BULGARIA SPENDING REVIEW MANUAL June 2018 ACKNOWLEDGEMENTS This manual is a deliverable under the technical assistance project “Spending Review and Institutional Strengthening� (P161180) under which the World Bank provides support to the introduction of spending reviews in Bulgaria. The manual was prepared by Jonas Arp Fallov (Sr. Public Sector Specialist) and Elena Georgieva (Public Sector Specialist) with inputs and guidance from Stella Ilieva (Sr. Economist and Task Team Leader) and Enrique Aldaz-Carroll (Sr. Economist and Co-Task Team Leader) and drawing from the deliverables prepared by the rest of the Bank team under the same project. The team received valuable advice from Lalita Moorty (Practice Manager, GMTE1), Roby Senderowitsch (Practice Manager, GGOEW), and Antony Thompson (Country Manager). The team also benefitted from helpful peer review comments received from William Leslie Dorotinsky (Adviser, GGODZ) and Theo David Thomas (Economic Adviser, OPSCE). The team is grateful for the support from and collaboration with a wide range of Bulgarian officials. In particular, the team would like to recognize the guidance and support received from the Ministry of Finance. 2 ABBREVIATIONS B/C Benefit/Cost CBA Cost Benefit Analysis CEA Cost Effectiveness Analysis CoM Council of Ministers CUA Cost Utility Analysis DEA Data Envelopment Analysis EC European Commission EU European Union FLBU First Level Budget Unit FY Fiscal Year GDP Gross Domestic Product GoB Government of Bulgaria ICT Information and Communications Technologies IFIs International Financial Institutions IMF International Monetary Fund IRR Internal Rate of Return MDAs Ministries, departments, agencies MoF Ministry of Finance MTBF Medium Term Budget Framework OECD Organization for Economic Cooperation and Development NPV Net Present Value O&M Operations and maintenance PFM Public Financial Management SOE State-owned enterprise TA Technical Assistance UK United Kingdom WB World Bank 3 TABLE OF CONTENTS Acknowledgements ......................................................................................................................... 2 Abbreviations .................................................................................................................................. 3 TABLE OF CONTENTS ................................................................................................................ 4 LIST OF FIGURES ........................................................................................................................ 5 LIST OF TABLES .......................................................................................................................... 5 LIST OF BOXES ............................................................................................................................ 5 I. INTRODUCTION ................................................................................................................ 6 A. Background, objectives and audience .................................................................................. 6 B. Structure of the manual ........................................................................................................ 7 II. ABOUT SPENDING REVIEWS ......................................................................................... 8 A. The nature and objectives of spending reviews ................................................................... 8 B. Main benefits of spending reviews ...................................................................................... 9 C. Typology of spending reviews ........................................................................................... 10 III. THE SPENDING REVIEW PROCESS............................................................................. 12 A. Main phases of the spending review process ..................................................................... 12 B. Integration with the budget process ................................................................................... 13 IV. PREPARATION OF SPENDING REVIEWS ................................................................... 16 A. Formulation of savings targets and identification of spending review topics ................... 16 B. Preparation of Terms of Reference and decision to initiate a spending review................. 19 C. Setting up the Steering Committee and working groups ................................................... 19 V. CONDUCTING THE SPENDING REVIEW ......................................................................... 23 A. Analyzing spending patterns, efficiency and effectiveness: key analytical questions, data requirements and analytical techniques..................................................................................... 23 B. Developing and analyzing savings/reallocation options .................................................... 35 VI. DECISIONS ON SPENDING REVIEWS ............................................................................. 39 A. Process of selection of savings or reallocations and approval of spending reviews .......... 39 B. Information required for ministers to make an informed decision .................................... 39 VII. REPORTING AND COMMUNICATION ........................................................................... 41 A. Presentation of data, analysis and recommendations ......................................................... 41 B. Dissemination .................................................................................................................... 43 ANNEXES .................................................................................................................................... 45 Annex 1: References ................................................................................................................. 45 Annex 2: Glossary of terms ....................................................................................................... 47 Annex 3: A Model Terms of Reference for a spending review ................................................ 49 Annex 4: Summary of SR roles and responsibilities................................................................. 50 Annex 5: Analytical questions typically raised in spending reviews ........................................ 53 Annex 6: Suggested Outline of a Spending Review ................................................................. 56 4 LIST OF FIGURES Figure 1: Main phases of the spending review process ................................................................ 12 Figure 2: Formulation of savings targets ...................................................................................... 17 Figure 3: Proposed institutional set-up ......................................................................................... 21 Figure 4: Example of profile analysis used in the first phase of the spending review in Bulgaria28 Figure 5: Example of trend analysis used in the first phase of the spending review on public pay in Bulgaria ..................................................................................................................................... 29 Figure 6: Example of DEA analysis performed during the first phase of the spending review in Bulgaria ......................................................................................................................................... 33 Figure 7: Example of DEA analysis performed during the first phase of the spending review in Bulgaria ......................................................................................................................................... 33 Figure 8: Prioritization matrix ...................................................................................................... 37 LIST OF TABLES Table 1: Spending review process and budget calendar ............................................................... 13 Table 2: Examples of savings targets from spending reviews ...................................................... 16 Table 3: Summary of analytical framework for spending reviews ............................................... 24 Table 4: Savings decision making template.................................................................................. 40 LIST OF BOXES Box 1: Dimensions of spending reviews ...................................................................................... 11 Box 2: Scenarios for identifying spending review topics ............................................................. 19 Box 3: Institutional set up for spending reviews: International approaches ................................. 20 Box 4: Commonly used decision criteria in CBA analysis........................................................... 