system development planning and the activities of some unnamed "vertical" programs within the sector which surfaced during the strategic planning process. Other sources report anecdotal reports of AIDS financing "crowding out" other types of health services and "distorting" priority-setting within government; however, this phenomena has yet to be systematically analyzed.8 8A detailed study on this topic will be undertaken by the World Bank in 2006. It is an urgent requirement as the US Government-financed PEPFAR will begin disbursements in Rwanda in 2006 as well, resulting in a substantial increase of available funds for HIV and AIDS programs -- $100 million USD for the 2006- 2008 period. 24 Glassman References Barkan J, Ladipo Adamolekun and Yongmei Zhou, "Emerging Legislatures: Institutions of Horizontal Accountability" in Brian Levy and Sahr Kpundeh (eds.) Building State Capacity in Africa. Washington, DC: World Bank Institute, 2004. Chen L, Evans T, Anand S, Boufford JI, Brown H, Chowdhury M et al. Human resources for health: overcoming the crisis. The Lancet 2004, 364: 1984-1990. Cohen J. The New World of Global Health. Science 2006, 311:162-167. Dmytraczenko T, De S, Chanfreau C, Kidane L. The Value of Beneficiary Analyses: Who Benefits from Funds Targeted For HIV/AIDS? Bethesda, MD: Abt Associates and PHRplus, 2006. www.phrplus.com Floyd K, Gilks CF. Impact of and response to the HIV epidemic at Kenyatta National Hospital, Nairobi. Report 1 (April) Liverpool, England: Liverpool School of Medicine, 1996. Fofack, Hippolyte, Chukwuma Obidegwu, Robert Ngong. Public expenditure performance in Rwanda: evidence from a Public Expenditure Tracking Study in the health and education sectors. Africa Region Working Paper Series No. 45. Washington, DC: World Bank, 2003. www.worldbank.org/afr/wps/index.htm Foster, Mick. MDG Oriented Poverty Reduction Strategies: Lessons from Experience in Health. Washington, DC: World Bank, 2005. http://www.mickfoster.com/docs/FosterMDGStrategiesFinal.pdf Government of Rwanda, Ministry of Health. Rwanda National Health Accounts 2004. Available at www.phrplus.com Harries, Anthony D, Erik J Schouten, and Edwin Libamba. Scaling up antiretroviral treatment in resource-poor settings. The Lancet 2006; 367; 1870-1872. www.thelancet.com/journals/lancet/article/PIIS0140673606688090/fulltext Hecht R, Adeyi O, Semini I. Making AIDS part of the global development agenda. Finance & Development, Washington, DC: International Monetary Fund, 2002. Guthrie T, Hickey A (Eds.). Funding the Fight: Budgeting For HIV/AIDS in Developing Countries. Cape Town, South Africa: IDASA, 2004. Keith-Brown K. Investing for life: making the link between public spending and the reduction of maternal mortality. Mexico D.F.: Fundar, Centro de Análisis e Investigación, 2005. http://www.internationalbudget.org/ How Parliamentarians Can Help Ensure Accountability 25 Lewis M (a). A war chest for fighting HIV/AIDS. Finance & Development. Washington, DC: International Monetary Fund, December 2005. Lewis M (b). Tackling healthcare corruption and governance woes in developing countries. CGD Brief. Washington, DC: Center for Global Development, 2006. www.cgdev.org Levine, Ruth. New HIV/AIDS Buffer Stock Service for Africa. Global health policy blog, October 11, 2006. www.cgdev.org Lindelow, Magnus, Inna Kushnarova and Kai Kaiser. Measuring corruption in the health sector: what we can learn from public expenditure tracking and service delivery surveys in developing countries. Pages 29-36, Transparency International, Global Corruption Report 2006, Berlin: Transparency International, 2006. http://www.transparency.org/publications/gcr National Democratic Institute for International Affairs (NDI) and the Southern African Development Community Parliamentary Forum (SADCPF). Survey of legislative efforts to combat HIV/AIDS in the Southern Africa Development Community region. November 2004. http://www.ndi.org/ndi/library/1795_sa_hiv_110104.txt Ndlovu, Nhlanhla. HIV and AIDS Allocations: A First Look at Budget 2005. IDASA, Budget Brief No. 152, 2005. http://www.internationalbudget.org/resources/howto/HIV.pdf Over, Mead and Martha Ainsworth. Confronting AIDS: Public Priorities in a Global Epidemic. Washington, DC: Oxford University Press, 1997. www.worldbank.org/aids- econ/confront/ Over, Mead. Coping with the impact of AIDS. Finance & Development. Washington, DC: International Monetary Fund, March 1998. www.imf.org/external/pubs/ft/fandd/1998/03/pdf/over.pdf Over, Mead, Peter Heywood, Julian Gold, Indrani Gupta, Subhash Hira, and Elliot Marseille. HIV/AIDS treatment and prevention in India: modeling the cost and consequences. Washington, DC: World Bank, 2004. www.worldbank.org/servlet/WDS_IBank_Servlet?pcont=details&eid=000090341_20040 