PUBLIC HEALTH EXPENDITURE FOR UNIVERSAL HEALTH COVERAGE IN GHANA DISCUSSION PAPER AUGUST 2023 Enoch Oti Agyekum Elisha Ngetich Owen Smith Soazic Elise Wange Sonne Dorothee Chen Paul Andres Corral Rodas PUBLIC HEALTH EXPENDITURE FOR UNIVERSAL HEALTH COVERAGE IN GHANA Enoch Oti Agyekum, Elisha Ngetich, Owen Smith, Soazic Elise Wang Sonne, Dorothee Chen, and Paul Andres Corral Rodas August 2023 Health, Nutrition, and Population (HNP) Discussion Paper This series is produced by the Health, Nutrition, and Population Global Practice of the World Bank. The papers in this series aim to provide a vehicle for publishing preliminary results on HNP topics to encourage discussion and debate. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or to the countries they represent. Citation and the use of the material presented in this series should take into account this provisional character. 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. For information regarding the HNP Discussion Paper Series, please contact the Editor, Jung-Hwan Choi at jchoi@worldbank.org or Erika Yanick at eyanick@worldbank.org. RIGHTS AND PERMISSIONS The material in this work is subject to copyright. Because the World Bank encourages the dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street, NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. © 2024 The International Bank for Reconstruction and Development / The World Bank 1818 H Street, NW, Washington, DC 20433 All rights reserved. ii Health, Nutrition and Population (HNP) Discussion Paper Public Health Expenditure for Universal Health Coverage in Ghana Enoch Oti Agyekum,a Elisha Ngetich,a Owen Smith,a Soazic Elise Wang Sonne,b Dorothee Chen,c and Paul Andres Corral Rodasd a Health, Nutrition, and Population, World Bank, Accra, Ghana. b Health, Nutrition, and Population, World Bank, Washington DC, USA. c Health, Nutrition, and Population, World Bank, Paris, France d Poverty and Equity, World Bank, Accra, Ghana. Ghana Abstract: Ghana has made substantial progress toward universal health coverage (UHC) and compares relatively well in health outcomes among its economic peers. International comparisons of government current health expenditure suggest that Ghana spends less than the sub-Saharan African norm for its income level. That notwithstanding, Ghana compares favorably with its peers in key health outcomes with a mixed pattern of equity in health service utilization. The trajectory indicates that better health outcomes could be achieved if health spending increases. Public funding through the Ministry of Health’s (MoH’s) budget and the National Health Insurance Scheme (NHIS) constitute the major sources of financing for the health sector in Ghana.. Total government health spending in Ghana almost trippled in nominal terms in 2015-20, but it rose by a modest 12 percent in real terms, and nearly all the increased spending was channeled through MoH instead of through NHIA. On average, 80 percent of the government budget is channeled through the MoH budget, while the share of NHIS spending has declined over the same period from 24 percent to 12 percent. There are preliminary signs that past achievements in financial protection largely attributable to the impact of the NHIS are being eroded. Although the social contract underpinning the establishment of the National Health Insurance Levy is not in question, the recently introduced Statutory Funds Capping Law further threatens the sustainability of the NHIS. There is a need to improve efficiency in the functional allocation of domestic health resources and upscale the use of public financial management (PFM) tools at the national and subnational levels to enhance budget credibility and expenditure tracking. To ensure no one is left behind, deliberate efforts should be made to improve equity in health care utilization and outcomes along income and regional dimensions. Keywords: Public expenditure, universal health coverage, financial protection, access, primary health care, Ghana Disclaimer: The findings, interpretations, and conclusions expressed in the paper are entirely those of the authors, and do not represent the views of the World Bank, its Executive Directors, or the countries they represent. Correspondence Details: Enoch Oti Agyekum, World Bank (Accra, Ghana), +233596912144; eotiagyekum@worldbank.org; https://www.worldbank.org. iii Table of Contents RIGHTS AND PERMISSIONS .................................................................................................. II PUBLIC HEALTH EXPENDITURE FOR UNIVERSAL HEALTH COVERAGE: THE ROLE OF GHANA’S NHIS........................................................................................................................................... III ACKNOWLEDGMENTS ............................................................................................................ 4 INTRODUCTION......................................................................................................................... 1 1.2. COUNTRY AND SECTOR CONTEXT ......................................................................................... 7 1.2.1. Macroeconomic Context ................................................................................................ 7 1.2.2. Health Sector Context .................................................................................................. 11 2. KEY HEALTH SECTOR PERFORMANCE INDICATORS ........................................... 13 2.1. DEMOGRAPHIC AND EPIDEMIOLOGICAL TRANSITION .......................................................... 13 2.2. HEALTH OUTCOMES ............................................................................................................ 15 2.3. BURDEN OF DISEASE ........................................................................................................... 20 2.4 PUBLIC HEALTH EMERGENCY PREPAREDNESS AND RESPONSE ............................................ 22 3. HEALTH FINANCING ......................................................................................................... 25 3.1 OVERVIEW OF HEALTH FINANCING IN GHANA ..................................................................... 25 3.2. PUBLIC HEALTH FINANCING SOURCES AND STRUCTURE ..................................................... 27 3.2.1. Ministry of Health Budget ........................................................................................... 29 3.2.2. The National Health Insurance Scheme ...................................................................... 30 3.3. DEVELOPMENT ASSISTANCE FOR HEALTH .......................................................................... 45 3.4 FINANCING VACCINES AND ESSENTIAL MEDICINES ............................................................. 47 3.5 PUBLIC FINANCIAL MANAGEMENT IN THE HEALTH SECTOR ................................................ 48 4. HEALTH SERVICES UTILIZATION ................................................................................ 52 5. FINANCIAL PROTECTION AND EQUITY ..................................................................... 58 7.1 CONCLUSIONS ...................................................................................................................... 63 7.2 RECOMMENDATIONS ............................................................................................................ 65 7.3 LIMITATIONS ........................................................................................................................ 66 8.0 ANNEX .................................................................................................................................. 68 REFERENCES............................................................................................................................ 69 iv Figures FIGURE 1: GHANA'S COMPARATIVE PERFORMANCE ON KEY FISCAL INDICATORS AS A PERCENTAGE OF GDP, 2021...............................8 FIGURE 2: COMPARISON OF SUB-SAHARAN AFRICAN COUNTRIES TOTAL REVENUE (EXCLUDING GRANTS) AS A PERCENTAGE OF GDP, 2019 .....................................................................................................................................................................9 FIGURE 3: POVERTY INCIDENCE IN GHANA BY GEOGRAPHIC LOCATION, 2005–2016....................................................................10 FIGURE 4: GHANA'S HEALTH SYSTEM ORGANIZATIONAL STRUCTURE ..........................................................................................12 FIGURE 5: GHANA'S POPULATION GROWTH TREND AND PROJECTION AS A PERCENTAGE, 2010–2050 ............................................13 FIGURE 6: GHANA'S TOTAL FERTILITY RATE BY REGION ............................................................................................................14 FIGURE 7: GHANA'S PROJECTED POPULATION SHIFT, 2020–2050 ............................................................................................15 FIGURE 8: TRENDS AND PROJECTIONS: GHANA'S CHILD MORTALITY RATES, 1993–2030 ..............................................................16 FIGURE 9: GHANA'S POSITION ON KEY HEALTH OUTCOME INDICATORS ......................................................................................16 FIGURE 10: COMPARISON OF U5MR AND STUNTING, BY REGION, GENDER, AND INCOME .............................................................18 FIGURE 11: CONCENTRATION INDEX TRENDS FOR STUNTING, U5MR, AND KEY HEALTH TREATMENTS .............................................20 FIGURE 12: GHANA'S CAUSE OF DEATH COMPARISON, 2009–2019 .........................................................................................21 FIGURE 13: COMPARISON OF FIVE PRIMARY CAUSES OF DEATH IN GHANA, 2019 TO 2040 ...........................................................21 FIGURE 14: PREVALENCE OF HIV IN GHANA, PERCENTAGE OF POPULATION AGES 15–49 ..............................................................22 FIGURE 15: GHANA'S HEALTH EXPENDITURES BY FUNDING SOURCE, 2014–2020 .......................................................................25 FIGURE 16: COMPARISON OF GHANA'S HEALTH SPENDING COMPARED WITH SSA PEERS AND THE LMIC AVERAGE ............................26 FIGURE 17: GHANA'S PUBLIC SECTOR HEALTH FUNDING FLOW .................................................................................................27 FIGURE 18: GHANA'S TRENDS IN REAL AND NOMINAL PUBLIC HEALTH EXPENDITURE IN MILLION, GH¢ ...........................................28 FIGURE 19: GHANA'S MOH VS. NHIS PUBLIC HEALTH SPENDING, 2015–2023..........................................................................29 FIGURE 20: ECONOMIC COMPOSITION AND SOURCES OF GHANA'S MOH BUDGET, 2015–2022 ....................................................29 FIGURE 21: MOH EXPENDITURE BY FUNDING SOURCE ............................................................................................................30 FIGURE 22: NHIA FINANCING GAPS, 2016–2023 ................................................................................................................31 FIGURE 23: UNEVEN IMPACT OF CAPPING ON STATUTORY FUND BUDGETS, 2022–2023 ..............................................................32 FIGURE 24: NHIS FINANCIAL PERFORMANCE EXCLUDING ARREARS, 2008–2021 .........................................................................33 FIGURE 25: NHIS FINANCIAL PERFORMANCE INCLUDING ARREARS, 2008–2021 ........................................................................33 FIGURE 26: NUMBER OF MONTHS OF NHIS CLAIMS REIMBURSEMENT DELAYS, 2017–2020 ........................................................34 FIGURE 27: PERCENTAGE CHANGE IN POVERTY GIVEN A PERCENTAGE DROP IN OUT-OF-POCKET EXPENDITURES ...............................35 FIGURE 28: GINI INDEX POINT CHANGE GIVEN A PERCENT DROP IN OUT-OF-POCKET EXPENDITURE................................................35 FIGURE 29: NUMBER OF PEOPLE LIFTED OUT OF POVERTY GIVEN AN X PERCENT DROP IN OUT-OF-POCKET EXPENDITURES .................36 FIGURE 30: COMPARISON OF OOP PAYMENTS VS. NHIS PAYMENTS AT PRIMARY HEALTH FACILITIES, 2017–2021 ..........................37 FIGURE 31: COMPARISON OF OOP PAYMENTS VS. NHIS PAYMENTS AT SECONDARY AND TERTIARY CARE FACILITIES, 2015–2021 ......38 FIGURE 32: NHIS MEMBER OOP COMPARISON, BY DECILE .....................................................................................................38 FIGURE 33: PERCENTAGE OF NHIS CLAIMS EXPENDITURE AS A SHARE OF TOTAL REVENUE.............................................................39 FIGURE 34: DISTRIBUTION OF NHIS-ACCREDITED FACILITIES IN GHANA, BY TYPE .........................................................................41 FIGURE 35: CLAIMS ADJUSTMENT RATE BY FACILITY TYPE, 2019–2021 ....................................................................................42 FIGURE 36: HEALTH INSURANCE SCHEME (ALL FORMS) COVERAGE FOR GHANA AND REGIONAL PEERS, 2021 .....................................43 FIGURE 37: NHIS ENROLLMENT RATE TRENDS, 2017–2021 ...................................................................................................43 FIGURE 38:NHIS POPULATION COVERAGE BY REGION, 2019–2021.........................................................................................44 FIGURE 39: COMPOSITION OF NHIS MEMBERSHIP, BY POPULATION SUBCATEGORIES ...................................................................45 FIGURE 40 COMPARISON OF MOH ALLOCATION AND EXPENDITURE, 2017–2022, IN MILLION GH¢ .............................................49 FIGURE 41: UHC SCI PROGRESS IN GHANA AND COMPARISON WITH REGIONAL PEERS .................................................................52 FIGURE 42: DISTRIBUTION OF GHANA'S HEALTH FACILITIES BY CATEGORY ...................................................................................53 FIGURE 43: PERCENTAGE OF HEALTH FACILITIES COMPLIANT WITH EMONC CRITERIA, BY REGION ..............................54 FIGURE 44: PERCENTAGE OF STOCKOUTS AT HEALTH FACILITIES AND COMMON CAUSES, THREE-MONTH PERIOD ..............................55 FIGURE 45: AVAILABILITY OF FUNCTIONAL EQUIPMENT IN EMONC FACILITIES .............................................................................55 FIGURE 46:MATERNAL AND CHILD CARE SPECIALISTS, ACTUAL VS. TARGET ..............................................................................56 FIGURE 47: NATIONAL DISTRIBUTION OF OUTPATIENT VISITS PER 1,000 POPULATION BY TYPE OF HEALTH FACILITY...........................56 FIGURE 48: TRENDS IN IMMUNIZATION COVERAGE FOR CHILDHOOD VACCINES, 2012–2021 ........................................................57 FIGURE 49: PERCENTAGE OF CATASTROPHIC HEALTH EXPENDITURES IN GHANA, 1991–2012 .......................................................58 FIGURE 50: COMPARISON OF CATASTROPHIC HEALTH EXPENDITURE AND OUT-OF-POCKET PAYMENTS.............................................58 FIGURE 51: UTILIZATION OF KEY CHILD HEALTH SERVICES BY WEALTH QUINTILE ..........................................................................60 FIGURE 52: HEALTH CARE UTILIZATION BY FACILITY TYPE AND WEALTH QUINTILE ........................................................................60 2 FIGURE 53: CHOICE OF HEALTH SERVICE PROVIDER BY WEALTH QUINTILE ..................................................................................61 Tables TABLE 1: OVERVIEW OF GHANA'S POVERTY RATES, 1998–2016 10 TABLE 2: CONCENTRATION INDEXES OF KEY UNDER-FIVE HEALTH VARIABLES 18 TABLE 3: COUNTRY COMPARISON OF GHANA'S HEALTH FINANCING INDICATORS, 2020 26 TABLE 4: NHIS EXPENDITURE BREAKDOWN 40 TABLE 5:NHIS CLAIMS EXPENDITURE BY FACILITY TYPE, 2021 40 TABLE 6: DEVELOPMENT ASSISTANCE ON AND OFF BUDGET COMPARISON, 2016–2018 46 TABLE 7: ROUTINE IMMUNIZATION EXPENDITURE IN GHANA, US$ 48 TABLE 8: GHANA'S HEALTH FACILITIES BY TYPE 53 TABLE 9: INCIDENCE OF CATASTROPHIC HEALTH PAYMENTS BY QUINTILE, 2017 59 TABLE 10: CONCENTRATION INDEX ON MATERNAL HEALTH OUTCOMES, BY LOCATION AND MOTHER'S EDUCATION 62 3 ACKNOWLEDGMENTS Enoch Oti Agyekum, Economist (Health), led the preparation of this paper. The paper benefited from inputs from World Bank and Global Financing Facility (GFF) teams including Owen Smith (Lead Health Economist), Elisha Kipkemoi Ngetich (Health Specialist), Soazic Elise Wang Sonne (Economist), Alexander Nartey (Consultant), Pearl Opoku Youngman (GFF Liaison Officer), Dorothee Chen (Senior Health Specialist), Akriti Mehta (Consultant), Cicely Thomas (GFF, Senior Health Economist), and Daniela Hoshino (Program Assistant). Paul Andres Corral Rodas (Senior Poverty Economist) prepared the simulation on the impact of reduced out-of-pocket payments due to increased funds for NHIS with inputs from Ambar Narayan (Lead Economist). The team would like to pay special recognition to Antonio Giuffrida (Senior Health Finance Specialist) at the Global Fund, who initiated the development of the Ghana Health Public Expenditure Review (PER), which formed the basis of the paper. The report also benefited from the insightful comments offered by its peer reviewers including Pia Schneider (Lead Economist) and Moritz Piatti (Senior Economist). The team worked closely with the government counterparts who provided data and valuable contributions. The team would like to express appreciation to Dr Bernard Okoe-Boye, Chief Executive Officer of the National Health Insurance Authority (NHIA); Ms Vivian Addo-Cobbiah, Deputy CEO, National Health Accounts (NHA); Mr. Oti Frimpong Director, Membership and Regional Operations; Dr. Yaw Opoku Boateng, Deputy Director, Claims; Mr. Daniel Adin Darko, Deputy Director Finance, NHIA; Mr. Richard Agyeman Badu, Director of Finance, NHIA; Mr. Kwakye Kontor, Head of Budget and Planning, MoH; Mr. Samson Awudanjong, Financial Reporting, MoH; Mr. Mustafa Hamidu, Director Finance, Ghana Health Service; Mr. Daniel Osei, Head of Resource Mobilization, Ghana Health Service; members of the Health Finance Technical Working Group and members of the Development Partners Health Financing Working Group. The team also collaborated with Stephen Duku, USAID, and Kingsley Addai Frimpong, WHO. Overall guidance and direction were provided by Pierre Laporte (Country Director, AWCW1); Magnus Lindelow (Practice Manager, HAWH3); Patrick Mullen (Program Leader, Human Development; and Toomas Palu (Adviser). The authors are grateful to the World Bank for publishing this report as an HNP Discussion Paper. 4 INTRODUCTION Backed by political commitment, Ghana has made substantial progress toward universal health coverage (UHC). Health outcomes are better than average among its African peers and commensurate with its (higher) income level. Equity in access to priority services is strong, financial risk protection for households has been good, and Ghana has been a trailblazer across the region in the adoption of health financing reforms. However, key bottlenecks persist that may hinder the country’s progress toward attaining its UHC goals by 2030. Additionally, Ghana needs to reorient the health system’s focus to meet the changing needs of the population as it undergoes demographic and epidemiologic transitions. The health system must also consider the dynamic economic and social landscape as it aims to meet UHC goals. With significant national economic gains over the last three decades, Ghana has experienced an overall reduction in poverty rates. This progress has stalled since post 2012, and geographical economic and health disparities have widened. The COVID-19 pandemic and subsequent economic crisis have further slowed the country’s economic growth. This paper examines the government’s health expenditure in line with its UHC policy objectives. It examines the government’s revenue-raising mechanisms and public sector expenditures for delivering effective, efficient, and equitable health services. The paper critically evaluates the performance of the National Health Insurance Scheme (NHIS). Additionally, the paper seeks to examine the flow of funds within the government system in line with the government’s public financial management (PFM) principles. The paper provides basic information on health sector performance and recommendations that will inform the Ministry of Health’s (MoH’s) health financing strategy. The strategy, currently under revision, elaborates how health financing in Ghana will contribute to the health system goals and objectives. Therefore, this paper will provide an analytical underpinning for the new health financing strategy, highlighting key issues that need to be addressed in the short, medium, and long terms. Quantitative and qualitative data were collected directly from relevant line ministries and agencies. Quantitative data sources include the Ministry of Finance’s (MoF) Budget Statement and Economic Policy, MoH budget documents, and officially requested data from the NHIS. Service delivery data were obtained from the District Health Information Management System (DHIMS2) database and representative national household surveys such as the Multiple Indicator Cluster Survey (MICS 2017/18), the Ghana Maternal Health Survey (MHS 2017), the 2019 Malaria Indicator Survey (MIS), and the 2017 Ghana Living Standard Survey (GLSS7). Global health databases used include the World Health Organization (WHO) Global Health Expenditure Database (GHED), the World Development Indicators (WDI), and the International Monetary Fund’s (IMF’s) World Economic Outlook. Despite its limitation, the GHED enabled global benchmarking of Ghana’s performance. To complement the quantitative data, qualitative data were also collected through interviews with national and subnational policy makers and program implementers. The paper was developed in close consultation with key health sector players and utilized existing analytical works that had relevant information to support the findings of this paper. Country coordination platforms such as the Health Development Partner’s Health Financing Working Group and the