30 Box 5: Good practice in spending review report writing .............................................................. 42 5 I. INTRODUCTION A. Background, objectives and audience 1. In 2016, the Government of Bulgaria (GoB) decided to initiate spending reviews as a way to improve efficiency and effectiveness of its spending in the context of a moderate fiscal consolidation process. The medium-term fiscal framework for 2017-2020 envisages a small deficit in 2017 with the fiscal balance improving to a small surplus by 20201. Bulgaria is one of the lowest spenders in the EU in general but spending outcomes lag those of other EU member states. 2. Improvements in the efficiency and effectiveness of government spending will be important for Bulgaria to sustain a relatively low spending level while delivering public services needed to put it on a trajectory to high-income country status. Bulgaria is heading for the steepest drop in the working-age population of any country. As the share of the elderly increases, demand for key public services, such as health, long-term care and pensions, will increase, which, in the absence of reform, is likely to increase spending pressures (World Bank 2014). The government will also need to improve spending efficiency within sectors through enhanced sector-specific policies and to re-allocate spending to underfunded sectors to achieve outcomes associated with a high-income country status. The European Commission (EC)2 has recommended introducing spending reviews in line with many other EU and OECD countries, which have benefitted from institutionalizing such reviews. 3. This manual aims to guide government officials in the Bulgarian government administration on how to plan, prepare, conduct, decide, and report on spending reviews. It seeks to assist GoB in introducing and sustaining spending reviews as a tool to continuously analyze on-going spending and identify savings and options to increase efficiency and effectiveness. The organization, process and methodology of spending reviews included in this manual are built on the experiences gained through the initial pilot spending reviews conducted in Bulgaria with World Bank support during 2016 and 2017. The pilot spending reviews were conducted in two phases. The first phase included a macro-level analysis of the wage bill and operations and maintenance spending in 7 ministries and 21 municipalities, which then motivated a selection of areas for in-depth spending review in the second phase: public expenditure on public order and safety, and waste management. 4. The manual is first and foremost a procedural guideline. It seeks to clarify the definition, scope, and duration of spending reviews; how to arrange the spending review process and its main phases; and the roles and responsibilities of stakeholders involved in each phase. At the request of the Ministry of Finance, the manual however also provides high-level methodological guidance on how to approach spending reviews from an analytical point of view. This includes a summary description of some commonly used analytical approaches: criteria for identifying candidates for 1 Bulgaria Convergence Program 2017-2020 2 European Commission, 2014: “Public Spending Reviews: Design, Conduct, Implementation�, Economic Working Papers 525 6 spending review, including topics and institutional coverage, how to set savings targets, what to look out for in collecting data, and various approaches to analyze spending patterns and options for savings, efficiency and effectiveness gains. 5. The target audience for the manual is government officials who are involved in spending reviews. The manual is aimed mainly at technical-level officials but may also inform decisions made by high level officials and policy-makers. The manual will benefit officials of the Ministry of Finance (MoF) leading the spending review processes or overseeing specific spending reviews as well as officials in line ministries and municipalities who participate in spending reviews in their specific areas. 6. The manual is designed to be a living document. The work on spending reviews is still in an early stage in Bulgaria. The manual will be adjusted by the MoF as experience is gained with the spending review process. B. Structure of the manual 7. The remainder of the manual is structured as follows: • Chapter II clarifies the concept of spending reviews, including its definition and scope in the Bulgarian context. • Chapter III lays out the spending review process, and its main phases, the roles and responsibilities of key actors at each phase and how the spending review intercepts with the budget calendar. • Chapter IV explains how spending reviews are prepared including identification of spending review topics and savings targets, formulation of terms of reference and how to set up the necessary working groups or other organizational arrangements. • Chapter V gives guidance on the conduct of the analytical spending review process, including data collection, and approaches to analyze spending patterns and savings options. • Chapter VI explains the decision-making process. • Chapter VII provides guidance on reporting and communication; and • Annexes 1-6 contain references, a Glossary of Terms; model Terms of Reference for a spending review; a summary of roles and responsibilities; a list by topic of analytical questions typically raised in spending reviews; and an example of a spending review report outline. 7 II. ABOUT SPENDING REVIEWS A. The nature and objectives of spending reviews 8. A spending review is defined in this manual as the conduct of identifying and proposing measures for reducing and/or reallocating government spending, based on the systematic scrutiny of baseline expenditure.3 Two key objectives of a spending review can be identified within this definition: On the one hand, a spending review can be used to achieve fiscal savings, i.e. reductions in the aggregate level of government spending. On the other hand, a spending review can be motivated by the need to reprioritize spending with a view to improve the overall effectiveness or efficiency of government programs and activities. 9. The need for fiscal adjustments can lead to a demand for spending reviews to reduce aggregate government spending. In this case, the savings identified by a spending review are not “reinvested� in new spending programs or increases in any existing spending programs in the budget. This approach would typically be used in situations where the maintenance or improvement of fiscal discipline demands reductions in overall spending for the fiscal year and/or the medium term (3 years). 10. A spending review can also be motivated by the need to improve effectiveness or efficiency through spending prioritization. In this case, the proceeds of the savings measures identified in one spending program are channeled to other budget headings in the same program or to other spending programs, and the spending review thereby becomes a mechanism of reallocating spending to where the resources are most needed or better match political priorities. It is thus important to note that setting targets for – and achieving – spending reductions in one area can be combined with increased spending in other areas. In this case, the spending review is a method of identifying financing for increased priority spending without jeopardizing the overall fiscal situation. 11. A spending review is a distinct form of public policy evaluation. Like other forms of public policy evaluations, a spending review assesses public policy programs or activities based on criteria such as effectiveness, efficiency, sustainability or overall value for money. What sets the spending reviews apart from other types of evaluations is the objective to use the assessment activities as a basis for identifying and proposing options for savings and reallocations. 12. Spending reviews can have both strategic and tactical dimensions in evaluating spending, each of which may correspond to different approaches to identifying and proposing savings measures: • The strategic dimension of a spending review is to evaluate or challenge the rationale of a 3 This definition is based on “Spending Review,� a note prepared by Marc Robinson for the 30 th annual meeting for the Working Party of Senior Budget Officials, OECD, Paris, 2013. Vandierendonck, C. (2014) uses a very similar definition of spending reviews as “coordinated and in-depth analysis of baseline expenditures with the prime objective to detect efficiency savings and opportunities for cutting low-priority or ineffective expenditures…� (Vandierendonck, C. (2014): Public Spending Reviews: Design, Conduct, Implementation; EU economic Papers 525/2014). 