51468 Empowerment Case Studies: Public Expenditure Tracking Surveys PUBLIC EXPENDITURE TRACKING SURVEYS ­ APPLICATION IN UGANDA, TANZANIA, GHANA AND HONDURAS I. Abstract Public Expenditure and Tracking Surveys (PETS) was launched in order to expose the anomalies in the process of flow of funds and to gauge the degree to which funds trickle down to their intended destinations. PETS consolidates information from frontline Service Providers who can impart useful insight into and the reasons for the disparity in the outcome of public spending (quantity or input) and quality and performance (output) of service delivery systems. Such information also provides a platform for enquiry into leakage of funds due to corruption or private gain. PETS has a multilevel focus with the service providers being the key unit of analysis. PETS was first introduced in Uganda in 1996. Thereafter a number of countries have implemented it. This case study highlights the application of PETS in Uganda, Tanzania, Ghana and Honduras. It analyses the linkage between public spending and development outcomes, the factors responsible for the emergence of inefficiencies in the system, and the role of the culture of accountability and transparency in fostering a sustainable alliance among the different institutional strata. Finally, it outlines the impact of such an initiative. II. Background: PETS, an exemplar of a diagnostic survey, is designed to provide crucial information that can help policy makers in either policy formulation or reformatory activities by identifying gaps in allocation and disbursement of funds to the intended recipients. Unlike evaluation and monitoring tools such as household surveys and enterprise surveys which critically examine the impact of interventions or policies on households and enterprises respectively, PETS focuses on the service provider as the key unit of analysis. The prime objective of the surveys is to track and evaluate the efficiency and effectiveness of public expenditures especially in the rural areas, identify the areas of concern and the impediments in resource flows. In addition, PETS also aims to assess and quantify improvements in the quality of service delivery at facility and community levels. There was a need to employ this diagnostic tool because despite the fact that government budgetary allocations were public, there was unavailability of information on actual public spending. This made it difficult to refute the logic that increasing allocation of funds for different sectoral development does not, in any way, guarantee improved quality and performance of those sectors. This could be explained by the fact that the allocated funds may not permeate though the different institutional strata to reach the intended facilities. By resorting to this tool, answers to questions like ­ were the funds spent as planned; was the expenditure effective and to what degree; and what has been the performance of service delivery in terms of quality? Using PETS becomes even more necessary when there are weak government institutions, unreliable data and poor Prepared by Parameeta Kanungo at the World Bank, Washington D.C. My sincere thanks go to Abigail Somma for her useful insights and valuable feedback. 1 Empowerment Case Studies: Public Expenditure Tracking Surveys reporting on execution, all of which inhibit the ability to gauge the efficacy of public spending and service delivery systems. In Uganda, PETS was deployed in the education and health sector. In Tanzania (1999 and 2001) PETS tracked pro-poor expenditures in priority sectors at all levels. In Ghana (2000) expenditure tracking was done based on data collected at the service facility, district and central level. In Honduras (2000) the objective of PETS was to diagnose and quantify ghost workers, absenteeism and job migration and the reasons that contributed to such phenomena. The primary objective in all the cases was to quantify "asymmetric information" in public spending. The results obtained from such surveys confirmed that leakage of non-wage public funds hinders the process of service delivery. UGANDA Uganda was the first country to implement PETS in 1996. The motivation for such an effort was the fact that though the official reports indicated that there was a substantial increase in public spending on education, primary school enrollment did not show any improvement. PETS focused on primary education and health care in its effort to compare budget allocation to actual spending through the different levels of government, including frontline service delivery units in both primary schools and clinics. Due to lack of adequate and sufficient public accounts to reflect actual spending, surveys were conducted in 19 districts (out of 39), 250 government primary school and 100 health clinics. These contributed to a five-year panel data on provider characteristics, spending (including in-kind transfers) and outputs. The school survey indicated that on average, only 13% of per-student non-wage funds provided by the central government reached the schools during 1991-1995. Eighty seven percent was misused either for personal gain or for purposes not intended for education. Roughly 70% of the schools did not