government-led Health Financing Technical Working Group were used to obtain stakeholder feedback on the report. Finally, the study was closely informed by key analytical works related to health systems such as the Resource Mapping and Expenditure Tracking (RMET) supported by the Global Financing Facility (GFF), the allocative efficiency analytical work Confidential supported by the Korea-World Bank Partnership Facility (KWPF), and the cross-programmatic efficiency analysis conducted by the WHO. Health financing overview: Spending under pressure, especially for nonwage recurrent categories Public funding through the Ministry of Health’s budget and the National Health Insurance Scheme (NHIS) constitute the major sources of financing for the health sector in Ghana (about 50 percent). Household out-of-pocket (OOP) payments are estimated to account for about 35 percent of total health spending, but the most recent robust evidence on OOP payments is now several years old, and there is a perception that OOP payments have been rising significantly. Development assistance for health (DAH), though relatively small, plays a significant role in the health sector financing. Over 90 percent of the government of Ghana (GoG) health revenue is usually allocated to workforce compensation, while DAH and internally generated funds (IGF) predominantly support capital expenditure and nonwage recurrent expenditures. NHIS is funded by the National Health Insurance Levy (NHIL) and the Social Security and National Insurance Trust (SSNIT). International comparisons of current health expenditure suggest that Ghana spends less than the sub-Saharan African norm for its income level, measured either as a share of gross domestic product (GDP) or as a share of the overall government budget. As of 2020, current government health spending accounted for about 2 percent of GDP and 7 to 8 percent of the overall government budget. However, these figures are higher than the lower-middle-income country average since many Asian countries in this income category spend relatively less on health. Global benchmarking can be obscured by issues of data comparability. Consistent with trends in other middle-income countries around the globe, the ongoing demographic and epidemiological transitions in Ghana, coupled with rising population expectations of the health sector, will put increasing pressure on government health spending. Total government health spending in Ghana almost doubled in nominal terms between 2015 and 2023, but it rose by a modest 12 percent in real terms, and nearly all the increased spending was channeled through MoH instead of NHIA. On average 80 percent of the government budget is channeled through the MoH budget while the share of NHIS spending has declined over the same period from 24 percent to 12 percent. This trend continued during the COVID-19 period as reflected in budget allocations (as distinct from actual expenditures which are reported with a delay). This implies that salary costs have been prioritized during the pandemic and subsequent economic downturn, while nonwage recurrent expenditure (the primary role of NHIA reimbursement) has been squeezed, with implications for an OOP burden on patients. Protecting households during the economic crisis is a key challenge. NHIA spending: Revenue shortfalls undermine progress Ghana has been at the forefront of health financing reforms in Africa since the launch of the National Health Insurance Scheme (NHIS) in 2003; after a strong start the NHIS evolution has stalled in recent years. A transition from MoH-led, supply-side financing toward a purchaser-led, demand-side approach is widely considered good practice reform in middle-income countries globally. Ghana had achieved a strong track record in this domain during the first 15 years of the NHIS, and is reflected in both robust institutional strengthening and a meaningful impact on financial risk protection among beneficiaries. The stagnation of NHIA spending in recent years is driven by two distinct financing gaps. First, a capping law, passed in 2018, limits flows to statutory funds as a share of total government 2 revenues. Initially the ceiling was set at 25 percent of revenues but was lowered to 17 percent in late 2022. As a result, by 2023 the NHIA budget allocation was 25 percent lower than in 2022, and amounted to just 54 percent of projected NHIL revenue. This undermines the implicit social contract whereby citizens pay the levy on the understanding that it will finance health insurance. While the logic of limiting the use of statutory funds to protect the preeminent role of ministries, departments, and agencies (MDAs) may be fully justifiable in other sectors, it makes less sense in the case of NHIA since this institutional model (a so-called purchaser-provider split with MoH) is consistent with good practice health financing reforms in other middle-income countries. Despite this, statutory funds in other sectors have not experienced a sharp decline in 2023 budgets, as was the case for NHIA. The second shortfall is because actual cash received from MoF is significantly less than the original budgeted amount. During the 2020–2022 period, transfers amounted to between 50 to 60 percent of the original budget. Taken together, the two gaps in NHIA funding are significantly affecting the scheme’s ability to reimburse providers in a timely manner, threatening the effectiveness of the scheme by incentivizing informal payments and ultimately imposing a financial burden on households. External financing shortfalls combined with other pressures on the NHIS are manifesting in significant balance sheet distortions. First, claims outlays as a share of revenues have declined from 85 percent in 2009 to 47 percent in 2021, suggesting a low prioritization of claims expenditure (the core mandate of NHIA) vis-à-vis other expenses. Among these other spending items are a transfer of funds to MoH equivalent to over 10 percent of total NHIA outlays (2020) for preventive care and support to district offices equivalent to about 2.5 percent of its budget. Second, claims arrears to providers now exceed GH¢ 1 billion. Taken together—a diversion of NHIA funds for other purposes while claims arrears accumulate—suggests a worrying pattern whereby NHIA is departing from its main purpose of ensuring financial protection to households. On a more positive note, NHIA is also making major investments in its IT infrastructure that should enable more strategic purchasing in the future. As and when NHIA’s revenues and balance sheet are stabilized and reoriented toward its core mandate of claim reimbursement, there is scope to strengthen its role as a strategic purchaser. The NHIS reimburses public and private health providers for variable costs of direct patient care using fee-for-service (FFS) for drugs, and diagnostic-related groups (DRGs) for inpatient and outpatient services. NHIS claims payment is concentrated at the district hospital level with only 17 percent going to the subdistrict and below. About 40 percent of outpatient services take place at these levels; and subdistrict facilities make up 82 percent of all accredited NHIS facilities, which highlights the need for improvements in claims payment for these critical lower-level health providers. Furthermore, the NHIS claims rejection rate is highest among frontline primary health care facilities compared to other facility types, which could be attributed to low capacity to correctly process claims and/or submission of claims for services above the prescribed level for the facility. More broadly, the ongoing transition to electronic claims payment will create opportunities to significantly improve the efficiency and quality of service delivery including through data analytics, fraud control, and medical audit interventions. A related priority will be to update the tariff schedule to more accurately reflect actual costs, as the current schedule is outdated. Lastly, there is a need to ensure that cost-effective preventive services including noncommunicable diseases (NCD) screening are adequately covered in the NHIA package. While NHIS enrollment of over 50 percent of Ghana’s population is the second highest in Africa, there is an opportunity to enhance coverage of poor and vulnerable groups going forward. Absolute numbers of NHIS enrollment have increased across membership categories, but there 3 is marked regional disparity and coverage for indigents, and free maternal care (FMC) has declined. The observed decline is likely due to challenges in registering new entrants and renewing memberships of the existing enrolled indigent population. Although the National Household Registry intends to collect and maintain a listing of the poor and vulnerable population, it lacks nationwide coverage and effective linkage to the NHIS database. As the registry is improved, it will provide an ideal platform to enhance coverage of disadvantaged socioeconomic groups, such as through the selective exemption of premium payments. To determine the potential impact on poverty of closing the financing gap in NHIS spending, different percentage reductions in out-of-pocket spending by registered households are assumed. This then translates to a purchasing power gain for NHIS-registered households who have reported out-of-pocket spending in the relevant expenditure categories covered by NHIS. For the analysis, the decile specific purchasing power gain is applied to all registered households in the decile. Specifically, the average purchasing power gain among those who reported out-of- pocket expenditures is used for all those registered. This is done to account for the fact that health expenditures are sporadic and not necessarily a fixed occurrence. The assumption is not without consequence: Health expenditures are sporadic, and this assumes that every registered individual in the decile would benefit similarly regardless of whether they had out-of-pocket expenses or not. The assumption simplifies the analysis considerably. Additionally, cheaper health care does not necessarily translate to a higher propensity to spend on health care. Finally, the poverty impact ignores the long-run benefits of a potentially healthier population. Through improved health outcomes the population can expect to become more productive, which will likely translate into larger incomes and potentially larger tax revenues in the future. A 10 percent reduction in out-of-pocket expenditures corresponds to 0.1 percentage point drop in absolute poverty and a 0.14 percentage point drop in extreme poverty. Improvements in inequality are modest, but also expected. Ministry of Health spending: Little room for discretion as wage bill predominates Despite a shift from activity-based to program-based budgeting, the MoH budget remains largely input- or line-item–based, providing little flexibility for discretionary spending. Additionally, the process of budget planning and allocation for the NHIS runs parallel to that of the MoH, resulting in possible inefficiencies. For example, as noted, while MoH spending (mainly for salaries) has increased significantly in nominal terms in recent years, NHIA spending (mainly for nonwage recurrent items) has been comparatively flat. This divergence represents a potentially large inefficiency in functional allocations. Human resource costs dominate the MoH budget, taking up about 90 percent of the domestic funds allocated to the sector and about 60 percent of the total health budget. In addition to the salary and compensation component of the MoH budget, health facilities spend about 15 percent of their IGF to pay contractual staff salaries and other compensation-related expenditures. Ghana has made some progress in scaling skilled human resources for health (HRH), but significant inequities in geographic distribution persist. Despite the rising trend of health workers’ density over the past five years, the health workforce remains insufficient to fulfill UHC and Sustainable Development Goal (SDG) targets. There are 3.79 medical staff (physicians, nurses, and midwives) per 1,000 people in Ghana, about 15 percent below the 4.45 SDG index threshold. However, this density is higher than the WHO threshold of 2.5 medical staff per 1,000 people needed to provide adequate coverage of primary health care (PHC) interventions. 4 Since 2017, the number of health care facilities in the country has increased, but there are gaps in the quality of care being delivered. For example, preliminary results from the Emergency Obstetric and Newborn Care (EmONC) survey show that a significant proportion of health facilities do not meet the required quality standards to deliver EmONC. The survey shows a wide disparity in the availability of EmONC service across regions, with some regions reporting no facilities that meet the required EmONC standards. In most cases, particularly in primary health care facilities, essential equipment is unavailable or nonfunctional. According to the EmONC report, only a third of EmONC facilities have functional equipment available. The gap in health care worker availability is more prominent for specialist doctors than general nurses and midwives. Institutional data on distribution of health services indicate high utilization of the primary health care level for inpatient and outpatient department (OPD) services. Ensuring a universal essential health services package with a focus on preventive and promotive services embedded in primary care requires significant policy attention. The government has shown commitment to providing essential health services to the population in the past years, but key challenges persist in the lead-up to donor transition, particularly for the financing of cost- effective interventions such as immunization. The NHIS and development partners have become Ghana's primary source of financing for essential medicines as the Ministry of Health’s budget is not sufficient to finance the procurement of vaccines and essential health commodities. There is an urgent need for policy dialogue on the financing of vaccine and essential medicines as major donors phase out their support and the NHIS refocuses attention to reimbursing providers. Public financial management: Overcoming inefficiencies Ghana has made consistent efforts toward improving public PFM through legal and policy reforms, but implementation at the level of ministries, departments, and agencies (MDAs) could be strengthened. In the health sector, weak PFM practices can result in planning and budgeting inefficiencies, weak expenditure controls, and lack of transparency. Available MoH budget and expenditure data from 2017 to 2022, indicate that actual expenditures of the MoH consistently exceed allocated budget. While this suggests that the MoH has the capacity to execute its Annual Program of Work, it also raises concern about the accuracy and comprehensiveness of the planning and budgeting process. Additionally, quasi-government health institutions that provide a significant quantity of health services to the population are consistently unaccounted for in the health sector budget and expenditure reports. The MoH has taken steps to integrate these institutions into the health sector planning process through a representative body, the Ghana Association of Quasi-Government Health Institution. However, in the absence of a well-defined framework to collect and organize budget and expenditure data from these institutions, government expenditure on health is potentially underestimated. Ghana has continuously improved its District Health Information Management System (DHIMS2) and has introduced innovative information management tools. However, these tools function in silos with little to no integration. Existing health policies aim to improve service delivery at the subdistrict level but fall short of addressing gaps in PFM capacity at that level. The development of the Networks of Practice (NoP) requires a strong hub and, in this case, strong health centers. However, over the years, health centers have received limited support, rendering them ineffective in acting as a hub for delivery and management of health care services at the subdistrict level. Thus, currently, the district management team has to support and handhold the subdistrict level to perform fiduciary functions. Additionally, health facility managers and district health management teams (DHMTs) have limited discretionary powers and funds to efficiently allocate resources for service delivery. 5 This results in gaps in the service availability and readiness of the health facilities to provide quality care. Forthcoming initiatives to strengthen NoPs offer significant promise. Development assistance for health: Preparing for the transition. Development assistance for health (DAH) in Ghana represented 11.0 percent of current health expenditure (CHE) in 2018, slightly lower than the SSA average, but can be expected to decline toward the lower-middle-income country (LMIC) average of 2.8 percent over time. Despite this outlook, DAH continues to fill critical gaps in financing capital investment and operational costs in the health sector, particularly at the subnational level. Donor funding prioritizes preventive services such as immunization and malaria control programs and essential commodities. The long-term sustainability of essential health programs should therefore be a priority. Development partners also provided substantial resources to manage the COVID-19 pandemic in the country, by rechanneling support from other sectors and providing additional funds. Beyond DAH levels, there is a lack of coordination in the modality of transfers and execution of earmarked funds from many donor agencies. While some partners transfer funds to MoF with specific instructions on their use, others transfer funds directly to MoH, Ghana Health Service (GHS), or other health sector agencies. Estimates indicate that from 2016 to 2018, between 61 and 73 percent of total DAH was not recorded in the MoF budget. In 2019, Ghana introduced its “Beyond Aid” agenda that outlined a strategy to reduce its dependence on external aid across sectors and charted out steps for implementation of the strategy. Looking ahead, Ghana must plan for eventual graduation from key global health financing sources such as GAVI (expected in 2029), with a focus on ensuring sustainability of both financing and implementation modalities. Financial protection and equity: Past achievements at risk in a context of economic crisis In the prepandemic, precrisis environment, Ghana’s health system appeared to be effective in preventing catastrophic health expenditures, as few households were allocating more than 10 percent of their total spending to health. Over the 1991–2012 period, less than 1 percent and 7 percent of households were allocating more than 25 percent and 10 percent of their household income to health, respectively. This level of financial protection is remarkable for a country like Ghana where OOP payments still represent close to 40 percent of CHE. Irrespective of the budget threshold, the households that exceed the catastrophic payments threshold tend to belong to lower socioeconomic quintiles. There are preliminary signs that past financial protection achievements are being eroded. The global “Findex” survey found a significant increase (from 10 percent to 16 percent) in the share of Ghanaian households reporting that they had borrowed money to pay for medical care between 2017 and 2021, especially among the poor. Despite the expansion in NHIS enrollment, households continue to pay OOP for health care services. The share of OOP payments in health facilities’ total revenue is increasing. Additionally, OOP has been reported among members and nonmembers of NHIS for services received at health facilities. Notably, 42 percent of insured clients belonging to the poorest quintile report paying OOP for services. Additionally, informal copayments appear to be more common in urban regions. Ghana has a mixed pattern of equity in health care utilization, with achievements in some areas but challenges elsewhere. For key child health services such as immunization, vitamin A provision, and care-seeking by mothers of sick kids, there is remarkable equity across quintiles. However, there are significant inequalities in the prevalence of stunting, under-five mortality, zero- dose children, and in broader patterns of health care utilization, with better-off households disproportionately visiting hospitals while the poor and vulnerable attend subdistrict facilities. 