8 government program (is public intervention still needed? Is it in line with government strategies or priorities?) or its effectiveness (to what extent does the program/activity meet its stated objectives?) and use this analysis to inform proposals for reductions or reallocations of spending. The strategic dimension often leads to formulating options for strategic savings, where expenditure reductions are achieved by cutting back services (outputs) or transfer payments delivered to the community. • The tactical dimension of a spending review assumes that the government program or activity has already passed the strategic test and is both strategically relevant and effective. The task is then to evaluate the efficiency of a spending program or activity, i.e. whether the same output can be delivered at a lower cost. The tactical dimension is often associated with searching for efficiency savings which are expenditure reductions achieved by changing the way in which services are produced to deliver the same quantity and quality of services (i.e. outputs) at lower cost. 13. Spending reviews are most effective if integrated with the budget process. Such an integration would imply, firstly, that the spending reviews take their starting point in an assessment of the fiscal situation, the medium-term budget outlook, and the need for savings or spending prioritizations in order to achieve policy objectives. Secondly, since any proposed measures for savings or reallocations are to be implemented in the budget, the process of deciding on the options should be closely aligned with the budget calendar and the key decision points for approving the budget at various stages and with some or full overlap of decision makers. 14. Spending reviews are usually focused on analyzing the spending of baseline spending programs. The analysis of new spending proposals, on the other hand, are not to be considered spending reviews, but spending reviews can provide an opportunity for decision makers to assess and prioritize new spending proposals against savings options within on-going spending programs. B. Main benefits of spending reviews 15. Introducing spending reviews offers benefits in addition to the benefits of the regular budget procedures. Among these are: • Adding specific savings measures to the savings tool box. Spending reviews provide opportunities for identifying savings or reallocation options, which are difficult to accomplish without systematic analysis. Spending reviews can provide options for savings with specific justifications and allow “deep diving� on specific issues. This is an important supplement to other types of savings such as across-the-board spending cuts. • Facilitating spending prioritization. As in many other countries, the ability to reallocate funds between spending programs has traditionally been a weak point of the Bulgarian budget process. This is partly because of lack of information about the efficiency and effectiveness of ongoing spending programs, partly because of resistance from line ministries and other stakeholders. Spending reviews can enhance the budget process on both these counts by providing a systematic analysis of program performance and by 9 offering analysis which can justify and legitimize spending prioritizations. • Matching savings options in baseline spending with new spending proposals. Integrating spending reviews in the budget process offers the opportunity to simultaneously consider new spending proposals with savings options, thereby comparing the relative merits of continued on-going spending and new spending. Knowledge about the scale of new spending proposals can also be used to inform the planning of spending reviews to ensure that new spending proposals can be accommodated within certain fiscal parameters. • Strengthening analytical capacity. Overall, the introduction of spending reviews is expected to enhance the analytical capacity of the MoF staff as well as staff of line ministries and municipalities, which will over time have a strong beneficial impact on the efficiency and outcomes of the budget process. C. Typology of spending reviews 16. Practices differ considerably internationally with respect to the coverage, scope and approach to spending reviews. Some countries have approached spending reviews as a comprehensive exercise covering most of the central government budget in which the specific topics or problems to be addressed have not been defined at the outset of the spending review campaign. This approach has been used in the UK for the 2010 Comprehensive Spending Review, in Canada for the 2011 Strategic Review, and in the value for money reviews in Ireland. Other countries, such as Netherlands or Denmark, have used a much more selective and problem-driven approach to spending reviews in which the number and scale of spending reviews vary greatly from year to year depending on the fiscal situation and the policy priorities of the governments. For spending reviews in these two countries, the topics of the spending reviews have typically been agreed beforehand in relation to specific problems or opportunities. 17. The spending reviews in Bulgaria are planned to follow a problem-driven approach by which the topics and problems to be addressed are defined at the outset. Within this approach the types of spending reviews can vary considerably. Some important dimensions are outlined in Box 1 below. 10 Box 1: Dimensions of spending reviews Topic of spending reviews Program reviews: analysis of the effectiveness or efficiency of one or more government programs or activities (i.e. specific categories of services or transfer payments). Program reviews may deliver either strategic savings (by reducing the services provided by the program) or efficiency savings (by lowering the costs of delivering services under the program). Process reviews: scrutiny of specific business processes used in the production of government services – for example, procurement processes; IT systems and practices; or human resources management practices. Process reviews aim to achieve efficiency rather than strategic savings. Agency reviews: review of a whole government organization (ministry, administration or other type of government entity), and may in principle cover all of the entity's programs and processes. Vertical or horizontal? A vertical review seeks to analyze most or all functions or processes within an organizational entity and subordinated agencies. A horizontal review examines one or a small number of processes or functions across a number of organizational entities. Source: Marc Robinson, “Spending Review,� OECD (2013). 11 III. THE SPENDING REVIEW PROCESS A. Main phases of the spending review process 18. The spending review process consists of four main phases (Figure 1). Each of the phases are summarized below. The subsequent chapters are structured according to these main phases of the process. Figure 1: Main phases of the spending review process •Identify and decide on spending review topics Prepare •Set savings targets •ToR and organizational set up •Collect data Conduct •Analyze spending & write report •Recommend savings options •Prepare Cabinet decision Decide •Conduct decision meeting •Inform stakeholders Implement •Disseminate to public •Implement & monitor 19. The preparation phase of spending reviews is where spending review topics are identified and any targets are set for savings to be achieved. The preparation phase also includes formulating terms of reference and defining roles and responsibilities of involved stakeholders, offering data sources, explaining how to analyze data, how to develop savings options, timetable, and setting up the organization around individual spending reviews, including working groups and steering committee. This phase also defines the public communication aspects and the timeline. 20. Conducting the spending review includes collecting data, analyzing spending patterns and developing savings and reallocation options. This phase also includes finalizing the written report to summarize the analysis and recommendations. 21. The decision phase includes preparing materials for a CoM decision on the recommended savings/reallocation options provided in each individual spending review . These materials could include a presentation of options and results, and selected savings options. 