receive anything. If the schools received some funds, it was a negligible amount. When we limit the survey results to the last year of the survey period (i.e. 1995), while improvements can be seen (increase of 7% of the capitation funds), only 20% of the central government capitation grants were estimated to have reached the schools. Yearly data reported less than 5% receipt of grants by 73% of schools while 10% of the schools received more than 50% of the intended funds. It was found that parents contributed about 73% of total school spending in 1991. In spite of increase in government share during the survey period, parents, on an average, still funded 60% of total primary funding by 1995. This trend seemed to continue despite higher public spending. Evaluation of the reforms, about 5 years later using two locally implemented follow-up PETSs showed that there that been a substantial improvement in the flow of funds reaching the schools - from 13% (on average) in 1991-95 to about 80 to 90 percent in 1999 and 2000. 2 Empowerment Case Studies: Public Expenditure Tracking Surveys TANZANIA Two public expenditure tracking surveys were implemented in Tanzania in 1999 and 2001 as a test of the hypothesis that there existed serious problems in the flow of funds from the central government through the local authorities to frontline service units. In the first Tanzanian PETS 45 primary schools and 36 health facilities in three districts was considered. The survey indicated diversion of a large portion of funds (about 57 percent in education and 41 percent in health care ) disbursed by the center for non-wage education and health expenditures to other non education sectors and also for personal gain. Though salaries were less prone to diversion, payrolls experienced the existence of ghost workers and there were delays in pay for the frontline staff. The second PETS had a similar focus as the first one (on health and education). However, additional information was collected that included pro-poor expenditures for rural water supply, rural roads, judiciary and HIV/AIDS. The survey was implemented in four primary schools and four clinics in each of the five selected districts. It was found that there were substantial delays in disbursement of funds at all levels of the government. Delays were more pronounced in case of non-wage expenditures and in rural areas. Rural district also suffered from receiving smaller shares of the intended resources than their urban counterparts. This could be explained by factors like cash budgeting that led to transfers on account due to fluctuations in revenue. Consequently, this led to the emergence of information asymmetries which prevented the beneficiaries from knowing the amount of their monthly allocation or entitlement. Information asymmetries proved beneficial for the council staff with respect to their dealings with the service facilities as the facilities did not have any basis to demand their entitlements. Moreover, government records also undermined the importance of transparency in public spending. It was observed that although the findings of the two PETS were disseminated in the national budget consultations, their impact in terms of transparency mechanisms in the central government, was not seen to be as significant as in case of Uganda. The treasury has adopted the strategy of circulating itemized local government budgets to Parliament members as well as publishing budget allocations aimed at pro-poor programs in both Swahili and English language papers. This provided information of budget allocations at the ministerial, regional and local authorities levels. GHANA : The rationale for employing PETS in Ghana in 2000 was similar to that of Uganda and Tanzania in intent (evaluate actual spending and appraise leakage of funds in the transfer from government through districts to service facilities) as well as sectors (basic education and primary health care). The Ghana PETS covered four districts in each of the ten regions. It included interviews of 40 district education officers and 40 district health officers, and a facility­level survey of a total of 119 primary schools, 79 junior secondary schools and 173 primary health 3 Empowerment Case Studies: Public Expenditure Tracking Surveys clinics. The results of this initiative indicated that 20% of non-wage public health expenditure and 50% of nonwage education expenditure reached the frontline facilities. There seemed to be a lesser proportion of leakage in salaries (about 20%) as compared to Uganda and Tanzania whereas, as opposed to the other two countries, a large proportion of leakage was observed to lie between line ministries and district offices during translation of public expenditure funds into in-kind transfers. Additionally, it was noted that there seemed to be greater leakage when the distributed material values are not known to the recipients. The transfers being in-kind by nature, were responsible for the emergence of information asymmetries and lack of accountability within the delivery system. Lack of information pertaining to funds allocation inhibited the frontline facilities from providing their feedback in terms of their resource needs or lodging complaints. PETS