6 1.2. COUNTRY AND SECTOR CONTEXT 1.2.1. Macroeconomic Context The Republic of Ghana is located in West Africa, covering approximately 240,000 square kilometers (sq. km), and is bordered by Burkina Faso, Togo, Cote d’Ivoire, and the Atlantic Ocean. Ghana is divided into 16 administrative regions, six of which were created in 2019. It is a multiparty parliamentary democracy with national elections held every four years. The average life expectancy at birth is 62.7 years, while the population density is about 133 per square kilometer. Since 2000, Ghana has been one of the strongest economic performers in SSA, despite volatile growth and relatively weak macro fundamentals. Between 2005 and 2019, annual GDP and GDP per capita growth averaged 6.6 percent and 4.1 percent, respectively—well above the average GDP growth rate for SSA (2 percent). However, GDP growth—propped by commodities—has historically fluctuated with changes to global commodity prices and market shocks. Over the past decade, the discovery of oil and the start of commercial production, the expansion of mining, and efficiency gains in export crops have increased the dependence of Ghana’s growth on commodities, heightening external vulnerabilities. Moreover, the government has struggled to implement effective fiscal policies to mitigate the growing economic volatility and achieve fiscal deficit targets. The result has been a spate of revenue shortfalls and expenditure overruns resulting in fiscal slippages. Since 2007, Ghana has relied extensively on expensive commercial borrowing (Eurobonds), progressively driving debt back to unsustainable levels. Against the backdrop of Ghana’s chronically weak macroeconomic framework, the shocks from the pandemic and Russia-Ukraine war led to a fast and deep deterioration of macroeconomic fundamentals between 2020 and 2022, compared to its structural and aspirational peers. In 2020, Ghana ran a large fiscal deficit (15.2 percent of GDP) reflecting revenue shortfalls and higher expenditures and the COVID-19-induced growth slowdown (0.5 percent in 2020 from 6.5 percent in 2019). Consequently, public debt reached 79 percent of GDP. In 2021, Ghana began to emerge from the crisis with stronger economic growth; however, the deficit remained high (reflecting low domestic resource mobilization) and the government issued a three billion Eurobond in March 2021, before progressively increasing direct monetary financing of the deficit after the country had effectively lost market access in the second half of 2021. The first half of 2022 saw two major inflexion points. The first was a sovereign credit downgrade reflecting the country’s debt vulnerabilities and the tightening of international financing conditions. The second was the start of Russia invasion of Ukraine, which slowed down the global economy and triggered large spikes in international food and fuel prices. These shocks compounded Ghana’s existing weaknesses and set in motion an unprecedented deterioration in macroeconomic stability. Inflation has increased steadily and exceeded 40 percent year-on-year in 2021-22 while the Ghana cedi has depreciated by over 50 percent against the US dollar so far in 2022. In response, the Central Bank has tightened monetary policy by incrementally raising the monetary interest rate to 24.5 percent and setting primary reserve requirements for banks to above 15 percent. The debt stock is projected to hit approximately 104.6 percent of GDP in 2022, and hover around 100 percent of GDP between 2023 and 2024. The Figure 1 below demonstrates how Ghana is lagging behind aspirational, regional, and structural peers.The figure illustrates Ghana’s comparative performance on key fiscal indicators as a percentage of GDP in the 2021 period. 7 Figure 1: Ghana's Comparative Performance on Key Fiscal Indicators as a Percentage of GDP, 2021 Source: Data from MTI benchmarking tool. Low domestic revenue mobilization and off-the-mark revenue projections remain cardinal drivers of fiscal vulnerabilities and are two areas ripe for reform. Ghana’s tax-to-GDP ratio lags behind that of structural and aspirational peers and is below the average for sub-Saharan Africa (see Figure 2 below ). Over the years, the government has repeatedly fallen short of revenue targets, even after revision. Improving revenue projection accuracy is crucial to anchoring spending, creating the right amount of fiscal space for pro-growth spending, and maintaining a credible budget process. For instance, the government’s gross overprojection of revenue from the E-levy, a 1.5 percent tax on electronic transactions, helped to garner support for the initiative. Unfortunately, the miscalculation put the budget at risk of a substantial revenue shortfall as receipts from the E-levy were substantially lower than expected. 8 Figure 2: Comparison of Sub-Saharan African Countries Total Revenue (Excluding Grants) as a Percentage of GDP, 2019 Source: IMF 2021, Article IV. Note: SSA = Sub-Saharan Africa. Prior to 2012, large economic gains resulted in reduced poverty rates. According to the most recent available data, between 2005 and 2012, the poverty rate dropped from 31.9 percent to 24.2 percent (see Table 1), which moved Ghana to the forefront of poverty reduction in Africa and indicated it was on track to achieving the Millennium Development Goal (MDG) of eradicating extreme poverty. However, despite continued economic growth between 2012 and 2016, the incidence of poverty only marginally decreased from 24.2 percent to 23.4 percent, while regional and geographic disparities widened. The poverty gap increased from 7.7 percent to 8.4 percent between 2012 and 2016, eroding some of the gains made between 1998 and 2012, when geographical disparity decreased from 13.8 percent to 7.7 percent. There are also geographic inequalities in the incidence of poverty. For example, poverty rates have worsened over time in rural Savanna while the highly urbanized areas have experienced a reduction in poverty between 2005 and 2016 (Figure 3). 9 Table 1: Overview of Ghana's Poverty Rates, 1998–2016 Year Poverty rates Poverty gap Severity of poverty 1998 39.5 13.8 6.5 2005 31.9 11 5.4 2012 24.2 7.7 3.5 2016 23.4 8.4 4.3 Source: Ghana Poverty Assessment 2020. Figure 3: Poverty Incidence in Ghana by Geographic Location, 2005–2016 Source: Ghana Living Standard Report 2017. The COVID-19 pandemic slowed the country’s economic growth. Economic growth slowed to 0.5 percent in 2020, improving to 5.4 percent in 2021. The proportion of the population living on less than US$1.90 per day is estimated to continue declining to 9.2 percent by 2023. In 2020, total public debt rose to 78.9 percent of GDP while the current account deficit widened slightly to 3.3 percent of GDP. With fiscal deficit rising to 15.2 percent of GDP in 2020, the country is estimated to be at high risk of debt distress with a narrow path to debt sustainability. At 0.45, Ghana’s Human Capital Index 1 (HCI) is higher than the sub-Saharan Africa average but falls below that of the average middle-income countries. The HCI measures the amount of human capital that a child born today can expect to attain by the age of 18 years. Ghana’s HCI 1 World Bank 2018. http://www.worldbank.org/en/publication/human-capital. 10 of 0.45 means that a child born in Ghana today can expect to be 45 percent as productive when s/he grows up, as s/he could have been if s/he had had complete education and full health. While Ghana’s health indicators suggest significant room for improvement, the country performs better than the sub-Saharan Africa average, with 95 out of 100 children born in Ghana surviving to age five, and 19 out of 100 children stunted. Seventy-six percent of 15-year-olds are expected to survive until age 60. However, out of the 11.6 average years of schooling in Ghana, the number of quality-adjusted learning years is just 5.7, implying that children are attending school for an average of nearly 12 years but not learning for nearly half of those years. 1.2.2. Health Sector Context Universal health coverage (UHC) is central to Ghana’s national health policy agenda. UHC refers to the ability of all people to receive quality services when needed, without being exposed to financial hardship (Etienne, Asamoa-Baah, and Evans 2010). UHC is one of the targets for Sustainable Development Goal (SDG) 3 and is monitored by two indicators: (i) timely access to quality care; and (ii) financial risk protection. 2 The key strategies adopted by the government of Ghana to foster UHC are the Community-Based Health Planning and Service (CHPS) strategy to bring primary health care (PHC) services close to the community and the NHIS to improve financial access to health services. The health sector’s development objectives are guided mainly by the 2019–30 National Health Policy, the Universal Health Coverage Roadmap, the Health Sector Medium Term Development Plan (2022–2025), and some specific health and health system policies and strategies. The 2019 National Health Policy, under the overarching goal to promote, restore, and maintain good health for all people living in Ghana identifies strategies to (i) strengthen the health care delivery system to be resilient; (ii) encourage the adoption of healthy lifestyles; (iii) improve the physical environment; (iv) improve the socioeconomic status of the population; and (v) ensure sustainable financing for health. In January 2020, the Ministry of Health released the Ghana Roadmap for Attaining Universal Health Coverage, which articulates a set of priority interventions to achieve UHC by year 2030. Ghana’s UHC Roadmap (MOH 2020) articulates a set of priority interventions to achieve UHC by year 2030 through the following health policy objectives: (i) Enable universal access to a better, efficiently managed, high-quality primary health system; (ii) reduce unnecessary maternal, adolescent, and child deaths and disabilities; and (iii) increase access to responsive clinical and public health emergency services. The UHC Roadmap has been operationalized in the Health Sector Medium Term Development Plan (HSMTDP) for 2022–25. 3 The 2016 National Social Protection Policy also lays out a range of programs and interventions to reduce poverty and vulnerability, with close ties to health insurance and health care provision for the most vulnerable. Ghana has a three-tier, plural, and decentralized health care system, which is organized into tertiary, secondary, and primary levels of care with a mix of public and private providers (see Figure 4). The Ministry of Health oversees the health sector, which comprises the public, private, 2 The UHC Service Coverage Index (SDS Indicator 3.8.1) combines 16 tracer indicators of service coverage organized by four components of service coverage (Reproductive, maternal, newborn, and child health; Infectious diseases; Noncommunicable diseases and Service capacity and access) into a single summary measure. A detailed methodology of computation of the UHC SCI is presented in Annex A. The UHC Financial Protection indicator (SDS Indicator 3.8.2) measures the proportion of population with large household expenditures on health as a share of total household expenditure or income. 3 It is a constitutional requirement for all sectors in Ghana to develop a four-years sector medium-term development plan following the guidance of the National Development Planning Commission (NDPC). 11 faith-based, and traditional medicine sectors. The public sector tertiary care facilities are self- governing agencies while the community, primary, and secondary care facilities are under the Ghana Health Service (GHS), an agency of the MOH. The private, faith-based, and traditional medicine sectors are regulated by law and organized into federations and associations. Both public and private facilities are accredited by the Health Facilities Regulatory Authority (HeFRA) and credentialed by the National Health Insurance Authority to provide health services after meeting set standards for accreditation and credentialing. Whereas it is optional for private facilities to deliver services under the NHIS, all public facilities that meet the NHIS credentialing requirements are expected to deliver services to NHIS card holders in accordance with the NHIS benefit package. Figure 4: Ghana's Health System Organizational Structure Source: Adapted by the author from MoH archives. Notes: MOH = Ministry of Health; RHMT = Regional health management team; PHC = Primary health care; DHMT = District health management team; SDHMT = Subdistrict health management team; CHAG = Christian Health Association of Ghana; GHS = Ghana Health Service. 12 2. KEY HEALTH SECTOR PERFORMANCE INDICATORS 2.1. DEMOGRAPHIC AND EPIDEMIOLOGICAL TRANSITION In Ghana, wide disparities in the population distribution between rural and urban areas pose unique challenges to the health system; rural areas demonstrate a sparse population density and a higher dependency ratio compared to urban areas. The country’s population is estimated at 30.8 million in 2020, an increase of 6.1 million since the last census in 2010 (Ghana Population and Housing Census 2020). The annual intercensal growth rate of 2.1 percent is the lowest since independence. At the observed growth rate, the country’s population is estimated to reach 52 million by 2050. At 15.6 million, the female population outnumbers the male population by about 3 percent. Ghana’s population pyramid differs between urban and rural areas; according to the 2020 population census, 60.5 percent of the younger population is concentrated in urban areas, resulting in a higher dependency ratio in rural areas. Additionally, the various regions of the country are in different stages of the demographic transition. There are about 17.5 million (57 percent) people living in urban compared to about 13.4 million (43 percent) in rural areas of the country. The urban population grew by 6.1 percentage points between 2010 and 2020. The population density in the capital city, Accra, is 1,678.3 per square miles; whereas in the Savannah region it is 18.7 per sq. km, a disparity resulting from the higher urban population growth rate (Figure 5). The low population density in rural regions and the growing urban population pose unique challenges to health care service delivery and must be considered in subnational health resource allocation. Figure 5: Ghana's Population Growth Trend and Projection as a percentage, 2010–2050 60 52 50 44.9 40 37.8 30.8 32.4 28.48 29.77 30 26 27.22 24.66 20 10 0 2010 2012 2014 2016 2018 2020 2022 2030 2040 2050 Source: Authors, based on UN World Population Prospects. The total fertility rate (TFR) in Ghana has been steadily declining since the year 2000, though the rate of decline has been slow and there are wide rural-urban and regional disparities. Ghana currently has an average fertility rate of 3.5. The country’s median projected TFR is expected to continue on a downward trend. However, the TFR is anticipated to remain above the replacement level until 2050, and with high-low population variance. Although 13 the TFR urban-rural disparity has narrowed over the periods 2000–2010 and 2010–2021, a substantial gap still exists. In 2021, the TFR for the average urban woman was 2.7, while the average TFR for women in rural areas was 3.8. There are also significant regional disparities, with the TFR in the rural areas being twice as high as that of urban (as shown in Figure 6). Figure 6 highlights the significant and persistent urban/rural regional differences. Figure 6: Ghana's Total Fertility Rate by Region Source: Ghana Statistical Services 2021. An epidemiologic transition is expected to accompany the demographic transition in Ghana, resulting in increased prevalence of age-related degenerative diseases and noncommunicable diseases (NCD) that will require reorientation of the health system. According to the 2020 Population Census, 4.3 percent of Ghana’s population is aged 65 years and over. Between 2020 and 2040, the life expectancy at birth is expected to increase from 62.7 years to about 72.3 years (WDI 2021). As per projected estimates, the low TFR, low mortality rates, and an increased life expectancy will contribute to the demographic transition and aging of the currently young population (Figure 7). The demographic transition will also contribute to the epidemiological change in disease patterns. Age-related illnesses are expected to increase as the population ages, requiring increased funding, planning, and resourcing for regenerative and palliative health care services. In addition, globalization and increased disposable incomes impact lifestyle choices and behaviors, which can increase the risk of NCDs. For example, the increased consumption of processed foods with high levels of cholesterol and saturated fats lead to an increased incidence of cardiovascular or cerebrovascular disease (McCracken and Phillips 2016). Health care services and financing thus need to be reoriented toward preventive and promotive services that can mitigate these emerging risks. 14 Figure 7: Ghana's Projected Population Shift, 2020–2050 Source: Population pyramid, 2019 revision of the World Population Projection. 2.2. HEALTH OUTCOMES Ghana has made noticeable progress in maternal and child health outcomes during the last two decades, performing better than comparable peers. The under-five mortality rate (U5MR) in the country fell from 111 to 52 deaths per 1,000 live births between 2013 and 2017. Although Ghana did not quite reach the Millennium Development Goal (MDG) 4 target, the progress achieved represents an overall reduction of 53 percent during the period. Additionally, when compared to peer countries, Ghana’s U5MR is below the LMIC 2017 U5MR average of 54.1 deaths per 1,000 live births. Current projections show that Ghana is on the path to achieving the SDG targets for infant and neonatal mortalities by 2030, if current efforts are sustained (Figure 8). The maternal mortality ratio (MMR) also decreased by almost 36 percent from 484 to 308 per 100,000 live births between 2000 and 2017. Although the improvement was insufficient to meet the MDG 5.1 target, it was slightly higher than the average for LMICs (253 per 100,000 live births). Figure 9provides four scatter plots of Ghana’s position in relation to peer countries with regards to key health outcomes. 15 Figure 8: Trends and Projections: Ghana's Child Mortality Rates, 1993–2030 Source: World Bank staff, based on Ghana Demographic Health Survey, 1988–2022, and Ghana Maternal Health Survey 2017 (GSS, GHS, and ICF 2018). Figure 9: Ghana's Position on Key Health Outcome Indicators Maternal mortality Under-five mortality Life expectancy Prevalence of stunting 16 Source: World Bank staff using GHED. Note: GNI = Gross national income. The average trends for national health outcomes mask significant differences among income groups, regions, and genders. According to the 2017/18 MICS data, U5MR varied from 79 deaths per 1,000 live births in Ashanti to 31 deaths per 1,000 live births in Greater Accra. Likewise, the U5MR was lower among the richest quintile of households (46 per 1,000 live births) compared to the poorest quintile (53 per 1,000 live births) (Figure 10). Mortality rates for female children (58 per 1,000 live births) were higher than for male children (54 per 1,000). However, stunting rates among male children were 3.9 percentage points higher than among female children. Stunting rates of children living in households belonging to the least wealthy quintile (24.5 percent) were roughly three times higher than for children living in households in the wealthiest quintile. There were also substantial variations in the prevalence of stunting across regions: children from Greater Accra demonstrated the lowest stunting rates, while those from Northern Ghana demonstrated the highest. 17 Figure 10: Comparison of U5MR and Stunting, by Region, Gender, and Income Source: Multiple Indicator Cluster Survey 2017/18. Note: U5MR = Under-five mortality rate. The data also showed that there is a greater degree of comparative inequality in urban Ghana compared to rural areas 4 (Table 2). Similarly, there is also a disproportionate concentration of access to treatment for diseases and health insurance coverage among children from wealthier households. Table 2: Concentration Indexes of Key Under-Five Health Variables Concentration Concentration index value index valuea Urban Rural Urban v. Greater Other Accra v. rural Accra health other differences regions health regions differences 4 Both statistical tests based on either the large sample or the equal variance assumptions fail to reject the null hypothesis that the concentration index is the same in rural v. urban areas. 18 Prevalence of -0.11*** -0.12*** -0.05** -0.066** - -0.10*** -0.003 stuntingb 0.11*** Prevalence of -0.080*** - - -0.025 -0.034 - 0.033 diarrhea 0.098*** 0.073*** 0.066*** Seek 0.065** 0.116** 0.046 0.071 -0.0714 0.067 -0.14 advice/treatment when sick Child covered by 0.13*** 0.130*** 0.062** 0.068 0.20*** 0.15*** 0.06 any health insurance (NHIS or private) Source: Authors computation using the 2017/18 MICS data. Notes: NHIS = National Health Insurance Scheme. a. The statistical difference is computed under the large sample assumptions. b. After cleaning up the WHO Height for Age Z-core by making sure it remains within the -6SD and 6SD interval, we define stunting as the probability of HAZ < -2 Standard Deviation (SD). The star indicates the level of statistical significance at a 95 percent confidence interval (legend: * p<.05; ** p<.01; *** p<.001). Concentration indexes (CI) confirm inequalities in health outcomes and health services utilization across income groups. 5 CI of the prevalence of stunting, U5MR, and access to treatment of diarrhea and acute respiratory infection (ARI) constructed from repeated cross- sectional survey rounds of the Ghana Living Standard Surveys (GLSSs) from 1993 to 2014 are presented in Figure 11. For stunting, the CI values are consistently negative across the survey years with the level of concentration deepening since 2008, confirming that stunting progressively concentrates among children from least wealthy households. Likewise, we found that CI values are consistently negative across the survey years for U5MR, with the level of concentration deepening post-2003. By contrast, CI values for access to treatment for diarrhea and ARI show gradual reductions in relative inequalities to values of near zero (equality) by 2014. 5 The concentration index and related concentration curve provide a means of quantifying the degree of income-related inequality in a specific health variable. The concentration index has a negative value when the health variable is concentrated among the disadvantaged. The concentration index has a positive value when the health variable is concentrated among the most advantaged/better off. 19 Figure 11: Concentration Index Trends for Stunting, U5MR, and Key Health Treatments Concentration index for the prevalence of stunting Concentration index for U5MR Years 0.00 1993 1998 2003 2008 2011 2014 Years 0 -0.05 1993 1998 2003 2008 2014 Concentration Index -0.02 Concentration Index -0.10 -0.04 -0.06 -0.15 -0.08 -0.1 -0.20 -0.12 -0.25 -0.14 Concentration index for access to diarrhea treatment Concentration index for access to ARI treatment 0.18 0.20 Concentration index 0.16 0.18 Concentration Index 0.14 0.16 0.14 0.12 0.12 0.10 0.10 0.08 0.08 0.06 0.06 0.04 0.04 0.02 0.02 0.00 0.00 1993 1998 2003 2008 2014 1993 1998 2003 2008 2014 Years Years Source: HEFPI World Bank database. Notes: U5MR = Under-five mortality rate; ARI = Acute respiratory infection. 