12 22. The implementation of the spending review involves adjusting the relevant expenditure ceilings and sections of the budget based on the CоM approved results. This phase also includes considerations as to whether the reports should be published and in which form, and disseminating results to other relevant stakeholders. Stakeholders may include the National Assembly Committees, unions, broader public, affected groups, as appropriate. B. Integration with the budget process 23. While international practices vary with respect to the linkages between the budget process and the spending review process, there are strong reasons for integrating spending reviews firmly within the budget process. The budget process generates a number of inputs for the spending reviews. For example, the setting of savings targets should be informed by the medium-term budget forecast and the expenditure estimates for new priorities in next fiscal year. The information from the budget forecasts is essential to define the necessary scale of savings options in order to close any gaps in the financing of expenditure priorities. The budget process also frames the timing and decision-making regarding the budget documents in which the results of the spending reviews will be implemented. 24. The spending review process therefore needs to be aligned with the budget calendar. Table 1 below shows the interception points between the budget process milestones and those of the spending review process exemplified with the budget process in Bulgaria for fiscal year 2018. Table 1: Spending review process and budget calendar Month Selected budget calendar Spending Review Process milestones Milestones Initiation (year 1) Finalization (year 2) January January 31: CoM adopts annual budget procedure February February 8: MoF prepares a February 1: Spending preliminary assessment of the review working groups main macroeconomic submit draft spending indicators for the medium term review reports to Steering February 10: MoF issues February 10: MoF Committee. instruction to prepare and Instruction includes Steering Committee present budget forecasts for the guidelines on spending evaluates spending review medium term review reports and requests changes as needed. February 15: MoF submits to CоM final spending review reports for approval including recommendations for savings and reallocations. February 22: CоM approves spending reviews. 13 March March 2: First Level Budget March 2: Based on March 2: FLBUs Units (FLBUs) submit draft medium term budget incorporate results of medium-term budget forecasts forecast, MoF starts work spending reviews in draft to the MoF on identifying spending medium-term budget review topics and setting forecast. March 31: MoF develops savings targets. consolidated draft Medium March 31: MoF prepares Term Budget Framework Spending Review ToRs (MTBF) with input from relevant FLBUs April April 13: MoF submits April 13: Consolidated consolidated draft MTBF to draft MTBF is CоM accompanied by MoF recommendations on proposed spending April 20: CoM approves reviews (ToRs) consolidated draft MTBF April 20: CoM decides on initiation of proposed spending reviews May - Working groups initiate August work September September 1: MoF issues instruction for the draft budget for next FY and updated budget estimates for out years September 15: FLBUs submit draft budget for next FY and updated budget estimates for out years September 25: MoF evaluates draft budget for next FY and updated budget estimates for out years October October 9: MoF develops draft Budget Bill and updated Medium Term Budget Forecast October 16: coordinate with municipalities and consult Fiscal Council on draft budget bill October 23: draft budget is October 23: MoF submits draft accompanied with a note budget to CоM explaining how spending review results have been incorporated. October 31: Draft Budget Bill includes results of 14 October 31: CoM Approval of Spending Review draft Budget Bill implemented in relevant October 31: Submission to budget ceilings and National Assembly. headings. Sources: Council of Ministers’ Decision on Budget process for 2018; MoF Instructions for Preparation and Presentation of Budgetary Forecasts 2018-20. 25. The timing of milestones in the spending review process implies an effective duration of spending reviews of about ten months. Spending review topics and savings targets are defined in March based on information from the draft MTBF. Spending reviews are then initiated in late April based on CoM approval and finalized in February of the following year in time for incorporating the results in the draft medium-term budget forecasts submitted by the FLBUs in March. Prior to that, the spending review must be approved by the Steering Committee chaired by the MoF (see further below). The results of the spending review are also to be implemented in the annual budget submitted to the National Assembly by end of October. 26. The one-year timeline allows sufficient preparation time for the spending reviews while at the same time ensuring alignment with key milestones in the budget process. Spending reviews are a novelty in the budget process in Bulgaria and officials will need some time to get used to applying a more systematic and analytical approach to scrutinizing the budget. The piloting of spending reviews has indicated that ten months is a realistic timeline. This however implies that spending reviews straddle two annual budget cycles – initiation in April of year 1 and final implementation in the budget prepared in year 2. The MTBF provides a multiyear framework to manage this. Over time, as the work on spending reviews matures, it may be an option to have shorter durations for some reviews in order to prepare, conduct and implement the results within one annual budget cycle. 15 IV. PREPARATION OF SPENDING REVIEWS 27. The preparation phase of spending reviews consists of the following main steps: • Formulation of savings targets • Identification of spending review topics • Preparation of Terms of Reference for spending review • Decision on initiation of spending review • Setting up steering committee and working group • Communication with the stakeholders A. Formulation of savings targets and identification of spending review topics 28. When preparing the annual budget, the MoF is often confronted with the challenge that the aggregated spending proposed by line ministries exceeds the total spending envelope. Achieving savings through spending reviews is therefore an important tool to bring the total spending estimate down to an affordable level. 29. Setting savings targets is a way to guide the work on identifying savings options in the spending reviews. Table 2 shows examples of savings targets set in different countries. Table 2: Examples of savings targets from spending reviews Country Review process Savings target Canada Strategic review Nominal 5% savings over 3 years Croatia Spending Review (2015) 10% nominal savings for four selected areas France General Review of Public 10% real savings over 3 years Policies (RGPP) Poland Spending Review (2015) Annual efficiency savings re- assigned to programs United Kingdom Spending Review (2004) 2.5% real savings per year for departmental spending United Kingdom General expenditure review 3% real savings per year for (2007) departmental spending. 5% cut in administrative budget Source: World Bank. 30. Savings targets consist of an aggregated savings target and individual targets or estimates for each individual spending review. Aggregated savings targets are formulated based on the MTBF and the spending estimate for next fiscal year by simultaneously taking into consideration the following factors (Figure 2): 16 • The financing gap, if any, between the spending envelope (the total consolidated spending ceilings for the government) and the proposed total spending for the fiscal year in question, including both baseline spending and new spending proposals. • An assessment of spending pressures and fiscal consolidation needs in the medium term. • The amount of savings that can realistically be achieved given the spending review opportunities that have been identified and are considered politically feasible. • Other possible savings measures which are at the disposal of the MoF in order to eliminate the financing gap. Figure 2: Formulation of savings targets Source: World Bank. 31. The formulation of savings targets typically starts with a “top-down� assessment of the amount of aggregated savings needed. The process is then reversed by assessing the possible savings for each individual spending review under consideration. Both the aggregate and individual savings targets may be revised through a number of iterations. This iterative process requires that spending review topics have been identified simultaneously with the formulation of savings targets. 32. Savings targets for individual spending reviews are estimated specifically based on the nature of the individual spending review and may therefore vary significantly across spending reviews and over time. It is recommended to focus first on “the low hanging fruit� – the areas with the best possible chance of delivering the savings required. This could be, for example, the largest or fastest growing programs or areas which are outliers in terms of spending based on international comparisons. 