has been successful in initiating a inter-ministerial collaboration in Ghana and has been able to evaluate frontline expenditures and service delivery. In spite of efforts toward increasing transparency and strengthening oversight functions, they have not been effective in hastening the process aimed at leakage reduction. It can be assumed that this being the first attempt as well as a pilot study, has not shown any significant outcomes but there are reasons to believe (based on the experiences of Uganda and Tanzania) that successive implementations would yield better reformatory outcomes. HONDURAS: In Honduras PETS was employed to identify moral hazards related to frontline health and education staff. The belief was that the central payroll office in Honduras had no way to verify whether public employees actually existed (ghost workers) and whether they were actually working where they were supposed to work (migration of posts). PETS was used as a response to mitigate the problems arising due to the post being assigned to the central ministry instead of individual facilities. Assigning posts to central ministry provided avenues for frontline staff to lobby the ministers and have their posts transferred to urban or more attractive locations. Therefore, application of PETS was proposed to measure the incongruity between budgetary and real assignments of staff and to determine the degree of attendance at work. Data were obtained from government sources (govt. payroll data indicated employee's place of work) and a nationally representative sample of frontline facilities in health and education was used. Unlike the other PETS application, the unit of analysis was the sector staff (both at the operational and administrative and from all levels of the two sectors from ministry to the service facility level) instead of the facility. In the health sector, it was observed that 2.4 percent of the payroll were ghost workers. General practitioners and specialists which accounted for 8.3% and 5.1% respectively stood out. Absenteeism was significantly high with an average attendance rate of 73% across all categories of staff (this means that in the 5 days prior to the survey, only 73% 4 Empowerment Case Studies: Public Expenditure Tracking Surveys of the staff were at work). Thirty-nine percent of absences was attributed to non- justifiable reasons (that is reasons other than sick leave, vacations, compensation for extra hours worked). There was prevalence of multiple jobs for general practitioners and specialists : fifty ­four percent of specialist physicians had two or more jobs of which 60% were in related fields. Moreover, 5.2% of the staff had drifted to posts other than the ones to which they were assigned in the central database whereas 40% had moved since their first assignment. In the education sector, ghost workers accounted for 3% of the staff members while 5 percent of the primary school teachers were unknown in their work place. As compared to health sector, absenteeism was lower in this case with an average attendance rate of 86% across all categories of staff and 15% of all absences being unaccounted for. Since PETS was carried out relatively recently, there as been little follow-up of the findings within the government. III. Impact/Results The information disclosure mechanism provided a sound base for the frontline service units to demand their entitlements from the district officials. As a result, they witnessed a significant reduction in capture and leakage of funds. Information campaign evaluation results (by using a repeat PETS) indicated a significant improvement. Though the entire fund was not received by schools and there were delays, the capture rate dropped from an average of 80% in 1995 to 20% in 20011. Additionally, pre and post assessment (1995 and 2001)of outcomes of the same schools indicated that while controlling for factors like household income, teacher education, school size, and degree of supervision, information campaigns contributed to two-thirds in the reduction of capture of funds2. By partnering with the service facilities and the beneficiaries of the capitation grant program, the government has been able to strengthen transparency mechanisms and increase accountability as a result of which there as been a substantial reduction in the capture of funds in both the health and the education sectors in Uganda. Given the Ugandan experience, it is reasonable to expect similar outcomes in the other three applications of PETS if there would have been follow-up surveys in these regions too. IV. Key Elements of Empowerment Access to Information Information dissemination regarding public spending has strengthened transparency and accountability of government to the public regarding goals, targets, and public expenditure management. This has provided the basis for the emergence of informed public participation and has shown an improvement in the flow of funds and the quality 1 Dehn, Reinikka, Svensson, "Survey tools for assessing performance in service delivery", 2003, World Bank, Washington D.C. 2 same as 1 5 Empowerment Case Studies: Public Expenditure Tracking Surveys of services provided. By employing mechanisms for information access and promotion