2.3. BURDEN OF DISEASE As a result of the demographic and epidemiologic transitions, deaths attributable to chronic NCDs are increasing in Ghana, while deaths attributable to communicable diseases and neonatal conditions are decreasing. In 2009, the three major causes of deaths in Ghana were malaria, HIV/AIDS, and neonatal disorders. Between 2009 and 2019, the share of deaths attributed to malaria and HIV/AIDS reduced by a third due to increased use of insecticide- treated bed nets among children less than five years and increased access to antiretroviral therapy, respectively. Deaths attributed to neonatal disorders also decreased by 18.6 percent. However, during the same period, deaths from stroke, ischemic heart disease and diabetes increased by 25.2 percent, 37.6 percent, and 24.6 percent, respectively (Figure 12). Additionally, the epidemiological transition is not consistent across Ghana. Wealthier communities in urban settings are at higher risk of chronic diseases, while less wealthy, rural communities are at greater risk of infectious diseases (Agyei-Mensah and Aikins 2010). Lower respiratory infections, which are predominant in rural populations, are expected to be the leading cause of death by 2040, along with NCDs such as ischemic heart disease, stroke, and diabetes (Figure 13). This implies that rural populations are likely to experience a dual burden of diseases in the medium-to-long term. Increasing urbanization and aging may also contribute to emerging causes of death such as road traffic accidents and injuries. 20 Figure 12: Ghana's Cause of Death Comparison, 2009–2019 Source: Adapted from IHME 2021. Note: HIV/AIDS = Human immunodeficiency virus/acquired immunodeficiency syndrome. Figure 13: Comparison of Five Primary Causes of Death in Ghana, 2019 to 2040 Source: World Bank based on IHME 2021. Communicable diseases currently contribute the most to mortality and morbidity combined, though the proportion shows a declining trend; meanwhile the proportion of NCDs is increasing annually. Communicable, maternal, neonatal, and nutritional diseases represent 52.3 percent of total disability-adjusted life years (DALYs), 6 though this proportion is 6The DALY was developed in the 1990s as a way of comparing the overall health and life expectancy of different countries. DALY includes both the potential years of life lost due to premature death and the equivalent years of “healthy'” life lost by 21 declining by 3.22 percent annually. On the other hand, NCDs (e.g., diabetes, cardiovascular diseases, mental and neurological diseases) contribute a smaller share at 40.66 percent of total DALYs, with the proportion increasing by 0.064 percent each year. 7 Given this trend, the total DALYs attributable to NCDs will overcome the share of DALYs attributable to communicable, maternal, neonatal, and nutritional diseases by 2026. While the prevalence of HIV has been on the decline since 2000, it has plateaued at 1.7 percent from 2018 to 2021. HIV prevalence in Ghana among people aged 15–49 years remains lower than the sub-Saharan Africa average of 3.2 percent (Figure 14). Further, Ghana has seen declines in mother-to-child transmission rates of HIV. The percentage of children born to HIV- positive mothers who tested negative at 18 months, currently stands at 74.4 percent of total children tested, and is approximately 20 percent lower than the 2025 target of >95 percent. Figure 14: Prevalence of HIV in Ghana, Percentage of Population Ages 15–49 Source: World Bank data. Note: HIV = Human immunodeficiency virus. 2.4 PUBLIC HEALTH EMERGENCY PREPAREDNESS AND RESPONSE Prior to the COVID-19 pandemic, Ghana had significant gaps in its public health emergency preparedness and response core capacities. In 2017, Ghana carried out a joint external evaluation (JEE), an objective, voluntary, independent peer-to-peer multisectoral assessment of the country’s health security preparedness and response capacities. The JEE identified huge virtue of being in states of poor health or disability, thus mortality and morbidity are combined into a single, common metric (Vos et al. 2020). 7Injuries and external causes represent the remaining category with 7.0 percent of total DALYs decreasing by 0.4 percent each year. 22 gaps in preparedness across all four public health functions with a rating of 45 out of 100, indicating the need to prioritize financial resources toward strengthening emergency preparedness and response. Ghana’s capacity was assessed to be limited in prevention (average score: 2.4 out of 5), detection (average score: 2.5 out of 5), response (average score: 1.8 out of 5), and other hazards (average score: 2.3 out of 5). The assessment identified two areas with minimal existing capacity (i.e., score of 1 out of 5). These include (i) antimicrobial resistance (AMR) detection, surveillance, and stewardship; and (ii) capacity for deployment and receipt of medical countermeasures, and the sending and receiving of health personnel during health crises. The report also identified nine areas in which Ghana fared only marginally better, scoring 2 out of 5. The emergence of Ebola viruses, Ebola disease, and COVID-19 has reemphasized the need to invest in public health emergency preparedness and response. Ghana is at high risk for epidemics due to the volume of cross-border air and land traffic. Although Ghana did not record any cases during the Ebola outbreak, the country activated its response mechanism at the national and subnational levels and provided leadership in laboratory testing and logistics to subregions. Ghana recorded its first COVID-19 outbreak on March 3, 2020. As of April 24, 2023, the country had reported 171,657 COVID-19 cases and 1,462 deaths. The COVID-19 pandemic posed a threat to human life directly through COVID-19 deaths, and indirectly by limiting access to and utilization of routine health services. These two major public health outbreaks within a decade attracted the attention of policy makers and moved public health emergency financing policy to the forefront of discussions. The government, notwithstanding the polycrisis maintained the level of funding to the health sector, as shown in Section 3.2. During the COVID-19 pandemic in 2020, general government revenues dropped sharply even as government spending surged to address social and economic impacts of the crisis. The government increased health spending by about GH¢2 billion (0.5 percent of GDP) in 2020 and provided nominal increases in the subsequent years. The core areas of government investment included the following: (i) Provision of protective gear, (ii) Procurement of medical and laboratory equipment; (iii) Procurement of testing kits; (iv) Community engagement and risk communication activities; (v) Construction of new treatment centers and improvements to existing facilities; and (vi) Social support to poor and vulnerable populations. In addition, the government recruited 37,000 health workers and granted an income tax waiver and life insurance to government health personnel. With support from the World Bank, the country strengthened institutional capacity for emergency preparedness and response by establishing an Inter-Ministerial Coordination Committee (IMCC) consisting of representatives from the Ministries of Finance, Health, Local Government, Gender, Children and Social Protection, Information, Transport, Interior, and the Office of the President to act as the apex coordinating body for emergency response. As the pandemic subsides, the country is setting up the Centre for Health Security to assume the role of the IMCC. Notably, the country has taken bold steps to procure essential equipment and protective gear to ensure effective treatment of infected persons while ensuring the safety of health care workers. In 2023, the government approved the Ghana Vaccine Institute Bill to enable the country to play a key role in the vaccine manufacturing value chain. The pandemic revealed that Ghana must make considerable investments in public health emergency systems to effectively respond to future emergencies. Ghana’s “National Strategy for COVID-19 Response” outlined an estimated budget of GH¢8.04 billion (US$400 million) between 2020 to 2024 to provide strong systems able to effectively respond to the COVID- 23 19 outbreak and to prepare for other public health emergencies in the short to medium terms. Despite recent efforts to strengthen emergency preparedness, Ghana remains vulnerable to major disease outbreaks. Ideally, as Ghana enters the post-COVID-19 era, the government should maintain the facility, infrastructure, and equipment baselines established during the pandemic and take the opportunity to enhance the health system’s diagnostic capacity and surveillance systems. However, financing constraints or debt sustainability may limit the scope of additional health allocations in government budgets. 24 3. HEALTH FINANCING 3.1 OVERVIEW OF HEALTH FINANCING IN GHANA In Ghana the main sources of funding for health care services are the annual government health budget (channeled through the MoH and NHIA); household out-of-pocket (OOP) payments; and external sources, primarily development assistance for health (DAH); 8 and other private health expenditures. Data from the WHO Global Health Expenditure Database show that Ghana’s domestic general government health expenditure, as a share of current health expenditure (CHE), increased from 35.1 percent to 49.7 percent, while the share of domestic private health expenditure declined from 49.2 percent to 41.1 percent between 2014 to 2020. The share of DAH in CHE decreased from 15.7 percent to 9.2 percent (Figure 15). Although the share of DAH has decreased over time, Ghana still receives a sizable amount of external funding. Notably, the most recent National Health Accounts (NHA) in Ghana was conducted in 2016, which means that consolidated health financing sources and expenditures data are not current. Therefore, the data presented on OOPs may be an underestimation of the current situation. Figure 15: Ghana's Health Expenditures by Funding Source, 2014–2020 Source: WHO GHED, 2022. Note: CHE = Current health expenditure. Ghana spends less on health as a percentage of GDP than the average in SSA and comparator countries, although the share of public spending as a percentage of current health expenditure is relatively higher than its comparator countries (Table 3). In 2020, Ghana’s CHE was 4.0 percent of GDP, lower than 6.5 percent and 5.4 percent, for SSA countries and LMICs, respectively. However, Ghana’s domestic general government health expenditure as a share of its CHE (49.7 percent) was higher than the average among SSA (36.2 percent), LMICs 8DAH includes grants, concessional loans, and aid in kind from bilateral, multilateral, or private foundations. Some DAH is channeled through the government budget while some of it finances health care expenditures at the facility or program level. 25 (45.7 percent), and a select number of relevant comparator countries such as Cote d’Ivoire (36.8 percent) and Senegal (33.4 percent). The ongoing demographic and epidemiological transitions, coupled with the population’s increased expectations of the health sector, will continue to put upward pressure on government health spending in the years ahead, consistent with global trends observed in other middle-income countries. We note that global benchmarking can be obscured by issues of data comparability since the latest National Health Account for Ghana was prepared in 2016. Table 3: Country Comparison of Ghana's Health Financing Indicators, 2020 Ghana SSA LMIC Cote d'Ivoire Senegal Zambia Kenya average average Current health 4.3 expenditure (% of 4.0 6.5 5.4 3.7 5.2 5.6 GDP) General 47.4 government health expenditure (% of 49.7 36.2 45.7 36.8 33.4 43.5 current health expenditure) GDP per capita PPP (current 2,129.6 2057 2,248 2,288.1 1,490.2 956.8 1,943.1 internal $) Population 32.18m 1.15bn 3.36bn 2,681m 16.43m 18.93m Source: 2018 WHO GHED. Notes: SSA = Sub-Saharan Africa; LMIC = Lower-middle-income country; GDP = Gross domestic product; PPP = Purchasing power parity. As of 2020, current government health spending accounted for about 2 percent of GDP and 7 to 8 percent of the overall government budget (Figure 16). However, these figures are higher than the lower-middle-income country average since many Asian countries in this income category spend relatively less on health. Figure 16: Comparison of Ghana's Health Spending Compared with SSA Peers and the LMIC average Govt health spending as % of GDP Govt health spending as % of total Govt budget 26 Source: World Bank staff. Notes: SSA = Sub-Saharan Africa; LMICs = Lower-middle-income countries; GDP = Gross domestic product; GNI = Gross national income. 3.2. PUBLIC HEALTH FINANCING SOURCES AND STRUCTURE The following section analyzes the level, sources, and composition of public health expenditures as well as the flow of funds, and presents a detailed analysis of NHIS expenditure and its contribution to the achievement of the UHC agenda. Additionally, it discusses how the public financial management (PFM) arrangement and system contributes toward efficient, effective, and transparent mobilization and utilization of limited health sector resources including development assistance. Figure 17: Ghana's Public Sector Health Funding Flow Source: Ministry of Health archives Notes: VAT = Value-added tax; SSNIT = Social Security and National Insurance Trust. 27 Each year, Ghana’s Ministry of Finance (MoF) allocates funds to the Ministry of Health (MoH) and NHIS through the annual budget and economic policy process. The bulk of public health financing is channeled through the Ministry of Health’s budget while earmarked funds are allocated to the NHIS for the purchase of health services from public and private health facilities (Figure 17). The MoH budget funds key public health service delivery, training, and the regulatory agencies. On the demand side, the NHIS contracts accredited government and privately owned health care providers to provide care to eligible populations. The MoF’s annual Budget and Economic Policy highlights the annual budget allocation to the NHIS, and therafter, the Parliament of Ghana approves the detailed budget lines for the NHIS budget through the allocation formula. Figure 18: Ghana's Trends in Real and Nominal Public Health Expenditure in million, GH¢ Source: Authors, using MoH, NHIA, and MoF expenditure data. Note: 2022 and 2023 data points reflect budget allocations, not actuals. Between 2015 and 2023, public health spending more than doubled in nominal terms, but only increased slightly in real terms as shown on Figure 18 above. Using 2013 as the base year GDP deflator, public health spending has remained relatively flat with only about a 12 percent increase during the period. During this period the composition of health care spending has changed. On average, 80 percent of the public health expenditure is channeled through the MoH budget, which has more than tripled in nominal terms from GH¢3,927 million to GH¢15,285 million from 2015 to 2023, an increase of 27 percent in real terms. At the same time, the NHIS expenditure has decreased by 36 percent in real terms. The NHIS budget, as a percentage of public health expenditure, has also declined considerably over the same period from 24 percent to 14 percent (Figure 19). 28 Figure 19: Ghana's MoH vs. NHIS Public Health Spending, 2015–2023 Source: Authors, using MoF and MoH budget and expenditure data. Notes: MOH = Ministry of Health; NHIS = National Health Insurance Scheme. Multiple factors continue to affect the government’s ability to reach the Abuja declaration 9 target of 15 percent spending on health. Since 2019, the total public health expenditure in Ghana has remained below 11 percent of the government budget. In contrast, since 2018, the education sector has received a gradually increasing share of the government budget from 11.8 percent in 2018, to 14.4 percent in 2022. The introduction of the government’s flagship free senior high school policy has in recent years led to increased budgetary allocations to the education sector. Additionally, and notably, interest payments constitute a prominent share of the budget at an average of 25.8 percent of the central government budget. With significant competing financial needs, the health sector remains at risk for underfunding by decision makers. 3.2.1. Ministry of Health Budget In Ghana, the majority of the MoH budget is spent on health workforce compensation. The MoH budget is financed by three main sources: Government of Ghana (GoG) revenue; internally generated funds (IGF), mainly OOP payments and NHIS payments at heath facilities; and development assistance for health (DAH). In 2022, government funding comprised over 60 percent of the MoH budget followed by IGF (27 percent) and donor resources (12 percent) as shown in Figure 20 below. Each year the MoF estimates the amount of internally generated funds and donor resources available to the MoH budget. Figure 20: Economic Composition and Sources of Ghana's MoH Budget, 2015–2022 Economic composition of MoH budget Sources of financing in Ghana 9In April 2001, heads of state of African Union countries met and pledged to set a target of allocating at least 15 percent of their annual budget to improve the health sector. At the same time, they urged donor countries to "fulfil the yet to be met target of 0.7 percent of their GNP as Official Development Assistance (ODA) to developing countries." 29 Source: MoH 2015–2020 expenditures and 2021–2022 budget data. Notes: MOH = Ministry of Health; Comp = Compensation; G&S = Goods and Services; GoG = Government of Ghana; IGF = Internally generated funds. MoH receives very limited discretionary funding from the MoF for operational expenditure and for preventive health and health promotion activities, which are expected to be financed through the MoH budget. As shown in Figure 21, about 90 percent of the share of the government’s contribution to the MoH budget is spent on salaries and compensation. The remaining ~10 percent is spent on capital investments and goods and services. Development assistance and IGF funds predominantly support capital investments and nonwage recurrent expenditures. Therefore, the MoH and its agencies rely on donor support to cover the cost of preventive health and health promotive activities and the IGF for operational expenditures. Figure 21: MoH Expenditure by Funding Source Source: Authors, based on MoH expenditure data. Notes: MOH = Ministry of Health; IGF = Internally generated funds; Capex = Capital expenditure. 3.2.2. The National Health Insurance Scheme Ghana’s National Health Insurance Scheme was launched in 2003 and positions the country as a regional trailblazer in health financing reforms. Most middle-income countries undertake a transition from MoH-dominated, input-based, supply-side financing systems to output-based, demand-side financing approaches including a “purchaser-provider split.” In principle such reforms can improve incentives for quality and efficiency, strengthen the accountability relationships within the health system, and generate valuable information on care provision. 30 The NHIS is primarily financed by the National Health Insurance Fund (NHIF), which receives funds collected through the National Health Insurance Levy (NHIL). Historically, the main financing mechanisms was a 2.5 percent levy on goods and services collected under the value-added tax (VAT), and 2.5 percent of Social Security and National Insurance Trust’s (SSNIT’s) contributions, together accounting for about 90 percent of NHIS revenue. Additional revenue sources include returns on NHIF investments and premiums paid by informal sector subscribers. 10 More recently, the 2021 Budget Statement and Economic Policy initiated a COVID- 19 levy intended to increase the NHIL from 2.5 percent to 3.5 percent. This extra source of revenue, if earmarked for the health sector, would significantly expand its buying power. The implementation of the levy, however, is still unclear and the ambiguity around the policy implementation could mean that the additional funding is not added to the NHIF. The tax computation by the Ghana Revenue Authority maintains a 2.5 percent tax for NHIL and separately charges a 1 percent levy on goods and services as the COVID-19 levy. The provisional collection under the levy for 2021 was approximately GH¢940 million; the budgeted collection for 2022 was GH¢1.1 billion. However, there is a lack of clarity as to where the 1 percent levy is accruing and whether the government intends to abrogate the tax post-COVID-19. Simultaneously, there is a broader effort to streamline the number of distinct levies used to mobilize revenues in Ghana, which may have future impacts on health care funding. NHIS is now facing significant financial difficulties due to two distinct financial gaps that have emerged in recent years. The first is a gap between NHIL revenues collected on behalf of NHIA and the actual budget allocation. In 2017, the “Earmarked Funds Capping and Realignment Act” imposed a ceiling on the share of total government revenue that can be transferred to statutory funds (Figure 22). This also interferes with the ability of NHIA to clear past arrears to providers. The cap was initially set at 25 percent, and in late 2022 was lowered to 17 percent, with severe implications for the 2023 budget in the form of a 25 percent decline from the previous year. The second gap is between the budgeted amount and the actual cash transferred to NHIA during the fiscal year. As a result of these two gaps, by 2022 NHIA was receiving only about half of the full NHIL amount due. Figure 22: NHIA Financing Gaps, 2016–2023 10Citizens under 18 years or above 70 years of age, SSNIT contributors (i.e., formal sector workers) and pensioners, indigents and pregnant women are exempted from premium contribution. The average NHIS premium paid by informal sector subscribers in 2020 was GH¢18.00 (US$3.12). 