33. The MoF is responsible for the formulation of aggregate and individual savings 17 targets and for recommending savings targets to the CoM for a final decision . The savings targets for individual spending review are validated with the relevant line ministries prior to finalizing the recommendations for the CoM. An overview of the financial gap and a consolidated list of new spending proposals and spending pressures detailing the budgetary impact for the fiscal year and three forward years should be presented to ministers alongside savings targets to clarify the context of the savings. Where there are significant cost pressures beyond the near-term horizon, these should also be made clear, and quantified where possible. 34. Identifying spending review topics is an important activity to make sure that a range of options are brought to the attention of the CoM, including options that would otherwise go unnoticed. Typically, the process of identifying spending review topics is led by the MoF to ensure that an objective opinion external to the relevant line ministry is brought to bear. The work will be carried out either by the designated spending review teams or by officials already assigned to oversee the individual budget areas. 35. The identification of spending review topics requires a combination of knowledge and skills about both budget and relevant policy areas. The officials tasked with identifying spending review topics should possess: • an overview of the relevant policy areas, including government policies and priorities, legal and regulatory requirements and the scope for changes within these requirements; • technical knowledge about the budget process and financial management practices in order to analyze how these policies and requirements – and any changes therein – might impact on the budget; • the ability to analyze expenditure patterns and identify outliers. 36. The identification of spending review topics should follow a problem-driven approach. As a starting point for identifying a spending review topic, officials should try to identify specific problems in a given budget area, the resolution of which might generate savings and opportunities for reallocation of funds. The scope of analysis, organizational coverage, and analytical methods to be applied should follow from the specific problem to be addressed and may thus vary between different spending reviews. While a comprehensive analysis is part of the spending review itself, some initial analysis of expenditures with potential for savings may have to be conducted in support of this problem identification before the launch of the spending review process. 37. The examples of problem areas and scenarios in Box 2 below may be used to identify problems to be solved. The officials who are tasked with identifying spending review topics will judge which scenario is relevant in each case. Chapter V further elaborates on the types of analysis, which can be applied in order to support the development of savings options within each problem area. 18 Box 2: Scenarios for identifying spending review topics Problem area: Improve effectiveness Scenario: A government program or activity is not achieving its intended objectives Problem area: Improve allocative efficiency Scenario: Some activities funded through the budget are low priority or not aligned with government strategies or policies. Problem area: Improve technical efficiency Scenarios: a) A budget area has received unchanged or increasing funding despite decreases in underlying cost drivers; b) The budget for specific budget headings is higher than comparable budget headings in other entities with no obvious reason. Problem area: Improve functional coherence Scenario: Indications of duplicative or inefficient functions carried out by an entity 38. The identified spending review topics and the estimated savings are summarized and recommendations are made to the CoM for a decision. The recommendations should include a brief with the supporting argument for why the spending review topic is recommended, the estimated results of the spending review including savings estimates as appropriate, any risks for achieving those results, and how the risks might be mitigated. B. Preparation of Terms of Reference and decision to initiate a spending review 39. Once the CoM has approved the recommended spending review topics and savings targets, the next step is to prepare the Terms of Reference (ToR) for the spending review . The ToR formalizes the objective, scope, methodology, deliverables and timeline for the spending review. The ToR is the basis for initiating the spending review and serves as a reference point for the quality review and approval of the final results, once the final spending review report and recommended savings are submitted. The ToR also serves to clarify the scope of the review for all concerned stakeholders. The Steering Committee discusses the ToR and approves it. Annex 3 contains a checklist for the contents of a spending review ToR. C. Setting up the Steering Committee and working groups 40. The approach to defining the institutional set up for spending reviews varies internationally. Box 3 below shows a range of alternative approaches. 19 Box 3: Institutional set up for spending reviews: International approaches • Bottom-up (used in Italy, UK, Ireland, Canada): in this approach, the spending ministries themselves are required by the government to identify savings options for presentation to the central agencies and political leadership. For this purpose, the spending ministries establish internal review teams which do not include representatives of the MoF; • Top-down (France): the senior decision-making level sets savings targets and reform options for each administrative unit; spending review teams are composed of MoF staff or nominees, and there is little or no participation of spending ministry staff. There is no process for requiring or requesting spending ministry endorsement of the savings options which are identified. This scheme is not commonly observed, since it marginalizes the spending ministries, does not tap into their knowledge of their own programs and creates a lack of understanding and acceptance of savings; • Joint hybrid scheme (Netherlands, Denmark): the spending ministries and the MoF form joint review teams to develop savings options. The savings options presented by these review teams to the political leadership are approved by both the spending ministry and the MoF; • Hybrid with alternative scheme: two sets of saving options are developed: one by the administration under review, and an alternative one by the MoF. Both sets are put forward to the senior levels for decision. Source: OECD (2013). Spending Reviews 41. The spending review process in Bulgaria is proposed to be governed by the institutional set up illustrated in figure 3 below. Annex 4 provides a summary of roles and responsibilities of all stakeholders in the spending review process. 20 Figure 3: Proposed institutional set-up Source: World Bank 42. A Steering Committee is established to provide strategic direction and guidance to the working groups throughout the spending review process. The steering committee is chaired by the Deputy Minister of Finance responsible for spending reviews and comprised of senior staff from the MoF (economic policy, budget and expenditure departments), and responsible line ministries. The Steering Committee is responsible for submitting the recommendations/policy options to the CoM for consideration and approval. 43. Typically, one steering committee is designated to cover all spending reviews in a given budget cycle. However, participation of line ministries in steering committee meetings may in some cases vary depending on which spending reviews are on the agenda. Line ministry representatives may for example only participate in meetings relevant for spending reviews in their own budget area, but not in meetings where the spending reviews of other ministries are on the agenda. A steering committee mandate will typically last for one budget cycle, since the composition of spending reviews and participating ministries will change from one budget cycle to the next leading to changes in the composition of steering committee members on the line ministry side. 44. The work at the technical level is organized through working groups. The working groups are responsible for conducting analysis, preparing reform options within the defined target savings and other qualitative recommendations. 45. The working groups are chaired by the Spending Review Leader (a MoF staff working under the direction of a senior MoF official). Other members are technical, planning 21 and budgeting staff of the MoF and relevant line ministries (budget analysts, internal auditors, strategic units, etc.), as well as local government representatives, if relevant. If relevant expertise exists, it is also an option to bring in external experts (auditors, consultants, academia). It is important that the working groups are non-political and consist of civil servants and independent experts. 