of transparency, the inadequacies and problems in the service delivery systems have been exposed and treated. For instance, in Uganda, in 1996 when PETS was first deployed, findings illustrated non-wage primary education spending by local bureaucrats. As a response to this, the government launched information campaigns both at the national, district and school levels by publishing monthly intergovernmental transfers of public funds in the newspaper, broadcasting on radio, and instructing the primary school to post information on inflows of funds. Results derived from a repeat survey done in 1999-2000 showed a remarkable increase in the school receiving non-wage education spending (from 20% in 1995 to over 90% ). This improvement can be attributed to information dissemination on public spending information by the government through the media and posting information at schools and districts3. Additionally, it was found that schools with newspapers, increased their funding by 10% more as compared to schools that did not have newspaper coverage. Thus, emergence of informed civic participation evolved through information disclosure whereby users of public services voiced their opinion and facilitated in the tracking process. This syndrome can assist in the design of an action plan for improved allocation and delivery of resources and services through citizens' assessment of government services. Accountability PETS, as a tool, has been shown to strengthen transparency and accountability in budget execution and service delivery. It is through the mechanisms of information access and accountability that the public agencies felt the pressure to comply with citizens' demands and were compelled to respond to accountability factors. Free circulation of accurate and complete information regarding the government budget allocations enabled the citizens to monitor and track whether public spending was consistent with the flow of funds to the agencies responsible for the delivery of goods and services. It is quite clear that accountability is manifested at two levels: at the government level and at the frontline delivery facilities. While the government exposes itself to public scrutiny through transparency in the budgetary system, the service providers subject themselves to a controlling mechanism (the users of the facilities) that enforce a certain degree of quality control in terms of timely and effective services. Additionally, facility level inputs have also encouraged accountability for national spending This is quite similar to the citizen report cards initiative which allows the citizens to monitor the quality of goods and services provided by government in relation to expenditures made for these goods and services. Inclusion and Participation Civic participation has allowed for the examination of the efficacy of the public expenditure systems. To arrive at answers to questions like if the resources are being 3 http://www1.worldbank.org/publicsector/pe/trackingsurveys.htm 6 Empowerment Case Studies: Public Expenditure Tracking Surveys channeled to the right sectors in the stipulated amounts, tasks like accounting, monitoring, auditing and evaluation should be carried out. This was done through citizen's participation which contributed to exposing the gaps and inadequacies in the delivery systems and has helped in better targeting of allocation and tracking of resources. The citizens participated in exposing the degree of effectiveness of the public sector. The process of demanding accountability for the powers bestowed in the public sector, has helped the facilities and institutions render themselves to efficient, effective and timely services. It has also proved to be a solid, civic, scrutiny mechanism to track the loop holes and thus promoted the attainment of the goals of funds allocation for specific sectors. In the process, it has resulted in providing a better understanding of the scenario and thus initiated inquiries into the gaps between allocations and disbursement, the reasons for delay, shortfalls and leakages. The outcome of such an initiative is analogous to that experienced with participatory budgeting where the link between the participation, formulation of and investment in pro- poor policies, greater public consensus and a more effective government have been established. Therefore, public participation has established a well functioning, concrete feedback loop in the planning process. V. Issues and Lessons Survey methods might be powerful tools for monitoring and evaluation of the impact of interventions or gauging the efficacy of the service delivery systems (as in this case) but they have to be very carefully designed owing to fact that such studies are complex and should incorporate context specific inquires to arrive at the relevant answers. Operationalization of concepts (defining and measuring) and designing questionnaires that point to the essential units of analysis and prediction are another important question. This was done by identifying implementing agency, piloting and training field workers to obtain quality data. Issues of not getting accurate data seems to be reasonable in a setting where there are strong incentives to misreport or not