31 Source: World Bank staff. Notes: NHIS = National Health Insurance Scheme; SSNIT = Social Security and National Insurance Trust; NHIL = National Health Insurance Levy. Figure 23 shows the impact of the lower cap on the 2023 budgets of key statutory funds. Notably, the District Assembly Common Fund (DACF) is exempted from the capping law due to a guaranteed transfer of at least 5 percent of revenues, and therefore received a significant increase in 2023. A similar exemption for NHIF could be considered. While significant financial flows outside of ministries, departments, and agencies (MCAs) might be a valid concern in the case of other sectoral statutory funds, the institutional design of NHIA is fully consistent with good practice in many middle- and high-income countries. Figure 23: Uneven Impact of Capping on Statutory Fund Budgets, 2022–2023 Source: World Bank staff. Financial sustainability is a challenge for the NHIS. The NHIS has shown a positive balance sheet since 2017, based on promissory notes/warrants issued by the MoF, which suggests the scheme is currently on the path to financial sustainability (Figure 24). However, on a cash basis, after adjusting for outstanding health care provider claims arrears, the scheme shows significant budget deficits (Figure 25). The MoF must honor the annual budget commitments to the NHIS and take steps to clear all the arrears owed the scheme to ensure its financial sustainability. As of 2022, claims arrears to providers amounted to GH¢1.27 billion, an amount equivalent to over half of actual cash received from MoF that year. 32 Figure 24: NHIS financial performance excluding arrears, 2008–2021 Source: World Bank staff, based on NHIA data. Notes: NHIS = National Health Insurance Scheme; NHIA = National Health Insurance Authority. Figure 25: NHIS Financial Performance Including Arrears, 2008–2021 Source: Authors, using NHIS revenue and expenditure data. Note: NHIS = National Health Insurance Scheme. The delay in reimbursing providers makes financial planning challenging and the provider claims reimbursement process unreliable. To meet costs, providers are increasingly obliged to charge insured clients informal fees for services received, which has implications for access and financial protection (Akweongo et al 2021). Figure 26 shows the minimum number of months of arrears owed providers. Although reimbursement improved from three to six months between 2019 and 2020, it rose again in 2021. Delayed payments by NHIA have also reportedly undermined the 33 framework contracts signed with selected drug suppliers/wholesalers, who are reluctant to supply at the agreed price unless they are paid in cash. Figure 26: Number of Months of NHIS Claims Reimbursement Delays, 2017–2020 Source: Authors, based on NHIA data. Note: NHIA = National Health Insurance Authority. Simulating the impact of reduced out-of-pocket payments due to increased funds for NHIS. As reported above, there is a large gap between projected collections and the allocation of funds to NHIS. It is presumed that this has led to larger out-of-pocket payments by NHIS beneficiaries. For this analysis, the 2022 Annual Household Income and Expenditure Survey (AHIES) for Ghana is used instead of the 2016/17 Ghana Living Standards Survey, which is the official survey for poverty monitoring in Ghana. The reasons are the following: 1) The AHIES data are considerably newer and are likely to better reflect the country’s current situation. 2) There has been a considerable increase in the share of the population registered: a. According to the 2016/17 GLSS, a little over 50 percent of the population resided in a dwelling where at least one member was registered with NHIS. b. According to 2022 data, nearly 95 percent of the population reside in a household where at least one member is registered with NHIS. To determine the potential impact on poverty of closing the financing gap in NHIS spending, different percentage reductions in out-of-pocket spending by registered households are assumed. This then translates to a purchasing power gain for NHIS-registered households that have reported out-of-pocket spending in the relevant expenditure categories covered by NHIS. 11 For the analysis, the decile-specific purchasing power gain is applied to all registered households in the decile. Specifically, the average purchasing power gain among those who reported out-of- pocket expenditures is used for all those registered. This is done to account for the fact that health expenditures are sporadic and not necessarily a fixed occurrence. The assumption is not without consequence: health expenditures are sporadic, and this assumes that all registered individual in the decile would benefit similarly regardless of whether they had out-of-pocket expenses or not. 11Apass-though rate of 0.7 is used. A 1.0 percent reduction in out-of-pocket expenditure corresponds to a 0.7 percent reduction in household-specific out-of-pocket expenditure. 34 The assumption simplifies the analysis considerably. Additionally, cheaper health care does not necessarily translate to a higher propensity to spend on health care. Finally, the poverty impact ignores the long-run benefits of a potentially healthier population. Through improved health outcomes the population can expect to become more productive, which will likely translate into larger incomes and potentially larger tax revenues in the future. The impact on absolute and extreme poverty from a percentage drop in out-of-pocket expenditures from households is presented in Figure 27. A 10 percent reduction in out-of-pocket expenditures corresponds to 0.1 percentage point drop in absolute poverty and a 0.14 percentage point drop in extreme poverty. In essence, a fifty percent reduction in OOP expenditure results in more than hundred thousand people removed from absolute poverty whereas about 800,000 removed from severe poverty (Figure 29). Improvements in inequality are modest, but also expected (Figure 28). Figure 27: Percentage Change in Poverty Given a Percentage Drop in Out-of-Pocket Expenditures Source: World Bank staff Figure 28: Gini Index Point Change Given a Percent Drop in Out-of-Pocket Expenditure 35 Source: World Bank staff Figure 29: Number of People Lifted Out of Poverty Given an X Percent Drop in Out-of-Pocket Expenditures Source: World Bank staff Out-of-pocket payments have been reported among NHIS members and nonmembers for services received at health facilities. As seen in Figure 30 and Figure 31, OOP payments are on the rise in both primary and secondary/tertiary facilities. While the share of NHIS as a proportion of IGF is higher in primary health care, the share of OOP payments continues to increase. Notably, OOP payments are observed by insured and uninsured patients, which raises concern around the effectiveness of NHIS to ensure financial protection and/or the appropriateness of the existing essential benefits package. A recent study on OOP payments for insured clients, published in 2021, corroborates anecdotal evidence of informal copayments at 36 health facilities 12. According to the study, about half of the patients at health facilities, irrespective of their NHIS membership status, pay OOP for services received. Specifically, 46.9 percent of insured clients paid OOP for health services compared to 49.7 percent of those who reported not having enrolled with the NHIS. Notably, 42 percent of the insured clients belonging to the poorest quintile paid OOP for services. Additionally, informal copayments are more common in urban regions. The Korle Bu Teaching Hospital, the country’s premier and largest hospital, did not accept NHIS for reimbursement of drugs and some laboratory analysis until recently when they partially relaxed their stance. Patients and their families have the option of either paying OOP for drugs or purchasing NHIS-covered drugs outside the hospital. Figure 30: Comparison of OOP Payments vs. NHIS Payments at Primary Health Facilities, 2017–2021 90% 79% 76% 76% 80% 73% 70% 66% 60% 50% 40% 34% 24% 27% 30% 21% 24% 20% 10% 0% 2017 2018 2019 2020 2021 NHIS share of IGF OOP share of IGF Source: Authors, using GHS data. Notes: OOP = Out-of-pocket; NHIS = National Health Insurance Scheme; IGF = Internally generated funds. 12 Akweongo et al. 2021 37 Figure 31: Comparison of OOP Payments vs. NHIS Payments at Secondary and Tertiary Care Facilities, 2015–2021 80% 72% 73% 72% 72% 74% 70% 69% 70% 60% 50% 40% 31% 30% 28% 27% 28% 28% 30% 26% 20% 10% 0% 2015 2016 2017 2018 2019 2020 2021 Share OOP Share NHIS Source: Authors, using data from the MoH. Notes: OOP = Out-of-pocket; NHIS = National Health Insurance Scheme. Recent OOP payment trends represent an unfortunate reversal of NHIS’s earlier achievements in providing financial protection to beneficiary households. Survey data from GLSS-7 in 2017 show that NHIS beneficiaries incurred significantly lower OOP expenses than nonbeneficiaries; the difference was most evident among the lowest deciles of the socioeconomic distribution (Figure 32). Figure 32: NHIS Member OOP Comparison, by Decile Source: Worl Bank estimate based on Raju and Younger 2022. Notes: NHIS = National Health Insurance Scheme; OOP = Out-of-pocket. 38 While finding a solution to revenue shortfalls is the most pressing priority, there is also an agenda to strengthen NHIA expenditure patterns. From 2009 to 2021, expenditure on claims reimbursements declined considerably. In 2009, 85.3 percent of NHIS revenue was channeled toward claims payment. In 2021, 59 percent of NHIS revenue was paid to provider claims, indicating a decline in the priority accorded to claims reimbursement compared to nonclaims expenditures (Figure 33). The situation has resulted in the previously discussed claims arrears issue. Notably, public facilities receive salary payments directly from the MoF, which may impact the urgency to resolve claims arrears. Nonclaims expenditure consists of support to the MoH for preventive care service, and operational and investment costs. The operational cost includes cost of information technology; claims processing, support to district office; support for district health project (money given to parliamentarians to undertake developmental projects in their constituencies); biometric identification cards and authentication systems; and office infrastructure (Table 4). The NHIS support to the MoH constitutes 10 to 12 percent of NHIS expenditure and covers the cost of essential health commodities and the government’s copayment commitment for vaccines. Detailed assessment of the NHIS contribution to MoH shows that funds are sometimes used to cover the cost of other programs such as payment for drone services or operational funding for health training schools. Essentially, MoH is tapping into NHIS for discretionary funding due to its own funding shortfall, at the cost of diverting NHIS attention from its primary function to reimburse claims expenditures. Ultimately, patients bear the burden of this dysfunctional system in the form of OOP payments. Figure 33: Percentage of NHIS Claims Expenditure as a Share of Total Revenue Source: Authors, based on NHIA actual revenue expenditure data. Notes: NHIS = National Health Insurance Scheme; NHIA = National Health Insurance Authority. 39 Table 4: NHIS Expenditure Breakdown Expenditure items GH¢, millions Percentage of funds Claims reimbursements for 2020 1303.11 54.93 NHIA operational cost 294.54 12.42 Support to MoH for public health and preventive care 244 10.29 Nationwide ICT network 122.17 5.15 Office buildings 94.31 3.98 Biometric ID cards and authentication system 87.4 3.68 Support to district offices 56.9 2.40 claims data capturing 42 1.77 Support for district health projects 34.5 1.45 Claims processing centers e-claims 33.4 1.41 Others including contingencies 59.97 2.53% Total 2372.3 100.00% Source: NHIS Allocation Formula 2021. Notes: NHIS = National Health Insurance Scheme; NHIA = National Health Insurance Authority; MOH = Ministry of Health; ICT = Information and Communication Technology. There is also scope to reorient NHIS claims expenditures toward subdistrict facilities under the Network of Practice (NoP) Initiative. About 17 percent of NHIS claims expenditure goes to health care providers at the subdistrict level; about 58 percent goes to district hospitals (Table 5). In 2021, 84 percent of total claims expenditure went to PHC facilities, which is the district hospital level and below. Meanwhile, only 17.37 percent of total services were purchased at the subdistrict level and below, although about 40 percent of outpatient services are provided at these levels. Most importantly, subdistrict-level facilities make up 82 percent of all accredited NHIS facilities and contribute significantly to outpatient care (Figure 34). The recently developed Ghana UHC roadmap proposes increasing the percentage of NHIS expenditures occurring at the subdistrict level to 50 percent. This may require a review of the prescribing levels of the NHIS benefit package to expand the services the NHIS purchases at the subdistrict level. Table 5:NHIS Claims Expenditure by Facility Type, 2021 Facility type ` Amount paid GH¢ Percentages Tertiary hospital 39,558,702.19 5.86% Secondary hospital 45,177,904.18 6.69% Primary hospital 393,085,123 58.19% Policlinic 14,108,802.15 2.09% Clinic 41,506,615.25 6.14% Health Centre 76,722,867.43 11.36% Maternity home 5,087,232.74 0.75% CHPS compound 35,506,317.48 5.26% Pharmacy 15,427,110.98 2.28% 40 Chemical shop 36,391.07 0.01% Laboratory 3,406,553.12 0.50% Diagnostic center 5,030,735.36 0.74% Ultrasound center 861,125.61 0.13% Total 675,515,480.58 100.00% Source: Authors, using data from the NHIA. Notes: NHIS = National Health Insurance Scheme; CHPS = Community health planning services. Figure 34: Distribution of NHIS-Accredited Facilities in Ghana, by Type Source: Authors, using data from the NHIA. Note: NHIS = National Health Insurance Scheme. NHIS medical claims rejection rate is highest among frontline primary health care facilities compared to other facility types. The NHIS claims management system involves multiple stages of reviews and vetting of claims. 13 The NHIA, in its effort to reduce claims processing time, has introduced an electronic claims processing system and will phase out paper claims in 2023. The vetting processes flag improperly filed claims, including those suspected to have been double-counted. An assessment of the quality of claims submitted by health facilities in the Volta Region showed that about 18 percent of the submitted claims were incomplete. Despite the relatively simple nature of outpatient department (OPD) claims forms, the claims rejection rate is highest at health centers and community health planning services (CHPS). Tertiary and secondary facilities have more positive and improved claims adjustment rates over time (Figure 35). Reasons for the high claims rejection rate at the frontline level, include low capacity to correctly process claims and submission of claims for services above the authorized level for the 13 Nsiah-Boateng et al. 2017. 41 facility type. NoPs can help address these challenges through knowledge-sharing among facilities and streamlining of financing mechanisms and service delivery. More generally, the transition to electronic claims opens the door to using data analytics to improve strategic purchasing. This can include medical audits and fraud control efforts, with significant potential to strengthen the quality, efficiency, and integrity of care provision. Regular- cost data collection could help to inform tariff schedule updates to ensure adequate reimbursement so that providers are not driven to seek informal OOP payments from patients. Ghana uses diagnostic-related groups (DRGs) to reimburse inpatient care, in line with many other countries, but these have not been updated in over 10 years, and there are concerns about upcoding. Benchmarking providers to promote rational drug use could be used as a counterweight to the current fee-for-service approach for medicines reimbursement. Figure 35: Claims Adjustment Rate by Facility Type, 2019–2021 Source: Authors, using data from NHIS. Note: CHPS = Community health planning services. Despite a generous benefits package, the NHIS does not explicitly cover cost-effective preventive and promotive health care services. The NHIS payment mechanism reimburses curative health services and some individually provided preventive health services such as family planning. However, the absence of NHIS provider purchasing mechanisms that link provider payments with population health needs represents a missed opportunity to put the patient at the center of service delivery. The current NHIS payment mechanism predominantly pays for curative services that are delivered at health facilities with limited incentives for the provision of preventive health services and health promotion. Meanwhile, the disease profile of Ghana and the epidemiological transition of the country, as discussed in earlier sections, require the delivery of preventive health services to mitigate the rise of noncommunicable diseases. 42 While the NHIS population coverage is the second highest in Africa (Figure 36), there is an opportunity to enhance coverage of poor and vulnerable groups. The NHIS is one of the main policy tools providing financial risk protection to the citizens of Ghana. After enrollment stagnated at about 35 percent for several years, the scheme has recently seen an increase in membership from 10.65 million to 16.31 million between 2017 and 2021, representing an increase in population coverage from 35.6 percent to 54.4 percent (Figure 37). The NHIS is leveraging information technology, such as integration of the National Identification Authority’s database to facilitate and ensure membership renewal. Notwithstanding the high national population coverage, regional disparities exist in NHIS active memberships across regions in Ghana (Figure 38). Data from 2019 to 2021 consistently show population coverage in the Greater Accra and Oti Regions at about 40 percent, whereas the Bono, Ahafo, Upper East, and Upper West Regions show about 80 percent population coverage. Also, this increase in absolute NHIS enrollment numbers masks the lower coverage for poor and vulnerable populations, as well as coverage for Free Maternal Care (FMC) program. To address some of the gaps identified above, the World Bank–supported Primary Health Care Improvement Program for Results (PforR) is providing incentives for the NHIS to increase membership enrollment with particular attention to the enrollment of vulnerable populations. Figure 36: Health insurance scheme (all forms) coverage for Ghana and regional peers, 2021 Source: World Bank Country PERs. Figure 37: NHIS Enrollment Rate Trends, 2017–2021 43 60 52.7 54.4 Population coverage (%) 50 40.57 35.6 35.77 40 30 20 10 0 2017 2018 2019 2020 2021 Years Source: Authors, using data from NHIS. Note: NHIS = National Health Insurance Scheme. Figure 38:NHIS Population Coverage by Region, 2019–2021 Source: Authors, using data from NHIS. Note: NHIS = National Health Insurance Scheme. Coverage for the exempted categories remains a priority for the MoH and the NHIA; targeted initiatives are required to increase enrollment for indigents. About 64 percent of NHIS active members are in the exempted category. This includes people under the age of 18 (42 percent) and pregnant women under the free maternal health care policy (4.9 percent) (Figure 39). The absolute number of the indigent population on the scheme has increased but at a low growth rate. Although the Ghana National Household Registry (GNHR) is meant to collect and maintain a listing of the poor and vulnerable population, it lacks nationwide coverage and effective linkage to the NHIS database. A memorandum of understanding (MoU) has been developed by 44 the NHIA and the GNHR to ensure linkage of the NHIA and the GNHR database. The aim of the MoU is to enable easy identification and enrollment of the poor and vulnerable population and also support enrollment policy changes such as the number of years for which an indigent stays on the register. Per the current policy, indigents are required to be reregistered each year, which presents logistical and data challenges. Figure 39: Composition of NHIS Membership, by Population Subcategories 45% 42% 40% 36% 35% 30% 25% 20% 15% 8% 10% 5% 5% 4% 5% 0% 2020 SSNIT pensioners SSNIT contributers 70 years and above Pregnant women Indigents Informal/(non-exempt group) Under 18 years Source: Authors, using data from NHIS. Notes: NHIS = National Health Insurance Scheme; NHIA = National Health Insurance Authority; SSNIT = Social Security and National Insurance Trust. 