46. The Steering Committee and Working Groups have regular coordination meetings to allow the spending review leaders to monitor progress and the Chair to provide guidance. It is recommended that no working group member may “veto� another member’s idea. This ensures that the CoM (and ultimately the wider public) is presented with a range of alternative policy options. 22 V. CONDUCTING THE SPENDING REVIEW 47. This chapter has two sections. The first section provides an overview of the data and techniques typically used in spending review analyses and their data requirements, while the second section focuses on developing and analyzing options for savings or reallocation. A. Analyzing spending patterns, efficiency and effectiveness: key analytical questions, data requirements and analytical techniques 48. This section provides guidance on the analytical techniques commonly used in conducting a spending review. As recommendations from a spending review are directed at identifying options for savings and/or reallocation, the framework for analysis should take as a starting point the identification of specific problems to be resolved – or opportunities to be harnessed – in order to realize the savings or reallocation potential. An analytical framework based on such a problem-driven approach could be based on the four generic problem areas suggested in Chapter IV: Effectiveness, allocative efficiency, technical efficiency, and functional coherence. Table 3 below summarizes this approach, together with the corresponding data requirements, types of analysis and recommendations that may be generated. Often, problems addressed in spending reviews require a combination of the different analytical techniques presented in the table. Annex 5 provides examples of analytical questions commonly asked in spending reviews, as well as questions asked when performing functional analysis. It should be noted that some of the types of analysis and data requirements listed under each problem area could also be used in other problem areas. 23 Table 3: Summary of analytical framework for spending reviews Problem area Examples of possible Data requirements Types of analysis Examples of areas of scenarios recommendations Effectiveness • A government program or • Government strategies, plans, • Desk reviews of documents • Terminate activities or activity is not achieving its reports; recent sector • Analysis of quantitative and reduce funding and intended objectives performance data; evaluation qualitative performance data reallocate to more effective studies; government and • Impact evaluation activities international statistics. Allocative • Activities not aligned with • Budget allocations by: year, • Profile analysis • Terminate activities or efficiency government strategies or functional composition, capital • Trend analysis reduce funding and policies vs. current, wage vs. nonwage, • Comparing/benchmarking reallocate to higher priority • Low socioeconomic returns public vs. private, national vs. budget activities of public funds subnational, across districts allocation/prioritization • Government strategies, plans, • Budget composition analysis reports; recent sector Estimating marginal returns performance data; evaluation • Cost benefit analysis studies; government and • Cost-effectiveness analysis international statistics. • Cost-utility analysis Technical • Constant or increasing • Budget allocations and actual • Data Envelopment Analysis • Reduced funding based on efficiency funding despite decreases spending by various categories (DEA) identified efficiency in underlying cost drivers, • Performance data on outputs • Budget deviation analysis potential e.g.: • Business processes review • Recommendations on how - decline in number of • IT systems gap analysis to implement new beneficiaries • Desk review of legal, technology or business - simplifications in the regulatory or international processes regulatory or requirements 24 international requirements - available new technology - new organizational models, processes or modes of service delivery • The budget for a specific budget heading is higher than comparable budget headings in other entities with no obvious reason. • • • Functional • Indications of duplicative • Functions by classification and • Review of mandates, goals • Legal and organizational coherence or inefficient functions spending levels per function and objectives; changes to carried out by an entity • Products/outputs of each • Benchmarking with other streamline/eliminate/outsou function countries rce/(de)centralize functions • Staffing levels per function • Gap analysis of systems and generate corresponding and per institution; savings • Employment types and contracts; • Unit labor costs; • Staff workloads and qualifications; Source: World Bank 25 Effectiveness 49. Effectiveness refers to the extent to which an intervention – program, project or policy – achieves or is expected to achieve its objectives, taking into account their relative importance. A spending review can identify and analyze government programs and activities which are not achieving their intended objectives and recommend measures for program adjustments, reduction of funding or termination. A major source of input into this analysis are evaluation reports which might have already been produced by existing monitoring and evaluation systems. 50. Types of analysis: • Analysis of performance information: this type of analysis focuses on understanding what the reviewed intervention was meant to achieve, how it measured success, whether the stated objectives were achieved and the extent to which the results can be attributed to the intervention funded by the government. It looks at quantitative and qualitative data on results at different levels (outputs, outcomes, impacts). Quantitative data can be obtained from program reports, national and international statistics, and surveys. Qualitative data can be obtained through interviews, focus groups, case studies and observations. During the first phase of the spending review on O&M and personnel expenditure in Bulgaria, a mapping of the sector objectives and targets for the seven ministries under review was undertaken by analyzing different documents such as program budgets, sector strategies, ministry mandates, plans, etc. This exercise showed that most programs have a multitude of objectives which are often difficult to measure by the specified indicators which focus mostly on outputs (e.g. number of meetings held), as opposed to outcomes (improvement in air quality). • Impact evaluations: this is a type of performance evaluation which determines changes (intended and unintended) in outcomes that are directly attributable to an intervention. This involves counterfactual analysis, which is a comparison between what actually happened and what would have happened in the absence of the intervention. Impact evaluations provide a definite assessment of an intervention’s effect, which can be a useful input into the decision about the intervention’s future funding. However, they are relatively complex, both in terms of data collection and analysis, and therefore time-consuming and costly. Realistically, impact evaluations can therefore only be expected to be used in more select cases. 51. Data requirements: program/project/policy documents, such as plans and reports; sectoral strategies; quantitative and qualitative information on outputs, outcomes and impacts; household and other surveys; evaluation reports. Allocative efficiency 52. Allocative efficiency refers to the capacity of the budget system to allocate resources to government programs and activities in a way that maximizes the net benefits to society. In practice, allocation in line with government priorities is often used as a proxy for allocative 26 efficiency. Determining allocative efficiency requires the government to establish and prioritize objectives and to assess the actual or expected net contribution of public expenditures to those objectives. Spending reviews can help identify and analyze government activities and programs which are low priority or not aligned with government strategies or policies. Spending reviews make recommendations on the future of these activities which could include their termination, reduced funding or reallocation of funds to higher priority activities. 