report key data, and there is also a possibility of not capturing relevant information for which measurement and analysis is required. PETS deals with these issues by first, employing a "multi angular data collection" strategy which means data are collected from varied sources and secondly, by identifying the sources that have a tendency to misreport and carefully choosing the right data sources.4 Lack of explicit allocation or allocation rules compounded by poor record keeping and sources of multiple financing added to the burden of efficient tracking systems. Additionally, feasibility of employing such surveys poses a challenge. How far the target 4 Dehn,Reinikka, Svensson, "Survey tools for assessing service delivery", June 2002, World Bank 7 Empowerment Case Studies: Public Expenditure Tracking Surveys audience can be identified and their willingness to participate becomes another issues which needs to be tackled. Key findings Apart from the leakage of funds, disparity in the allocated and disbursed funds, and inefficiencies in the public service delivery systems, there are other indicative factors that complemented the overwhelmingly poor outputs. Firstly, in cases where the funds reached their intended destinations, lack of, or insufficient, incentives for the service providers in terms of pay, absence of effective monitoring system led to syndromes like "absenteeism" among frontline workers. Secondly, the interplay between demand and supply side failures resulted in low level of public services resulting in less desirable outcomes. For instance, even when services are provided, households, for varied reasons (one of which could be economic) pull their children out of school or may not even entertain the idea of their children taking advantage of public service systems. Some lessons that can be drawn are listed below: · A standardized approach cannot be employed for PETS ­ since the issue and the context have implications on the process and outcome of PETS. The effective administration of the PETS technique is dependent on the complexity and context and the specific administrative and logistics of systems under examination. As a result, in some instances, case studies may be more applicable than a survey. · Collecting data "upstream" from the facility has shown that low levels of financing, capacity constraints and information asymmetries contribute to the degree of inequities, inefficiencies and resource inadequacies that is observed at facility level. · Understanding the impact of public spending is improved by examining the determinants of performance of the facilities. Issues like quality, efficiency and appropriateness of services have to be taken into consideration while studying the impact of public spending in terms of performance of the facilities. · Civil society organizations should mount capacity building exercises for communities to get involved in input and output monitoring at the local level. Further, the existing government ministries' monitoring and evaluation systems are be enhanced to allow for consistent and quality tracking of expenditures. · Implementation of such tools on a sustained basis is essential for long term viability of such endeavors. The results obtained from deploying such tools can help in reforming policies and public management systems. However, this can also be coupled with performance-based budgeting of service providers. 8 Empowerment Case Studies: Public Expenditure Tracking Surveys · The performance of service delivery systems might vary within the same jurisdictions. This might be due to the differences in the management and operations of the service delivery units. For instance, in Uganda, good record keeping by primary schools was observed while health clinics displayed inefficiencies(in fact there was complete lack of information) pertaining to in and out patients, financial information like user fees and cash and in-kind transfers of public resources5. Additionally, one can also argue that divergence in institutional arrangements can be associated with the variations seen in the impact of the service delivery systems. To conclude, PETS has provided useful information regarding broad allocation of resources, problems associated with budget execution(discretionary reallocation of resources resulting from favoritism and lack of sharing of information), leakage of financial and material resources, misappropriation of resources (for instance, stealing of medicines), problems in service delivery (for e.g absenteeism) and lack of demand for services due to inefficiency, high cost and low quality of the service providers. The data obtained from such surveys help reveal operational problems and can be used for empirical research on capture of funds. This is an effective tool to evaluate the effects of responsibility assignment at different levels including central, local governments and community groups as well as to examine the impact of decentralization. 5 Emmanuel and Reinikka (1998), Do budgets really matter? Evidence from public spending on education and health in Uganda, Policy Research Working Paper, 1926, World Bank, Washington D.C 9 Empowerment Case Studies: Public Expenditure Tracking Surveys VI. Further Information: References and World Wide Web Resources 1. Ablo, Emmanuel, and Ritva Reinikka. 