3.3. DEVELOPMENT ASSISTANCE FOR HEALTH Ghana has been the recipient of significant development assistance over the years, but external financing is expected to decrease in the future. DAH in Ghana represented 11.02 percent of current health expenditure (CHE) in 2018, which was slightly lower than the average of 12.5 percent among SSA countries, but higher than the average of 2.8 percent among LMICs. However, some of the relevant comparator countries, namely Cote d’Ivoire and Zambia, receive two to four times more DAH, respectively. Over the medium term, as Ghana’s per capita GDP is expected to grow to middle-income status, DAH will decline as the country graduates from GAVI, Global Fund, and the International Development Association (IDA) financing (MOH and Oxford Policy Management 2018). In 2019, the country introduced its “Beyond Aid” agenda that laid out a strategy to reduce its dependence on external aid across various sectors and identified steps to implement the strategy (Ghana Beyond Aid Charter and Strategy Document 2019). While the strategy is an important first step in financial independence, there is limited information on the implementation of the strategy. Additionally, there have been no significant policy changes that confirm the government’s willingness to assume full responsibility to financing expenditures related to the procurement of essential commodities and vaccines. 45 Today, donor funding fills a critical gap in the government budget and supports the government’s prioritization of primary health care (PHC). Despite the significant drop in donor funding over the years, from US$20.09 per capita in 2015 to US$8.5 per capita in 2019, it continues to fill critical gaps in financing capital investment and operational costs in the health sector, and helps the government balance its budget allocation between recurrent and capital investments. Donor funding prioritizes preventive services such as immunization and malaria control programs. The use of donor funds is more profound at the subnational level due to low or insufficient operations funding from the government. Therefore, the decline of donor support will have significant impacts on the sustainability of essential health programs if the government budget is not allocated to compensate the declining donor funding. Declines in donor funding will also undoubtedly impact the supply of essential commodities for HIV/AIDS, family planning, tuberculosis (TB), and vaccines for childhood immunization, which are not covered by the NHIS. In 2022, the NHIS added family planning services to the benefits package. However, family planning commodities are are not covered by procured by the MoH with donor support. Development partners provided significant support for the COVID-19 health response in Ghana, though the surge in financial support is likely to be episodic and unsustainable. Development partners rechanneled support from other sectors and provided additional funds to manage the COVID-19 pandemic in Ghana. Between 2020 and 2021, development partners mobilized about US$640 million for COVID-19 and emergency response activities, of which COVID-19 vaccine–related activities are estimated at US$264 million. 14 As with the fluctuation in development support experienced during the 2014 Ebola virus outbreak, decision makers should prepare for the inevitable decline of COVID-19 surge funding. There is a lack of consistency in the modalities for transfers and execution of DAH; many development partners provide health funds outside the government budget. Most donors earmark funds allocated to the health sector. However, the modality of transfer and execution of these earmarked funds differ from one donor agency to the other. While some development partners transfer funds to MoF with specific instructions, others transfer funds directly to MoH, Ghana Health Service (GHS), or other health sector agencies. Estimates indicate that from 2016 to 2018, between 61 percent and 73 percent of total DAH was not recorded in the MoF budget (Table 6). Table 6: Development Assistance On and Off Budget Comparison, 2016–2018 2016 2017 2018 GH¢ % GH¢ % GH¢ % DAH on-budget 446,816,525 27 718,875,771 39 413,511,210 29 DAH off-budget 1,182,760,892 73 1,109,390,998 61 1,000,887,199 71 Total DAH (both on and 1,629,577,417 100 1,828,266,769 100 1,414,398,409 100. off budget) Source: Authors’ calculation using WHO GHED and MOF budget statements, 2016–2018. 14 COVID-19 donor mapping tool. This includes secured funds programmed up to 2023. 46 Note: DAH = Development assistance for health. 3.4 FINANCING VACCINES AND ESSENTIAL MEDICINES The government of Ghana has shown commitment to providing essential health services to the population in the past years, but key challenges persist in the lead-up to donor transition. The country has made significant progress in the reduction of childhood morbidity and mortality through its Expanded Programme of Immunization (EPI). For instance, Ghana attained elimination status of maternal and neonatal tetanus in 2011. Currently, immunization coverage is at over 90 percent for most childhood vaccines, ranking Ghana among the top performers on the continent for childhood vaccination coverage. As donor resources decline, the government is expected to assume full responsibility for purchasing and delivering essential medicines and vaccines. As indicated in Table 7, the government has steadily increased the percentage of routine immunizations financed by government funds from 35 percent in 2017 to 45 percent in 2021. Nonetheless, government funding constituted less than half of the total immunization expenditure as of 2021. Total immunization expenditure more than doubled in 2022, compared to the average immunization expenditure between 2017 to 2021, but the increase in government budget was inadequate to cover the gap. As a result, Ghana for the first time in over 15 years, ran out of key routine vaccines in early 2023. Notably, the share of government funds in total expenditure on vaccines used in routine immunization reduced from 49 percent in 2021 to 20 percent in 2022. The current financing mechanism for vaccines and essential commodities leaves critical gaps. The Regional Medical Store (RMS) and tertiary hospitals procure, primarily from private wholesalers, most medicines used in Ghana. District hospitals rely on the RMS for the supply of medicines and supplement their supplies directly from wholesalers and private pharmacies. Medicines on the approved NHIS list are paid for by the NHIS. Other drugs and essential commodities, including HIV, AIDS, TB, and malaria medications, and anti-snake, family planning commodities and vaccines, are financed outside of the NHIS. The government of Ghana has, over the years, relied on development partner support to partially finance these essential commodities, including vaccines. Notably, the Global Fund, US government, Canada, and GAVI have been the major partners in the supply of essential medicines and commodities through copayment arrangements. The Ministry of Health has in the recent past relied on the NHIS to fulfill its copayment obligations through an allocation of about 12 percent of NHIS expenditures in support of preventive health service delivery. This arrangement however has three major challenges: the inadequacy of the NHIS transfers to cover all the GoG cofinancing obligations; delays in the transfer of funds from the MoF to the NHIS; and lastly, the lack of clarity in the allocation of the insufficient budget among priority objectives by the MoH. Moreover, the use of NHIS funds for the purchase of vaccines and essential commodities has the potential risk of detracting the attention of the NHIA from its core mandate of paying for medical claims, as already noted in the low priority to claims payment Figure 33 above. The Ministry of Health estimates the funding gap for childhood vaccine, HIV AIDS, tuberculosis, and malaria commodities at about US$71,250,058 (GH¢883,520,812); this includes 2022 arrears and 2023 commitments for GAVI and the Global Fund. 15 15 MoH Annual Program of Work, 2023. 47 Table 7: Routine Immunization Expenditure in Ghana, US$ Immunization-related expenditures 2017 2018 2019 2020 2021 2022 Total expenditure (from all sources) on 37.8M 20.2M 38.7M 27.2M 27.2M 64.4M routine immunization, including vaccines (US$) Total expenditure (from all sources) on 32.7M 16.2M 38.7m 18.6M 23.6M 57.8M vaccines used in routine immunization (US$) Government expenditure on routine 13.2M 6.0M 14.9M 18.0M 12.3M 23.9M immunization, including vaccines (US$) Government expenditure on vaccines 9.4M 3M 11.6M 12.1M 11.6M 11.5M used in routine immunization (US$) Percentage of total expenditure on 35.00 30.00 38.00 66.34 45.23 37.12 routine immunization financed by government funds (%) Percentage of the total expenditure on 29.00 19.00 44.00 65.33 48.88 19.95 vaccines financed by government funds (%) Source: WHO Immunization expenditure data, 2023. 3.5 PUBLIC FINANCIAL MANAGEMENT IN THE HEALTH SECTOR Ghana has made consistent efforts toward improving PFM, although scale up of PFM intervention to ministries and departments at the subnational level has been slow. The government has undertaken significant and continuous PFM reforms over the past two decades. From 1997 to 2003, the government implemented a public financial management reform program. Between 2006 and 2016, it followed the initial work with a short- and medium-term PFM action plan. In 2010, the MoF introduced the PFM Reform Strategy, which was renewed in 2016. In 2016, the PFM Act (Act 291) laid out clear public fund management directives and revised the original PFM Reform Strategy for implementation between 2022 to 2026. These legal and policy reforms have led to a strengthened and better-integrated PFM system. However, these PFM interventions are weakly implemented at the level of ministries, departments, and agencies (MDA). In the health sector, weak PFM practices result in planning and budgeting inefficiencies, weak expenditure controls, and lack of transparency, all of which can negatively impact health outcomes. There are challenges in implementing an integrated financial management information system that increases expenditure visibility across programs and levels. The government has adopted the Ghana Integrated Financial Management Information System (GIFMIS) to prepare and execute budgets, prepare financial reports, and conduct audits. By design, the GIFMIS covers budgeting and expenditure processes of all government funds including IGF, statutory funds (SFs), and project/donor funds (DFs), in addition to the consolidated funds (CFs). In line with the 2016 PFM Act, the health sector–specific PFM objective is to institute strategies and procedures that ensure budget predictability; adopt standardized, scheduled fund flow procedures; ensure timely budget execution; and enable better reporting, accountability, and transparency. Despite strong commitment and support from donor partners, the health sector’s rollout of the GIFMIS has been slow, due in part to the undefined roles of the MoH and the GIFMIS Secretariat at the MoF. Currently, the GIFMIS is only deployed at the national level and covers 48 the government’s portion of the health sector budget (excluding donor partner [DP] funds and IGFs). This means that health facilities (including the majority of public facilities) are unable to leverage the effective budget and expenditure controls embedded in GIFMIS. Further, a Financial Management and Accounting Information System (FMAIS) should be deployed for efficient and effective management and monitoring of allocation of inputs at the facility level vis-à-vis their outputs. Figure 40 Comparison of MOH Allocation and Expenditure, 2017–2022, in Million GH¢ Source: Authors, based on budget and expenditures data from MoH and holistic assessment. Note: MOH = Ministry of Health. The MoH expenditures are consistently higher than its budget, indicating a high execution rate but also signaling possible concerns with planning and budgeting inefficiencies. Available MoH budget and expenditure data from 2017 to 2022, indicate that actual expenditures of the MoH consistently exceed allocated budget and, in most cases (except the year 2020), the revised budget allocation (Figure 40). These data demonstrate the MoH capacity to execute its annual budget, but also raise concerns about the accuracy and comprehensiveness of the planning and budgeting process. Notably, the capped budget ceiling of the Medium-Term Expenditure Framework (MTEF) limits the MoH from submitting an ideal budget for approval. Ideally, the MoH planning and budgeting process should be comprehensive and include all streams of funding aided by a tool that captures inputs by type and tracks the spending in each category throughout the year. This would allow the MoH to maximize government, quasi- government, and donor support in a more integrated and effective way. 49 Quasi-government health institutions that provide health services to the population are consistently unaccounted for in the health sector budget and expenditure reports. Quasi- government institutions in Ghana include secondary- and tertiary-level health facilities operated by ministries and agencies other than the MoH. These institutions and health facilities receive full or partial budgetary support from the GoG through their parent agencies/ministries to provide health services to the population. The MoH has taken steps to integrate these institutions into the health sector planning process through a representative body, the Ghana Association of Quasi- Government Health Institution. However, in the absence of a well-defined framework to collect and organize budget and expenditure data from these institutions, government expenditure on health is likely underestimated. The process of NHIS budget allocation and approval is separate from that of the MoH, and this could be a source of budgetary and planning inefficiency. The NHIS plays a crucial role in the implementation of the MoH’s work program as the largest purchaser of health services in the country. The NHIS provides a generous benefits package covering all costs for about 95 percent of common health conditions. 16 Notably, the NHIS budget allocation and approval process runs parallel to the MoH budget approval process. Although the program-based budgeting (PBB) and medium-term development planning process of the MoH accounts for NHIS program outputs, the Parliament of Ghana separately determines the programmatic allocation of NHIS expenditure through an allocation formula while the MoH budget is allocated through the appropriation bill. These separate budget approval processes often leave gaps in financing essential health care such as procurement of essential health commodities, and potentially causes double-counting as in the case of IGFs for health facilities. Existing health policies aim to improve service delivery at the subdistrict level but fail to address gaps in PFM capacity at that level. The development of the NoP requires a strong hub model and, in this case, strong health centers. However, over the years, health centers have received limited support, which has limited their ability to function as subdistrict hubs for the delivery and management of health care services. As a result, the district management team must provide significant financial management support to the subdistrict level. The MoH budget remains largely input- or line item–based despite the rollout of PBB. The transition from activity-based budgeting (ABB) to PBB theoretically allows budget management centers (BMCs) greater flexibility to allocate resources according to priorities and objectives. Currently, the PBB remains largely input-based, and covers capital expenditure, goods and services, and compensation. In contrast, NHIS payments are output-based even though the current PFM arrangement requires that the service component of the NHIS reimbursement be recorded as part of IGFs, which can then be classified as budget and allocated along input items. The workforce wages component of the MoH budget and the drug component of NHIS reimbursement are protected. Therefore, health care managers are left with a small amount of discretionary funds to allocate based on set outputs and needs. IGF are the second-largest source of the MoH budget after domestic tax revenues, and predominantly comprise NHIS reimbursement to health facilities for services rendered. IGF include fines, penalties, forfeitures, fees and charges including OOP payments, rent on government lands and buildings, interest on government investments, dividends, and all other 16NHIS benefits package covers both outpatient and inpatient services, oral care and eye care services, maternity care and all emergencies, with few exclusions: http://www.nhis.gov.gh/benefits.aspx. 50 revenue generated from the activities of BMCs. A proportion of funds (about 30 percent) generated by the MoH revenue-generating agencies is, by law, transferred to the central treasury to support the government’s expenditure while the remaining funds are retained by the BMCs. Each year, the MoF estimates the expected funds that will be generated by BMCs and allocates a percentage to be retained on the entity’s budget as “IGF.” Thus, the internally generated funds in the MoH budget is an aggregate of internally generated funds from the various BMCs under the MoH. For health care facilities, however, a significant proportion of IGFs come from NHIS reimbursement for services delivered. The MoF allows health care facilities to retain 100 percent of internally generated funds. These funds are used to offset the cost of delivering care in addition to financing operational expenditures such as support for home visits and salary payments for casual staff. This structure underscores the importance of NHIS reimbursements for health facilities. Health facility managers and district health management teams (DHMTs) have limited discretionary powers and funds to efficiently allocate resources for service delivery. IGF and earmarked donor funds are the main sources of discretionary funding for health facilities. Retained IGFs are spent on goods and services that can be put into two categories: drug and nondrug expenditure. To ensure the regular supply of drugs, funds are specifically protected for drug-related expenditures. Thus, health facilities and DHMTs only have nondrug IGFs for operational activities and maintenance of health facility infrastructure and equipment. Although health facilities are supposed to receive funding from the MoH budget to finance operational activities, fund flows to health facilities, particularly at the primary care level, are consistently low. As a result, equipment and supplies at health facilities are frequently unavailable, impacting the quality of health services. Although Ghana has established robust information management systems, lack of integration of the existing data systems and inadequate coordination among relevant implementing agencies hinder their potential. Ghana has continuously improved its District Health Information Management System (DHIMS2) and has introduced innovative tools such as the e-tracker to enable tracking of individual service utilization data. Besides the DHIMS2, other data systems such as the NHIS claims and membership databases, the Integrated Disease Surveillance and Reporting system, and the Ghana Integrated Logistics Management Information Systems (GhiLMIS) Human Resource Information Management System exist. However, these databases function in silos with limited to no integration or cross-validation, limiting their potential. 51 4. HEALTH SERVICES UTILIZATION Ghana has made substantial progress toward universal health coverage (UHC) over the past three decades. The UHC Service Coverage Index (SCI) 17 for Ghana gained roughly 20 points, equivalent to a relative increase of 66 percent, between 1990 and 2019. The SCI combines service coverage values for 14 essential health services and ranges from 0 to 100. Over the past 30 years, the UHC SCI has increased annually by 1.75 percent (Figure 41). More specifically, between 2015 and 2019, the UHC SCI score for Ghana has been constantly above or equal to that of comparator countries Despite this progress, an absolute SCI value of 49.1 in 2019 indicates the need to accelerate efforts toward UHC. Figure 41: UHC SCI Progress in Ghana and Comparison with Regional Peers Benchmark UHC Service Coverage countries Index 2015 2017 Ghana 46 47 Cameroon 43 46 Cote d’Ivoire 46 47 Mauritania 36 41 Liberia 37 39 Nigeria 42 42 Source: IHME Ghana Country Profile. Source: WHO Global Health Observatory database. Notes: UHC = Universal health coverage; SCI = Service Note: UHC = Universal health coverage. Coverage Index. The number of health care facilities in the country has increased, but there are gaps in the quality of care. As of 2020, there are over 9,300 health facilities in Ghana (Table 8), delivering a range of services, compared to 3,500 facilities reported in the 2017 PER. The distribution of health facilities is shown in Figure 42. 17 The UHC Service Coverage Index combines 16 tracer indicators of service coverage organized by four components of service coverage (Reproductive, maternal, newborn, and child health; Infectious diseases; Noncommunicable diseases Service capacity and access [[AU: What is the fourth component?]]) (is it okay to delet this as UJC coverage has been explained in earlier chapter) into a single summary measure. A detailed methodology of computation of the UHC SCI is presented in Annex A. 52 Table 8: Ghana's Health Facilities by Type Facilities Number Percentage CHPS 6,319 67.95 Health centers 992 10.67 Maternity homes 343 3.69 Clinics 1,020 10.97 Polyclinics 64 0.69 Psychiatric hospital 4 0.04 District hospital 110 1.18 Hospitals 435 4.68 Teaching hospital 7 0.08 University hospital 6 0.06 Total 9,300 100.00 Source: World Bank based on GHS EmoNC survey. Note: CHPS = Community health planning services. Figure 42: Distribution of Ghana's Health Facilities by Category Source: World Bank staff based on DHIM2 data accessed in 2021. Note: CHAG = Christian Health Association of Ghana. 