53. Types of analytical techniques4: Common methods of descriptive expenditure analyses used to analyze allocative efficiency are profile, trend, and composition analysis. These analytical methods can also be used to undertake benchmarking and country comparisons. More advanced analytical techniques include socioeconomic assessment tools such as cost benefit analysis. The various types of analysis are described below. • Profile analysis: profile analysis captures expenditure at a point in time and usually presents data on levels of expenditure or components of expenditure in absolute numbers or as shares of GDP. Figure 4 provides examples which were used in Phase 1 of the spending review in Bulgaria. A profile analysis can provide answers to many useful questions in a spending review: how much was public expenditure in a given year? Who spent the most, on what and for what function? How did recurrent spending compare to investment spending? How much was allocated to a given sector? Did the country spend more on that sector than regional peer countries? How much was spent on priority sectors, such as health? What was the expenditure profile of this priority program? Profile analysis might also be used as an analytical tool when analyzing technical efficiency and effectiveness. Profile standard tables present spending by administrative, functional, or economic classification, or any combination of these classifications, as well as information on financing sources and expenditure types (Table 1, Annex 8). 4 Source: World Bank. 2016. BOOST Guidance Note No 2- Strengthening Public Expenditure Analysis. 27 Figure 4: Example of profile analysis used in the first phase of the spending review in Bulgaria General government O&M spending , General government O&M spending 2015 (in % of general government (in % of GDP) expenditures) 10 20 16 8 12 6 8 4 4 2 0 0 2015 2009 Source: World Bank based on Eurostat. • Trend analysis: trend analysis measures progress over time, presenting expenditures in absolute numbers, constant numbers, as growth rates, or as shares of GDP. Figure 5 provides an example used in the first phase of the spending review on public pay in Bulgaria. This analytic approach can help policymakers understand broad movements or tendencies over time: for example, what is driving the growth or decline of public expenditure over the last five years – investment, wages, and/or social transfers? If wages, which sectors exhibit the most growth? What is the trend in sector expenditure across the region? What has been the trend in recurrent expenditure in this sector? Trend standard tables present spending by administrative, functional, or economic classification, or any combination thereof. 28 Figure 5: Example of trend analysis used in the first phase of the spending review on public pay in Bulgaria General government wage bill, 2016 Percent of GDP Spending on compensation of employees (% of 12.0 11.5 BG EU10 EU28 11.0 10.5 10.0 GDP) 9.5 9.0 8.5 8.0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Source: World Bank based on Eurostat. • Comparing/benchmarking budget allocation/prioritization: this is one of the most commonly used techniques to prepare allocative efficiency analysis. If sufficient input data is available, comparative analysis can be done within a country, within the same industry in the same country, over time within a country or across geographic locations within a country. For example, spending reviews often try to identify inefficiencies in service delivery. One way to do that is to compare the ratios of wage and non-wage expenditures to total expenditures and compare that within the country – if regional data is available – or to other comparator countries to shed light on the relative allocative inefficiency in service delivery. Another technique is to compare unit cost of the service delivery among various service providers (in transport, publishing of the textbooks, administrative service, water utility, etc.) that are currently financed out of budget, and also with commercial service providers. • Budget composition analysis: an analysis of budget composition lays out the government’s priorities based on its expenditure outcomes. The analysis determines the composition of public spending at the national and subnational level (if relevant). The analysis could include decomposition by (i) functional area or priority sector (e.g. administration and general, social and municipal functions); (ii) economic area (e.g. current vs. capital, wage vs. nonwage, public vs. private) and (iii) transfers or subsidies to state- owned enterprises. The analysis should reflect how the spending composition has changed over the last 5 to 10 years. It may be useful to cite some comparisons with other countries as possible benchmarks, if they are available. • Estimating marginal returns: can the sector objectives be achieved more efficiently by reallocating resources to other areas? Marginal returns to public spending across spending 29 categories are usually estimated across functional areas. These types of marginal returns analyses typically require time series data and are often done as stand-alone studies. Such studies typically develop a simultaneous equation model to estimate the various effects of government expenditure on production, inequality, and poverty through different channels. Through subsequent differential equations, they estimate the marginal impact of different types of government expenditures. • Cost-benefit analysis (CBA): CBA is a technique used to compare the total costs of a program/project with its benefits as expressed in monetary units. This enables the calculation of the net benefits associated with the program. The technique assumes that a monetary value can be placed on all the costs and benefits of an intervention, including tangible and intangible returns to other people and organizations in addition to those immediately impacted. A major advantage of cost-benefit analysis is that it forces the explicit and systematic consideration of the various factors that influence strategic choices. Box 4 below presents the commonly used criteria to make decisions in a CBA. Box 4: Commonly used decision criteria in CBA analysis Three commonly used criteria to make decisions in CBA are Net Present Value (NPV), Internal Rate of Return (IRR) and the Benefit-Cost Ratio (B/C). When using NPV, decisions are based on whether there is a net benefit or cost to the approach, i.e. total benefits less total costs. Costs and benefits that occur in the future are typically seen as having less value than costs and benefits occurring today. To account for such a time preference, it is necessary to ‘discount’ or reduce the value of future costs or benefits to place them on a par with costs and benefits incurred today. The ‘discount rate’ will vary depending on the sector or industry, but public sector activity generally uses a discount rate of 5-6%. The sum of the benefits of an option minus the sum of the costs, all discounted to the same base date, is the ‘net present value’ of the option. NPV is a very simple and precise performance indicator. A positive NPV, NPV>0, means that the project generates a net benefit (because the sum of the weighted flows of costs and benefits is positive) and is generally desirable either in financial terms or in economic terms. The IRR is the maximum rate of interest that a proposal can afford to pay for resources used which allows the proposal to cover its investment and operating expenses and still break even. In other words, the IRR is the discount rate which will make the NPV of a proposal equal zero. When calculated, it can be compared against a pre-determined target rate of return. If it exceeds this cut-off rate, IRR indicates acceptance of a project proposal. The IRR is an indicator of the relative efficiency of an investment, and should be used with caution. If the sign of the net benefits (benefits minus costs) changes over the project’s lifespan (for example - + - + -) there may be multiple IRRs for a single project. In these cases, the IRR decision rule is impossible to implement. Examples of this type of projects are mines and nuclear power plants, where there is usually a large cash outflow at the end of the project because of decommissioning costs.5 5 European Commission. 2008. Guide to Cost-Benefit Analysis of Investment Projects. 