1998. "Do Budgets Really Matter? Evidence from Public Spending on Education and Health in Uganda." Policy Research Working Paper 1926. World Bank, Africa Region, Macroeconomics 2, Washington, D.C URL: http://econ.worldbank.org/docs/539.pdf 2. Dehn Jan, Reinikka Ritva, Svensson Jakob, Survey tools for assessing Service Delivery, June 2002, Development Research Group, Public Services Research, World Bank. URL: http://econ.worldbank.org/files/16094_survey.tools.for.assessing.sd.june2002.pdf 3. Dehn D et al, Assessing Performance in Service Delivery: Tools for assessing public service delivery, The World Bank URL: http://216.239.57.104/search?q=cache:Hjd8tDU8d94J:www.ppmu.moldova.md/en/cactive/5s eminar/Okrasa_3en.ppt 4. Dehn Jan, Reinikka Ritva, Svensson Jakob, 2003, Survey tools for assessing Performance in Service Delivery, World Bank. URL: http://www1.worldbank.org/publicsector/pe/PETS1.pdf 5. Edward Miguel, Zaidi Farhan, "Do Politicians reward their Supporters? Public spending and Incumbency advantage in Ghana", October 2003. URL: http://elsa.berkeley.edu/users/emiguel/miguel_ghana.pdf 6. Foster Mick, Mijumbi, P. When and Why does Poverty get Budget Priority:Poverty Reduction Strategy and Public Expenditure in Uganda, 2002, ODI Working Paper 163. URL: http://www.odi.org.uk/pppg/publications/working_papers/163.html 7. Foster Mick, Zormelo D, How, When and Why does Poverty get Budget Priority:Poverty Reduction Strategy and Public Expenditure in Ghana, 2002, ODI Working Paper 164. URL: http://www.odi.org.uk/pppg/publications/working_papers/164.html 8. Fozzard, Adrian, Naschold Felix, How, When and Why does Poverty get Budget Priority:Poverty Reduction Strategy and Public Expenditure in Tanzania, 2002, ODI Working Paper 165 URL: http://www.odi.org.uk/pppg/publications/working_papers/165.html 9. Foster Mick, Fozzard, Adrian, Naschold Felix , Conway Tim, How, When and Why does Poverty get Budget Priority:Poverty Reduction Strategy and Public Expenditure in Five Africab Countries, Synthesis Paper, ODI Working Paper 168, May 2002. URL: http://www.odi.org.uk/publications/working_papers/wp168.pdf 10 Empowerment Case Studies: Public Expenditure Tracking Surveys 10. Healey John , Foster Mick , Norton Andy, and Booth David, "Towards National Public Expenditure Strategies for Poverty Reduction", ODI Poverty Briefing, 7, March 2000 URL: http://www.odi.org.uk/publications/briefing/pov7.html 11. Lindelow, Magnus, Public expenditure tracking and Service Delivery Surveys, 11th International anti corruption conference, Seoul, May 26, 2003, Development Research Group, The World Bank URL: http://www.icgfm.org/downloads/MagnusLindelow.ppt 12. Magnus, Lindelow, Reinikka, Ritva, Public Expenditure Tracking surveys, Health Facility Survey workshop, December 12, 2001. URL: http://216.239.57.104/search?q=cache:- BplJQNANSgJ:wbln0018.worldbank.org/HDNet/HDdocs.nsf/0/03d62f55c37fe6e085256b19 005ed6e2/%24FILE/PETS%2520presentation.ppt 13. Murali, D. Business Line, Internet Edition, Financial Daily from THE HINDU group of publications, Saturday, Jul 05, 2003 URL: http://www.thehindubusinessline.com/bline/2003/07/05/stories/2003070500100800.htm 14. Reinikka, Ritva, Presentation on Public Expenditure Tracking and Service Delivery Surveys WBI Learning Activity: Empirical Tools for Governance Analysis, June 18, 2002, DECRG, The World Bank URL: http://216.239.39.104/search?q=cache:cQmZwTG67K4J:www1.worldbank.org/publicsector/d ecentralization/ReinikkaJune18.ppt 15. Reinikka, Ritva, Transparency and Accountability: Expenditure Tracking Surveys for Public Sector reform and PRSCs URL: http://www.worldbank.org/participation/webfiles/RitvaMay2001.pdf 16. Reinikka, Ritva, Recovery in Service Delivery:Evidence from Schools and health Centres, World Bank. URL: http://www.worldbank.org/research/projects/publicspending/tools/Uganda%20PETS/Chapt11. health.education.2001.pdf 17. Reinikka, Ritva, Svensson, Jakob, Explaining leakage of funds, October 2001 URL: http://www.econ.ku.dk/epru/Seminars/files/svensson261001.pdf 18. Reinikka, Ritva, Svensson, Jakob, Assessing Frontline Service Delivery, 2002, World Bank, Washington D.C URL: http://www.sti.ch/pdfs/swap158.pdf 19. Schacter Mark, The Heart of the Matter: Donors, Development Assistance and Public Sector Reform, Policy Brief, Institute of Governance, Canada URL: http://www.iog.ca/publications/policybrief10.pdf 11 Empowerment Case Studies: Public Expenditure Tracking Surveys 20. Wagle, Swamim and Shah, Parmesh; Public Expenditure Tracking and Facilities Surveys: a general note on methodology, Social Development Department, World Bank. URL: http://km.worldbank.org/LacCivilsociety/docs/puntacana2003/web/Docs%20Eng/Utzigpaper/ PCE_exptrack.pdf 21. Wagle, Swamim and Shah, Parmesh. 2002. Participation in Public Expenditure Systems: An issue paper, World Bank, Social Development Department, Washington D.C. URL: http://216.239.37.104/search?q=cache:Dc0yF_NBEMgJ:www.oneworldaction.org/download/ WBPublicExpenditureandParticipationIssuePaper.doc Web Resources: http://www.kippra.org/institute/division/project1.asp?pass=20 http://www1.worldbank.org/publicsector/pe/trackingsurveys.htm 12