53 One important area for improvement is the category of emergency obstetric and newborn care (EmONC). Preliminary results from an EmONC survey 18 show that a significant proportion of health facilities do not meet the required quality standards to effectively deliver EmONC. The survey shows a wide disparity in the availability of EmONC services across regions, with some regions reporting no facilities that meet the required EmONC standards (Figure 43). The percentage of EmONC facilities nationwide meeting more rigorous EmONC criteria is below 20 percent. Figure 43: Percentage of Health Facilities Compliant with EmONC Criteria, by Region Source: EmONC survey 2021. Note: EmONC = Emergency obstetric and newborn care. A significant proportion of the health facilities report stockouts of essential medicines, including antiretroviral drugs, most commonly due to administrative bottlenecks and stockouts in the regional medical store (Figure 44). 18The Ghana Health Service conducted an EmONC survey in 2019–2020, but the full report was yet to be published at the time of concluding the PER. 54 Figure 44: Percentage of Stockouts at Health Facilities and Common Causes, Three-Month Period Source: EmONC survey 2021. Notes: ARVs = Antiretroviral. Most health facilities lack the full complement of human resources, equipment, supplies, and drugs required for comprehensive health service delivery. In most cases, particularly in primary health care facilities, essential equipment is unavailable or nonfunctional. According to the preliminary results of the 2021 EmONC survey, 19 only a third of EmONC facilities have functional equipment available such as suction catheter, disposable uncuffed tracheal tubes, mucus trap for suction, and other basic essentials (Figure 45). Figure 45: Availability of Functional Equipment in EmONC Facilities Source: EmONC survey 2021. The gap in resources extends to health care worker availability and is more prominent for specialist doctors than general nurses and midwives (Figure 46). For instance, there are only 171 pediatricians posted at government facilities compared to the targeted need for 338. The mismatch between need and specialist availability has a significant impact on health care utilization behaviors and rates. 19 Ghana Health Service (2021) Emergency Obstetrics and Newborn Care (EmONC) survey, draft report. 55 Figure 46:Maternal and Child Care Specialists, Actual vs. Target Source: EmONC survey 2021. Neonatologist Pediatrician General surgeon Obstetrician/Gynecologist 0 200 400 400 600 600 800 800 Available Available Target Target Service utilizaton of both in- and out-patient health services is concentrated at the primary care level. Institutional data on distribution of health services indicate high utilization of the primary health care level for hospital bed admissions and outpatient department (OPD) services. Whereas hospital admissions mostly occur at district hospitals, the subdistrict level (health centers, CHPS, clinics, and maternity homes) contributes significantly to OPD services. Private and faith-based facilities labeled “other hospitals” in the graphs are predominantly primary-level hospitals (Figure 47). Figure 47: National Distribution of Outpatient Visits per 1,000 Population by Type of Health Facility University/Teaching Regional hospital 3% Health centers, 2% District clinics, CHPS and 20% maternity homes 36% Polyclinics 3% Other hospitals 36% Source: World Bank staff, based on DHIMS data. Note: CHPS = Community health planning services. Ghana’s immunization utilization rates show sustained progress; childhood vaccination coverage is higher than with comparator countries. Since 2012, measles’ second dose vaccination coverage has steadily improved from 52 percent to 83 percent in 2021 (Figure 48). 56 This is, however, lower than the WHO-recommended target of ≥ 95 percent. Pneumococcal and DPT third dose coverage has remained high during the last three years, with coverage of 98 percent in 2021. Figure 48: Trends in Immunization Coverage for Childhood Vaccines, 2012–2021 100 80 60 40 20 0 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 DTP-containing vaccine, 3rd dose Measles-containing vaccine, 2nd dose Pneumococcal conjugate vaccine, final dose Source: WHO Immunization data. 57 5. FINANCIAL PROTECTION AND EQUITY Ghana reports a low incidence of catastrophic health expenditure. Ghana’s health system appears to be effective in preventing catastrophic health expenditures, as few households are allocating more than 10 percent or 25 percent of their entire income to health (Figure 49 and Figure 50). 20 Over the 1991–2017 period, less than 1 percent allocated more than 25 percent of their household income to health, and 7 percent allocated 10 percent of household income to health. Figure 49: Percentage of Catastrophic Health Expenditures in Ghana, 1991–2012 7 6 6 5 4.7 Percentage 4 3.1 3 2 1.1 1 0.6 0.8 1 0.5 0.1 0.1 0 1991 1998 2005 2012 2017 catastro_25% catastro_10% Source: World Bank Health Equity and Financial Protection Indicators. Figure 50: Comparison of Catastrophic Health Expenditure and Out-of-Pocket Payments 20 Catastrophic health expenditure is defined as out-of-pocket spending for health care that exceeds a certain proportion of a household’s income with the consequence that households suffer the burden of disease. 58 Sources: WHO Global Health Expenditure Database and World Bank Health Equity and Financial Protection Indicators. This level of financial protection is remarkable for a country like Ghana where OOP payments still represent close to 40 percent of CHE (Figure 50). In 2017, only 1 percent of households devoted more than 10 percent of their total household expenditures to health. Table 9 illustrates the decline in catastrophic health payments as the threshold increases. 21 The 3.4 percent of households that spend more than 5 percent of their total budget on health, spend an average of 9.7 percent. Whereas the 1 percent of households devoting more than 10.0 percent of their total household expenditures on health tend to spend an average of 16.7 percent. Table 9: Incidence of Catastrophic Health Payments by Quintile, 2017 5% 10% 25% Headcount (H) Lowest quintile 6.0 1.8 0.2 2 2.8 1.0 0.1 3 2.3 0.3 0.1 4 2.9 0.8 0.1 Highest quintile 3.1 1.2 0.0 Total 3.4 1.0 0.1 Source: Authors, using Adept on the 2017 GLSS. Note: Total expenditures were used as a proxy for household income. Households exceeding the catastrophic payment threshold tend to belong to the least wealthy quintiles. Table 9 shows that irrespective of the budget threshold, households incurring catastrophic payments by devoting more than 5, 10, or 25 percent of their total household budget to health care payments are in the least wealthy quintile. Importantly, the most recent comprehensive OOP data in Ghana are from 2017 and may not accurately reflect the current reality. The World Bank Findex survey 22 found that between 2017 and 2021, there was an increase (from 10 percent to 16 percent) in the number of households reporting the need to borrow money for health expenses, with an even larger increase among the lower quintiles. As previously discussed, the decrease in NHIA funding during recent years and the COVID-19 pandemic likely caused and may continue to cause higher (informal) OOP for patients seeking care. The planned implementation of GLSS-8 during 2023–24 should 21According to Pradhan and Prescott (2002); Wagstaff and Van Doorslaer (2003), [[AU: Pls. include full citations for both in References.]] (done) when total expenditures are used as the denominator to compute the incidence of catastrophic health expenditures, the most common threshold to use is 10 percent. The latter tends to represent the approximate threshold at which the household is being forced to sacrifice for basic needs, sell productive assets, incur debts, or become impoverished. However, the WHO has been using the threshold of 40 percent when “capacity to pay” or nonfood expenditures is being used as the denominator. 22 Demirgüç-Kunt et al. 2021. 59 provide new OOP payment data and help clarify the current state of catastrophic and impoverishing personal health expenditures in Ghana. Ghana has a mixed pattern of equity in health care utilization. For key child health services such as immunization, vitamin A, and care-seeking by mothers of sick children, there is remarkable equity across quintiles (Figure 51). However, the share of zero-dose children in the bottom 40 percent is double the number of children in the top quintile (3.9 percent vs. 1.9 percent), and the prevalence of stunting in the poorest quintile is almost triple the level of the richest quintile (24.5 percent vs. 8.5 percent). With regard to general health care seeking, the share of households consulting a public health care practitioner is similar from the second-poorest decile to the second-richest (between 3.5 and 4.0 percent), with only the bottom and top deciles as significant outliers. As expected, there is a significant gradient with regard to private health care (see Figure 52). Figure 51: Utilization of Key Child Health Services by Wealth Quintile Source: MICS 2017. Notes: BCG = Bacillus Calmette-Guerin; DPT3 = Third dose of diphtheria-pertussis-tetanus. Figure 52: Health Care Utilization by Facility Type and Wealth Quintile 60 Consulted any health practitioner 12% Percentage of population 10% 8% 5.5% 5.4% 6% 3.4% 3.8% 4.1% 3.8% 4.6% 4.4% 3.3% 4% 2.0% 2% 4.1% 3.7% 3.3% 3.5% 3.8% 3.5% 3.6% 4.5% 3.0% 3.6% 0% 1 2 3 4 5 6 7 8 9 10 Decile Public Private Alternative Source: GLSS-7, 2017. A closer look at utilization patterns within the public sector shows that richer households are more likely to seek care at hospitals, whereas poorer households tend to seek care at community health posts and health centers (Figure 53). Since unit costs are higher at hospitals, this implies a more unequal capture of health care subsidies by the better-off households than the raw utilization numbers imply. A benefit incidence study could explore this issue more closely. Figure 53: Choice of Health Service Provider by Wealth Quintile Provider choice among population consulting public 100% facilities 4% 6% 15% 16% 13% 18% 11% 26% 21% 15% 36% 22% 80% 19% 28% 39% 40% 29% 60% 40% 35% 32% 40% 74% 79% 70% 58% 53% 20% 39% 39% 46% 44% 32% 0% 1 2 3 4 5 6 7 8 9 10 Decile Public hospital Health Centre MCH clinic or CHPs Source: GLSS-7, 2017. Note: MCH = Maternal and child health; CHPS = Community health planning services. Gender and education attainment play an important role in health care utilization and the persistent inequity of poorer households. There are inequalities in access to antenatal care 61 and social health insurance coverage along dimensions of mother’s education and region/place of residence. Among pregnant women of childbearing age (15–49 years), those from poor households in both rural and urban areas and possessing low education levels have disproportionately lower access to antenatal care (see Table 10). Further, women with higher education levels and from wealthier households have greater coverage under any form of health insurance compared to those who are less educated and belong to less wealthy households. 23 Regarding equality in access to malaria eradication, the analysis suggests that irrespective of the place of residence and the level of education, enrollment in a malaria program was disproportionately concentrated among women from least wealthy households. However, the degree of inequality in access to malaria support programs was greater among women in rural areas and those possessing lower education. Table 10: Concentration Index on Maternal Health Outcomes, by Location and Mother's Education Maternal Concentratio Concentration Index value Health n Index value Urban Rural Urban Vs Greater Other Accra Vs Variables rural Accra health other differencesa regions regions differences Access to 0.038*** 0.025* 0.035*** -0.009 0.047** 0.037*** -0.010 antenatal care Ownershi 0.185*** 0.183*** 0.14*** 0.039 0.17*** 0.19*** -0.025 p of a social health insurance Advanced level of education Low level of education Differences (Primary 2 & secondary plus (No education or education school) primary 1) Access to antenatal 0.025** 0.0378*** -0.0125 care Ownership of a 0.16*** 0.082*** 0.080** social health insurance Source: Authors computation using 2017/2018 MICS data. Note: a. The statistical difference is computed under the large sample assumptions. 23 TheT-test under the large sample assumption rejects the null hypothesis that the index is the same among low and highly educated mothers. 62 7. CONCLUSIONS AND RECOMMENDATIONS 7.1 CONCLUSIONS Domestic resources for health are inadequately mobilized, allocated, and utilized. The country’s higher proportion of public spending, coupled with expansion of financing sources through the NHIS and effective population pooling contributes to financial protection and relatively good macroefficient health system performance compared to peer countries. Unfortunately, NHIS receives only a fraction of earmarked funds, which along with untimely fund release raise concerns about effective allocation. For example, after accounting for arrears payments, only 6.7 percent of the NHIS budget for 2021 was released within the budget year by the MoF. Given the reliance of health facilities on NHIS reimbursements, such practices affect health care provider capacity to deliver quality services and incentivize informal payments. Additionally, the parallel budget processes of the NHIA and MoH limit the ability of health care managers to allocate resources, in particular both salary and nonsalary recurrent costs, efficiently and effectively. There is limited incentive to improve health workforce efficiency due to the fragmented release of health worker compensation and lack of performance benchmarks. Additionally, health care providers are limited in their ability to efficiently allocate the mix of inputs required for delivery of quality health services. Over the past few years, Ghana has made substantial progress toward UHC, performing better than regional peers of comparable income level. However, the rate of improvement is slow and the average national health outcomes hide some disparities along various equity dimensions. Despite an increase in OOP payments for health services, Ghana has maintained low levels of catastrophic health spending among households. Regarding health system inputs, the supply of human resources for health (HRH) has increased over the years, although inequities in distribution persist. Similarly, the distribution of hospital beds varies across the country’s regions. While Ghana has made important strides in improving maternal and child health outcomes, disaggregation of key population health outcomes such as stunting and U5MR indicate inequality along dimensions of socioeconomic status, regions (Accra or other regions), gender, and place of residence (urban or rural). Similarly, access to preventive and curative health services for common childhood diseases as well as maternal health services like antenatal care is concentrated in wealthier households. Likewise, health insurance coverage of women and children is concentrated among wealthier households. The likely decline in DAH, lack of clear donor transition plan in government budgets, and squeezing of the fiscal space for health threaten health financing sustainability for UHC. Following the shock of the COVID-19 pandemic, Ghana’s economy was expected to rebound but progress has been slow, and the Ukraine-Russia war has further impacted the early post-COVID- 19 gains. With assistance from development partners, government health spending increased in response to the COVID-19 crisis but is expected to decline to restore fiscal balance. Domestic health resource mobilization is below the average for SSA countries and LMICs and opportunities have been identified to further improve revenue collections in Ghana. The share of government spending on health has fluctuated significantly over the last two decades and is declining, indicating the need to reprioritize health in the government budget. Interest payments constitute a growing proportion of the government budget, squeezing fiscal space for health and other discretionary spending on social determinants of health. The ongoing COVID-19 pandemic has created limited opportunities to prioritize health spending in the government budget. Ghana has been able to mobilize significant DAH, which is likely to decrease as the country transitions to middle-income country status. However, there are opportunities in the short to medium terms to 63 improve coordination and alignment of external assistance and reduce volatility and fragmentation of external funds. It will be important for the country to focus on strategies to improve the allocation of existing resources. There are inefficiencies in the health sector budget planning and expenditure processes that need to be addressed by implementing resource mapping and expenditure tracking interventions. MoH expenditures outpace its budget, raising concerns of planning and budgeting inefficiencies. Whereas domestic resources are fully accounted for in the planning and budget processes, a significant portion of donor funds are implemented outside of the government system. There appears to be limited appetite or capacity to incorporate donor funding into the MoH’s budget and planning process. Although domestic funding sources constitute the greatest share of the MoH budget, most of it goes toward workforce compensation, resulting in reliance on DAH and IGF to finance capital expenditure and goods and services. This model results in limited health sector discretionary spending, particularly at the district and subdistrict levels. Fragmented budgeting and expenditure processes at the individual donor level, irregular use or absence of financial planning tools, and the piecemeal implementation of an integrated financial management information system diminish the transparency of health sector resource availability and allocation. The health sector has adopted program-based budgeting, but the MoH budget is still input- and line item–based. This limits the ability of the sector managers to allocate resources to specific program objectives. The low NHIS medical loss ratio raises concerns of significant administrative inefficiency and provides justification to improve strategic purchasing. The rising operational cost of the NHIS needs to be monitored and managed. NHIS’s financial sustainability appears to have improved recently with revenues exceeding expenditure over a four-year period, not considering arrears payments. However, increasingly NHIS’s revenues are being used to finance nonclaim expenditures. With recent cost containment interventions such as claims auditing and expansion of electronic claims processing, the declining share of claims expenditure could mean decreasing utilization of health care services by members. In addition, the current payment mechanism excludes cost-effective preventive and promotive health services that could be actuarially examined and included in the NHIS benefits package, particularly in light of the increasing share of NCDs in Ghana’s disease burden. There is a policy priority and financing mismatch as frontline primary health care services are less prioritized in the government budget despite being a policy priority in Ghana’s UHC roadmap. The major sources of government funding to the health sector disadvantage frontline PHC facilities. Lower-level facilities that focus on preventive and promotive health services receive less funding from the NHIS since the current payment mechanism is activity- based and curative-care oriented. At the same time, the subdistrict receives a small share of public funds through the MoH budget for its operations. The absence of strategic purchasing and population-based payment mechanisms that incentivize delivery of preventive, high-quality care services exacerbate the situation. PHC relies heavily on DAH funds, which to a large extent are managed off-budget. Off-budget modalities limit MoH/GHS capacity to allocate and utilize health funds critical to financing PHC goods and services, resulting in higher transaction costs and inefficiencies. While key health sector DPs are managing their transition, there are opportunities for DPs and the government to better manage the transition by adopting programmatic approaches such as results-based financing to shift attention to results rather than financing of inputs. 