921112225 PHRPlus. Using the National Health Accounts Subanalysis to Track Resource Flows for HIV/AIDS. NHA Global Policy Brief. Bethesda, MD: Abt Associates, June 2005. http://www.phrplus.org/Pubs/sp16.pdf 26 Glassman PHRPlus Project. AIDSTREATCOST (ATC) model software to help determine costs of ARV programs, and user manual, developed by PHRplus. PHRPlus project. System-wide effects of the Global Fund. http://www.phrplus.org/swef.php Reinikka, Ritva and Jakob Svensson. Survey techniques to measure and explain corruption. Washington, DC: World Bank, 2006 http://web.worldbank.org/WBSITE/EXTERNAL/TOPICS/EXTPUBLICSECTORANDG OVERNANCE/EXTPUBLICFINANCE/0,,contentMDK:20235447~pagePK:148956~piP K:216618~theSitePK:1339564,00.html Savedoff, William. How much should countries spend on health? FER/EIP Discussion Paper Number 2. Geneva, Switzerland: World Health Organization, 2003. http://www.who.int/health_financing/en/how_much_should_dp_03_2.pdf Schneider P, Bhatt P. Linking Indicators from National Health Accounts and the NHA HIV/AIDS Subanalysis to Health Policy Goals. Bethesda, MD: PHRplus 2004 Scorraille, Guy de, Jacquie Nachtigal, Jean Munyenpenda. Revue des Depenses Publiques dans le Secteur de la Sante 2002-2005. Republique de Rwanda, Ministere de la Sante, March 2006. Smith O. Human resource requirements for scaling up antiretroviral therapy in low- resource countries (Appendix E). In: Curran J, Debas H, Arya M, Kelley P, Knobler S, Pray L, eds. Scaling up treatment for the global AIDS pandemic: challenges and opportunities. Washington, DC, National Academies Press (Board of Global health), 2004. Soeters, Robert, Jean Perrot, Etienne Sekaganda, Antonio Lozito, and Laurent Musango. Improving health services delivery through performance-based financing in Rwanda: what are the factors associated with success? Submitted to Health Policy and Planning, 2006. Stapenhurst, F. C. and R. Pelizzo (2002). `A Bigger Role for Legislatures.' Finance and Development 39(4): 46-48. http://www.imf.org/external/pubs/ft/fandd/2002/12/stapenhu.htm Tayler Liz and Clare Dickinson (2006). "The link between corruption and HIV/AIDS" in Transparency International, Global Corruption Report 2006, Berlin: Transparency International, 104-111. http://www.transparency.org/publications/gcr UNAIDS. Resource needs for an expanded response to AIDS in low and middle-income countries. UNAIDS, August 2005. How Parliamentarians Can Help Ensure Accountability 27 http://data.unaids.org/publications/irc-pub06/resourceneedsreport_en.pdf UNAIDS. UNGASS indicators. http://data.unaids.org/publications/irc-pub06/jc1126- constrcoreindic-ungass_en.pdf UNAIDS. National Spending For AIDS 2004. Geneva, Switzerland: UNAIDS 2004. Wehner, Joachim. Effective financial scrutiny: the role of parliament in public finance. Washington, DC: The World Bank, 2004. World Bank (1998). Public Expenditure Management Handbook. Washington, D.C., World Bank. http://www1.worldbank.org/publicsector/pe/english.htm World Bank. World Development Report: Making Services Work for Poor People. Washington, DC: Oxford University Press, 2004. http://publications.worldbank.org/ecommerce/catalog/product?item_id=1764996 World Bank. Global Monitoring Report 2006: Strengthening Mutual Accountability ­ Aid, Trade and Governance. Washington, DC: Oxford University Press and the World Bank, 2006. http://web.worldbank.org/WBSITE/EXTERNAL/EXTDEC/EXTGLOBALMONITOR/E XTGLOBALMONITOR2006/0,,menuPK:2186472~pagePK:64218926~piPK:64218953 ~theSitePK:2186432,00.html 28 Glassman Annex: Civil Society Organizations that Watchdog Budgets in Africa (source: http://www.internationalbudget.org/groups/) Cameroon: www.eitdr.org www.cameroonbudgets.org www.e-citizens.net www.vs-aprm.net Ghana: http://www.isodec.org.gh Malawi: Malawi Economic Justice Network mejn@sdnp.org.mw Civil Society Coalition for Quality Basic Education (CSCQBE) cscqbe@sdnp.org.mw Namibia: Institute for Public Policy Research www.ippr.org.na Namibian NGO Forum (NANGOF) nangof@iafrica.com.na Namibia Chamber of Commerce and Industries (NCCI) http://goss.rho.net/ncci/indexp.htm Nigeria: Integrity www.theconvention.org Social & Economic Rights Action Center (SERAC) serac@linkserve.com.ng; serac@hyperia.com; seracnig@aol.com South Africa Institute for Democracy in South Africa (IDASA)'s Budget Information Service (BIS) shun@idasact.org.za How Parliamentarians Can Help Ensure Accountability 29 SA National NGO Coalition (SANGOCO) http://www.sangoco.org.za Public Service Accountability Monitor (PSAM) http://www.psam.org.za http://www.myrights.org.za Tanzania Hakikazi Catalyst www.hakikazi.org Tanzania Gender Networking Programme (TGNP) http://www.tgnp.org Action Aid admin@actionaid.org