64 7.2 RECOMMENDATIONS Enhance allocation and transfer of earmarked funds to the health sector. The MoH and NHIA, with the support of civil society and DPs, should collaborate with the MoF for timely allocation and release of earmarked resources to the health sector. Impact analysis of low and delayed fund release from MoF to NHIA for health care delivery can be used to initiate the dialogue with MoF. It is imperative that the engagements focus on both the level and timeliness of fund release. The dialogue should aim to reach a consensus among parties to clear arrears owed to providers and to abide by the scheme’s minimum reimbursement waiting time of three months. Further, MoH and NHIA should engage with the MoF on implementation plans for the COVID-19 levy introduced in 2021, since it potentially increases the fiscal space for the health sector. Scale up implementation of public financial management tools to enhance consolidated budget planning and reporting, including accounting for DAH. A more transparent process of on-budget reporting will improve efficiency in planning and executing resources by reducing double-counting and underreporting (off-budget); and in turn, reduce fragmentation of DAH. The MoH, GHS, and NHIA should ensure regular production of the National Health Accounts; publish timely annual financial reports; review the 2015 Health Financing Strategy, revise the Accounting, Treasury, and Finance manual and revise the guidelines for use of IGF; leverage the country’s “Beyond Aid” strategy and develop an implementation plan with measurable goals and time lines to transition out of external aid; leverage GFF-support for resource mapping and expenditure tracking to strengthen resource mobilization functions through use of Resource Mapping and Expenditure Tracking (RMET) tools; and pursue deployment of GIFMIS to at least district levels of health care in the short term and to subdistrict levels in the medium term. To enable effective resource mapping and expenditure tracking, the resource-mapping tools should be integrated or be made interoperable with the GIFMIS. The MoH should accelerate the development and implementation of the Health Information Management Strategy, prioritizing activities aimed at integration of health information systems in the sector. To enhance the effective and efficient management of health facilities, a Financial Management and Accounting Information System (FMAIS) should be deployed for management and monitoring of allocation of inputs at the facility level vis-à-vis their outputs. Accelerate NHIS coverage by prioritizing the poor and indigent, as well as explicitly including cost-effective PHC services, particularly preventive health services in the essential benefits package. This will enhance the pooling function of NHIS and prioritize financial protection of vulnerable subpopulations. Efforts should include identifying barriers to registering and renewing membership of the indigent population through interventions such as linking the NHIS database to the National Household Registry, collaborating with the Ministry of Gender, Children, and Social Protection, and focusing resources on enhancing enrollment in regions that are lagging behind. Additionally, reduction in NHIS claims expenditure should be investigated to ensure that it is not the result of reduced health services utilization by members. The prevalence of informal copayments from members and nonmembers should be investigated and steps taken to minimize these. In the face of low medical loss ratio, the NHIS should revisit its benefit package as well as tariffs, and collaborate with health service delivery agencies to improve utilization of the existing benefit package. The proportion of NHIS expenditure on claims is declining. The NHIS should reexamine its expenditures and track its operational expenditures, particularly salaries. It should also examine its capacity to purchase additional services following actuarial analysis and 65 gradually shift expenditures toward claims payments as opposed to operational overhead. A dialogue between NHIS and MoH should be undertaken regarding payment for vaccines and essential commodities. The NHIS should collaborate with service delivery agencies to streamline the claims process. Delays in submission and poor quality of claims contribute significantly to delays in NHIS claims processing and payment. The NHIS has expanded its operational capacity including an increase in its number of staff. The NHIA should use its regional and district staff strength to support the GHS and other providers to improve the claims generation and submission process. This could include reposting excess NHIS staff to health facilities to support claims management or provide supportive supervision to service providers. Revisit NHIS payment methods for PHC services toward population-based modalities and strategic purchasing. A blended payment system with population-based mechanisms at its core would provide better value for money and align provider incentives with population health needs and outcomes. Currently, fee-for-service (FFS) and diagnostic-related groups (DRGs) promote productivity; however, population-based payment systems such as capitation or cost-sharing arrangements that vary across levels of care should be piloted for implementation feasibility, as recommended in the 2017 PER. This is likely to promote efficiency in service utilization with a potential shift to preventive services. With the introduction of the NoP service delivery arrangement, the need for an alternative payment mechanism at the primary level has become even more relevant. Strengthen PHC service delivery system, particularly at the subdistrict level and below, by enhancing allocation of resources to frontline facilities and developing networks of practice (NoPs). This includes increasing the MoH budgetary allocation to PHC facilities and shifting IGF resources from district hospitals and health centers toward preventive and promotive services at the district level. It will also require reviewing the Ghana Essential Health Service Package (GEHSP) and aligning it with the NHIS benefits package. To further support PHC service delivery, cost-effective PHC services should be actuarially tested for inclusion in the NHIS benefits package. Additionally, this would include ensuring availability of supplies, drugs, and diagnostics at the subdistrict facilities through the scale-up of the Ghana Integrated Logistics Management Information System; training the subdistrict-level workforce to submit NHIS claims; strengthening referral networks to address inefficiencies at the hospital level; and enhancing availability of and capacity to utilize MoH discretionary funds at the subnational level for need-based operational and strategic costs. Collaboration with the Health Facilities Regulatory Agency will be required to initiate steps toward accreditation and credentialing of health facilities under NoP. 7.3 LIMITATIONS A key challenge encountered in the preparation of this study was the lack of readily available data and information on budget and expenditure. Notably, Ghana has not completed and published its National Health Accounts (NHA) since 2016. The NHA serves as the primary data source for many global health-financing databases, especially the WHO Global Health Financing Database. The absence of an updated NHA since 2017 undermines the credibility of Ghana’s health financing data in the global health expenditure database and hence compromises the reliability of the global benchmarking of Ghana’s performance in health financing. It also makes it challenging to conduct a comparative analysis of Ghana with other countries. Despite being a public institution, the NHIS has not published an annual report since 2014. The absence of an up-to- 66 date annual reports of the NHIA resulted in significant delays in the acquisition and validation of data provided by the NHIS. Similarly, the team could not conduct a detailed HRH analysis due to the lack of readily available data. As salaries of the health workforce constitute the biggest chunk of public health sector expenditure, the team intended to perform a health workforce efficiency analysis. However, limited availability of data on HRH performance limited the team’s ability to conduct a meaningful HRH efficiency analysis. It is recommended that a HRH productivity analysis be conducted to identify key HRH efficiency gaps in Ghana. Stakeholder consultations highlighted recommendations to include COVID-19 expenditures in the paper. However, the team did not incorporate this recommendation due to the lack of readily available data and the fact that the WHO and UNICEF are conducting COVID-19 expenditure reviews, which could complement this paper. 67 8.0 ANNEX Annex A. UHC SCI Detailed Methodology of Computation Source:UHC2030 report Notes: UHC = Universal health coverage; SCI = Service Coverage Index; NCD = Noncommunicable disease; TB = Tuberculosis; HIV = Human immunodeficiency virus; ART = Antiretroviral; WASH = Water, sanitation and hygiene; IHR = International health regulation. 68 REFERENCES Abiiro, G. A., K. A. Alatinga, and G. Yamey. 2021. “Why Did Ghana’s National Health Insurance Capitation Payment Model Fall Off the Policy Agenda? A Regional Level Policy Analysis. Health Policy and Planning. Abor, Patience Aseweh, and Joshua Yindenaba Abor. 2020 "Implications of COVID-19 Pandemic for Health Financing System in Ghana." Journal of Health Management 22.4: 559–69. Adua, E., K. Frimpong, X. Li, and Wang, W. 2017. "Emerging Issues in Public Health: A Perspective on Ghana’s Healthcare Expenditure, Policies and Outcomes." EPMA Journal 8.3: 197–206. Akweongo, P., Aikins, M., Wyss, K. et al. 2021 Insured clients out-of-pocket payments for health care under the national health insurance scheme in Ghana. BMC Health Serv Res 21, 440 (2021). https://doi.org/10.1186/s12913-021-06401-8. Agyei-Mensah, S. and Aikins, A.D.G., 2010. Epidemiological Transition and the Double Burden of Disease in Accra, Ghana. Journal of Urban Health 87 (5): 879–97. Ango, P., M. Aikins, K. Wyss, P. Salari, and F. Tediosi. 2021. “Insured Clients’ Out-of-Pocket Payments for Health Care under the National Health Insurance Scheme in Ghana.” BMC Health Services Research 21 (1): 1–14. Awoonor-Williams, J. K., P. Tindana, P. A. Dalinjong, H. Nartey, and J. Akazili. 2016. “Do the Operations of the National Health Insurance Scheme (NHIS) in Ghana Align with the Goals of Primary Health Care? Perspectives of Key Stakeholders in Northern Ghana.” BMC International Health and Human Rights 16 (1): 1-11. World Bank. 2022. Pandemic Preparedness Assessment in Ghana. Boachie Kofi, Micheal, K. Ramu, and Tatjana Põlajeva. 2018. "Public Health Expenditures and Health Outcomes: New Evidence from Ghana." Economies 6.4: 58. Bokhari F, Y. Gai, and P. Gottret. 2007. “Government Health Expenditures and Health Outcomes.” Health Econ 16: 257–73. Cashin, C., S. Sparkes, and D. Bloom. 2017. “Earmarking for Health: From Theory to Practice” (No. WHO/HIS/HGF/HFWorkingPaper/17.5). Geneva: World Health Organization. Dieleman, J., M. Campbell, A. Chapin, E. Eldrenkamp, V. Y. Fan, A. Haakenstad, J. Kates, et al. 2017. “Evolution and Patterns of Global Health Financing 1995–2014: Development Assistance for Health, and Government, Prepaid Private, and Out-of-Pocket Health Spending in 184 countries.” The Lancet 389 (10083): 1981–2004. Etienne, Ghanaissa, Anarfi Asamoa-Baah, and David B. Evans. 2010. Health Systems Financing: The Path to Universal Coverage. Geneva: World Health Organization. Farahani, M., S. V. Subramanian, and D. Canning. 2010. “Effects of State-Level Public Spending on Health on the Mortality Probability in India.” Health Econ 19: 1361–76. Heller, P. 2006. “The Prospect of Creating ‘Fiscal Space’ for the Health Sector.” Health Policy and Planning 21 (2): 75–79. 69 Fan, Victoria Y., and William D. Savedoff. 2014. "The Health Financing Transition: A Conceptual Framework and Empirical Evidence." Social Science and Medicine 105: 112–21. Filmer, D., and L. Pritchett. 1999. “The Impact of Public Spending on Health: Does Money Matter?” Soc Sci Med 49: 1309–23. Fusheini, Adam. 2016. "The Politico-Economic Challenges of Ghana’s National Health Insurance Scheme Implementation." International Journal of Health Policy and Management 5. 9: 543. Gaddah, Mawuli, and Alistair Munro. 2011. "The Progressivity of Health Care Services in Ghana." National Graduate Institute for Policy Studies (GRIPS) Discussion Paper 11–14. Global Financing Facility 2021. Monitoring Essential Health Services in Times of COVID-19. Ghana July 2021 Update. Washington, DC: GFF Ghana Statistical Services (GSS). 2018. Ghana Living Standards Survey 7 (GLLS7) Main Report. Accra: GSS. ———. 2019. Multiple Indicator Cluster Survey (MICS2017/18), Survey Finding Report. Accra: GSS. Ghana Statistical Services (GSS), Ghana Health Service (GHS), and ICF. 2018. Ghana Maternal Health Survey 2017. Accra, Ghana: GSS, GHS, and ICF. Ghana Demographic and Health Survey 2022 Key Indicators Report Ghana Statistical Service Accra, Ghana The DHS Program ICF Rockville, Maryland, USA May 2023 Ghana Statistical Services (GSS), Population and Housing Census, 2021. PHC Report, 2021. Green, C., B. Hollingsworth, and M. Yang. 2021. “The Impact of Social Health Insurance on Rural populations.” Eur J Health Econ 22: 473–83. https://doi.org/10.1007/s10198-021-01268-2. IFC (International Finance Corporation). 2018. Ghana Private Health Sector. Summary Sub-Sector Deep- Dives. Washington, DC. Kittoe, J. D., and S. K. Asiedu-Addo. 2017. “Exploring Fraud and Abuse in National Health Insurance Scheme (NHIS) Using Data Mining Technique as a Statistical Model.” African Journal of Educational Studies in Mathematics and Sciences 13: 13–31. Demirgüç-Kunt, Asli, Leora Klapper, Dorothe Singer, Saniya Ansar. 2021. “Financial Inclusion, Digital Payments, and Resilience in the Age of COVID-19.” The Global Findex Database 2021. Kurowski, Christoph, David B. Evans, Ajay Tandon, Patrick Hoang-Vu Eozenou, Martin Schmidt, Alec Irwin, Jewelwayne Salcedo Cain, et al. 2021. “From Double Shock to Double Recovery— Implications and Options for Health Financing in the Time of COVID-19 (English).” Health, Nutrition, and Population (HNP) Discussion Paper. Washington, DC: World Bank Group. Langenbrunner, J., C. Cashin, and S. O'Dougherty, eds. 2009. Designing and Implementing Health Care Provider Payment Systems: How-to Manuals. World Bank Publications. Ma, X., and T. Oshio. 2020. The Impact of Social Insurance on Health among Middle-Aged and Older Adults in Rural China: A Longitudinal Study Using a Three-Wave Nationwide Survey.” BMC Public Health 20, 1842. https://doi.org/10.1186/s12889-020-09945-2. Makinen, Marty, Stephanie Sealy, RiGhanado A. Bitrán, Sam Adjei, and Rodrigo Muñoz. 2011. Private Health Sector Assessment in Ghana. Working Paper 210. Washington, DC: World Bank Group. 70 Mba C. J. 2010. “Population Ageing in Ghana: Research Gaps and the Way Forward.” Journal of Aging Research 672157. https://doi.org/10.4061/2010/672157. McCracken, K. and D. R. Phillips. 2016. Demographic and Epidemiological Transition. International Encyclopedia of Geography: People, the Earth, Environment and Technology: People, the Earth, Environment and Technology, 1–8. Ministry of Health of Health Ghana. 2023. Annual Program of Work. https://www.moh.gov.gh/wp- content/uploads/2023/07/2023-POW.pdf Miracolo, A., M. Sophiea, Mills, and P. Kanavos. 2021. “Sin Taxes and Their Effect on Consumption, Revenue Generation and Health Improvement: A Systematic Literature Review in Latin America.” Health Policy and Planning 36 (5): 790–810. MOH (Ministry of Health). 2020. Ghana’s Roadmap for Attaining Universal Health Coverage 2020–2030. Accra, Ghana: Ministry of Health. MOH, and Oxford Policy Management. 2018. Ghana Health Roadmap for Sustainability and Transition from External Finance. Accra, Ghana: Ministry of Health. Moreno-Serra, R., and P. C. Smith. 2012. “Does Progress towards Universal Health Coverage Improve Population Health?” The Lancet 380 (9845): 917–23. Moreno-Serra, R., and P. C. Smith. 2015. “Broader Health Coverage is Good for the Nation's Health: Evidence from Country Level Panel Data.” Journal of the Royal Statistical Society. Series A (Statistics in Society) 178 (1): 101. Nsiah-Boateng, E., F. Asenso-Boadi, L. Dsane-Selby, F.-X. Andoh-Adjei, N. Otoo, P. Akweongo, and M. Aikins. 2017. “Reducing Medical Claims Cost to Ghana’s National Health Insurance Scheme: A Cross-Sectional Comparative Assessment of the Paper- and Electronic-Based Claims Reviews.” BMC Health Services Research 17 (115). DOI 10.1186/s12913-017-2054-1. Okoroh, Juliet, Samuel Essoun, Anthony Seddoh, Hobart Harris, Joel S. Weissman, Lydia Dsane-Selby, and Robert Riviello. 2018. "Evaluating the Impact of the National Health Insurance Scheme of Ghana on Out of Pocket Expenditures: A Systematic Review." BMC Health Services Research 18.1: 1–14. Osei, D., S. Sparkes, K. A. Frimpong, and S. Sapatnekar. 2021. Implementing Programme Based Budgeting in Ghana’s Health sector. (Health Financing Case study 18. Budgeting in Health). Geneva: World Health Organization. Pradhan M, Prescott N. Social risk management options for medical care in Indonesia. Health Econ. 2002 Jul;11(5):431-46. doi: 10.1002/hec.689. PMID: 12112492 Saleh, Karima. 2012. The Health Sector in Ghana: A Comprehensive Assessment. Washington, DC: World Bank. Savedoff, W. D., D. de Ferranti, A. L. Smith, and V. Fan. 2012. “Political and Economic Aspects of the Transition to Universal Health Coverage.” The Lancet 380: 924–32. Schieber, G., C. Cashin, K. Saleh, and R. Lavado. 2012. Health Financing in Ghana. Washington, DC: World Bank. Starfield, B., L. Shi, and J. Macinko. 2005. “Contribution of Primary Care to Health Systems and Health.” The Milbank Quarterly 83 (3): 457–502. 71 Tandon, A., and C Cashin. 2010. “Assessing Public Expenditure on Health from a Fiscal Space Perspective.” Health, Nutrition, and Population Discussion Paper. Washington, DC: World Bank. Tsimpo, C., and Q. Wodon. 2012. “National Health Insurance Scheme in Ghana.” In Improving the Targeting of Social Programs in Ghana, ed. Quentin Wodon. Washington, DC: World Bank. Vos, T., S. S. Lim, C. Abbafati, K. M. Abbas, M. Abbasi, M. Abbasifard, M. Abbasi-Kangevari, et al. 2020. “Global Burden of 369 Diseases and Injuries in 204 Countries and Territories, 1990–2019: A Systematic Analysis for the Global Burden of Disease Study 2019.” The Lancet 396 (10258): 1204–22. Wagstaff A, van Doorslaer E. Catastrophe and impoverishment in paying for health care: with applications to Vietnam 1993-1998. Health Econ. 2003 Nov;12(11):921-34. doi: 10.1002/hec.776. PMID: 14601155. Wang, Huihui, Nathaniel Otoo, and Lydia Dsane-Selby. 2017. Ghana National Health Insurance Scheme: Improving Financial Sustainability based on Expenditure Review. Washington, DC: World Bank. WHO (World Health Organization). 2010. Health Systems Financing: The Path to Universal Coverage. Geneva. ———. 2019. Global Spending on Health: A World in Transition. WHO/HIS/HGF/HFWorkingPaper/19.4. Geneva. ———. 2020. Ghana Country Assessment. Data Collection Template Web Annex. In The Health Financing Progress Matrix: Country Assessment Guide. Geneva. ———. 2021. Ghana: Cross-Programmatic Efficiency Analysis. Health Financing Case Study 11. Geneva. Wodon, Q., ed. 2012. Improving the Targeting of Social Programs in Ghana. Washington, DC: World Bank. World Bank. 2014. Raising Additional Government Revenues in Ghana by Raising the Excise Tax on Tobacco and Alcohol. Washington, DC. ———. 2017a. Fiscal Consolidation to Accelerate Growth and Support Inclusive Development: Ghana Public Expenditure Review. Washington, DC. ———. 2017b. Republic of Ghana GH: Public Financial Management Bottlenecks to Service Delivery. Washington, DC. ———. 2020. Ghana: Enhancing Revenue Mobilization through Improved Tax Compliance and Administrative Systems. Washington, DC. ———. 2021. Ghana 2021 Health Budget Highlights. Ghana HNP Task Team. Mimeo. Washington, DC. Wright, A., K. E. Smith, and M. Hellowell. 2017. “Policy Lessons from Health Taxes: A Systematic Review of Empirical Studies.” BMC Public Health 17 (1): 1–14. Yackovlev, I., Z. Murgasova, F. Liu, G. Minasyan, and K. Wang. 2020. “How to Operationalize IMF Engagement on Social Spending during and in the aftermath of the COVID-19 Crisis.” IMF How To Notes 2020 (002). Washington, DC: International Monetary Fund. Zhang, Cherri, M. S. Rahman, M. M. Rahman, A. E. Yawson, and K. Shibuya. 2019. "Trends and Projections of Universal Health Coverage Indicators in Ghana, 1995–2030: A National and Subnational Study." PloS One 14.5: e0209126. 72 73 Ghana has made substantial progress toward universal health coverage (UHC) and compares relatively well in health outcomes among its economic peers. International comparisons of government current health expenditure suggest that Ghana spends less than the sub-Saharan African norm for its income level. That notwithstanding, Ghana compares favorably with its peers in key health outcomes with a mixed pattern of equity in health service utilization. The trajectory indicates that better health outcomes could be achieved if health spending increases. Public funding through the Ministry of Health’s (MoH’s) budget and the National Health Insurance Scheme (NHIS) constitute the major sources of financing for the health sector in Ghana.. Total government health spending in Ghana almost trippled in nominal terms in 2015-20, but it rose by a modest 12 percent in real terms, and nearly all the increased spending was channeled through MoH instead of through NHIA. On average, 80 percent of the government budget is channeled through the MoH budget, while the share of NHIS spending has declined over the same period from 24 percent to 12 percent. There are preliminary signs that past achievements in financial protection largely attributable to the impact of the NHIS are being eroded. Although the social contract underpinning the establishment of the National Health Insurance Levy is not in question, the recently introduced Statutory Funds Capping Law further threatens the sustainability of the NHIS. There is a need to improve efficiency in the functional allocation of domestic health resources and upscale the use of public financial management (PFM) tools at the national and subnational levels to enhance budget credibility and expenditure tracking. To ensure no one is left behind, deliberate efforts should be made to improve equity in health care utilization and outcomes along income and regional dimensions. ABOUT THIS SERIES: This series is produced by the Health, Nutrition, and Population Global Practice of the World Bank. The papers in this series aim to provide a vehicle for publishing preliminary results on HNP topics to encourage discussion and debate. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations or to members of its Board of Executive Directors or the countries they represent. Citation and the use of material presented in this series should take into account this provisional character. For free copies of papers in this series please contact the individual author/s whose name appears on the paper. Enquiries about the series and submissions should be made directly to the Editor Jung-Hwan Choi (jchoi@worldbank.org) or HNP Advisory Service (askhnp@worldbank.org). For more information, see also www.worldbank.org/hnppublications. 1818 H Street, NW Washington, DC USA 20433 Telephone: 202 473 1000 Facsimile: 202 477 6391 Internet: www.worldbank.org E-mail: feedback@worldbank.org