30 The B/C ratio is the ratio of discounted benefits to discounted costs. If it is greater than one, spending proposals will be accepted. Like the IRR, this ratio is independent of the size of the investment, but in contrast to IRR it does not generate ambiguous cases and for this reason it can complement the NPV in ranking projects where budget constraints apply. • Cost-effectiveness analysis (CEA): CEA is used as an alternative to CBA when the benefits of a government program or activity cannot be expressed in monetary terms. It compares the costs of alternative ways of providing similar kinds of outputs. CEA is also used to compare the costs of achieving different degrees of an objective, e.g. the most cost- effective way of increasing literacy rates within a certain cohort of the population. CEA measures costs in a common monetary value (BGN) and the effectiveness of an option in terms of physical units, e.g. BGN spent per life saved. CEA can also be used to analyze effectiveness (see next section). • Cost-utility analysis (CUA): CUA is useful for evaluating and comparing the non- monetary utility or value of programs that aim to reach the same goal. CUA develops an overall measure of utility or value based on the preferences of individuals. A well-known application of cost utility analysis is in the health sector, with the use of Quality Adjusted Life Years (QALYs). The QALY allows each potential program to be measured according to the extent to which it extends life expectancy while also improving the quality of each year lived. Developing this indicator involves determining satisfaction derived from different health states. 54. Data requirements: descriptive statistics; program/project/policy documents, such as plans and reports; sectoral strategies; quantitative and qualitative information on inputs (in particular data on costs of physical resources and human effort needed to produce benefits/outputs), outputs, outcomes and impacts, as well as monetary and non-monetary estimates of their benefits; private sector data; household and other surveys; evaluation reports; comparisons with more advanced countries whose systems display characteristics which Bulgaria aspires to. Technical efficiency 55. Technical efficiency is concerned with achieving maximum outputs with the least cost. Spending reviews can contribute to improving the technical efficiency of public spending by, e.g. comparing actual spending with the minimum spending theoretically sufficient to produce the same output. A spending review might analyze programs which have received unchanged or increasing funding despite decreases in underlying cost drivers. Examples of decreases in cost drivers may include: • A decrease in the number of beneficiaries of a given public service; • Simplifications in the regulatory requirements or requirements from the EU or other international standard setting bodies leading to simplified and less time-consuming processing, controls or documentations requirements; 31 • Available new technology, e.g. introduction of new common IT platforms, electronic signatures, etc. leading to potential efficiencies; • New organizational models, processes, or modes of service delivery leading to potential efficiencies. The analysis could also focus on specific services or functions whose allocations are higher per unit than comparable functions or services in other entities without an obvious reason. In this case, a spending review could include a horizontal analysis of similar functions, typically various support and “back office� functions, across different ministries or other entities and recommends savings. 56. Types of analysis: • Frontier analysis: one of the most frequently used methods for evaluating efficiency is Data Envelopment Analysis (DEA). In DEA, the productivity of a unit is evaluated by comparing the amount of output(s) produced in comparison to the amount of input(s) used. The performance of a unit is calculated by comparing its efficiency with the best observed performance in the data set. This methodology defines efficiency as the distance from the observed input-output combinations to an efficiency frontier, defined as the maximum attainable output for a given input level. Once the DEA efficiency frontier is estimated, both input-inefficiency (excess input consumption to achieve a level of output) and output- inefficiency (output shortfall for a given level of inputs) can be scored. Efficiency is defined as the ratio of the weighted sum of outputs to the weighted sum of inputs and is measured on a scale from 0 to 1 (“efficiency score�). A value of 1 indicates that the unit is the most efficient among the units included in the study. A value less than 1 indicates that the unit is inefficient compared to the most efficient unit. An example of DEA analysis performed during the first phase of the spending review in Bulgaria is presented in Figures 6 and 7. 32 Figure 6: Example of DEA analysis performed during the first phase of the spending review in Bulgaria Government expenditures on public order and safety and trust in local police (from Gallup) Confidence in local police (Gallup World Poll) 1 IS 0.9 DK AT DE 0.8 IE FR FI NO CH SI NL BE ES HR PT LU SE CZ EEIT UK 0.7 HU EL LT 0.6 MT SK LV RO PL CY 0.5 BG 0.4 0.0% 0.5% 1.0% 1.5% 2.0% 2.5% 3.0% Expenditures on public order and safety (% of GDP, 2015) DEA frontier Source: World Bank based on Eurostat and Gallup. Figure 7: Example of DEA analysis performed during the first phase of the spending review in Bulgaria Efficiency score, using DEA method BG CY RO PL LV SK LT EL HU IT EE UK CZ NL BE ES FR SI HR MT PT SE IE DE AT DK CH NO FI IS LU 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Source: World Bank based on Eurostat and Gallup. • Budget deviation analysis: this is another technique used to determine the technical efficiency of budget executing units. It tries to answer questions such as whether budgets are over- or under-spent, how good government spending units are at executing the planned 33 budget, which issues are coming through when comparing expenditure out-turn and budgeted expenditures and how the discrepancies can be explained. • Business processes reviews: this involves breaking down selected business processes into individual tasks, activities or steps, estimating their cost or resource use and analyzing options for simplified (“reengineered�) processes. Cost estimates can be based on time recording (if available) or specific surveys of resource use. • IT systems gap analysis: analysis of current IT systems use can be compared with good practice in the particular field in order to identify the gaps and estimate the potential efficiency gains from automation of business processes currently done manually or further integration of stand-alone systems. IT analysis is often applied together with business process analysis, since automation is one of the key drivers of simplified processing, and hence lower costs of producing a given output. At the same time, re-engineering of underlying business processes are typically a prerequisite for achieving higher technical efficiency through upgrades or replacements of IT systems. • Desk reviews of legal, regulatory or international requirements: another driver of lower costs is to simplify the legal/regulatory basis for a given activity. This requires desk reviews of national and international requirements. In some cases, simplifications may have already occurred in requirements, but may not have been fully implemented in practice. The spending review can thus provide a push for “cashing in� on such simplifications. In other cases, the spending review can provide suggestions for new simplifications in the legal environment. 57. Data requirements: budget vs. actual expenditures and performance information at output level of the spending under review. Sources of data could be sector strategies and reports, program budgets, legal and regulatory documents, in some cases collective agreements, national and international statistics, and surveys. Functional coherence 58. Often savings can be achieved by identifying duplicative or inefficient functions carried out by government entities. A spending review may use functional analysis in two ways. First, functional coherence can be defined as a problem area in and of itself used to identify savings or reallocation options as a result of consolidating functions, doing away with duplication, terminating non-core functions, etc. Second, functional analysis can be used to complement analysis in other problem areas to assess whether the current organizational and functional set up of the sector/policy/program/entity under review is conducive to achieving the savings target. It is recommended to conduct the functional analysis in a participatory fashion and validate the results with the institutions under review. The functional analysis should result in actionable recommendations as to the organizational and staffing changes needed in order to produce savings, proposed streamlining of functions, likely resource requirements to implement the recommendations, possible legislative changes, training needs and change management program. Modeling of the costs and benefits of the different scenarios (e.g. merger of departments, staff cuts, etc.) can also be done. 34 59. Types of analysis: the functional analysis could begin by a horizontal review and mapping of all institutions involved in the sector under review, together with a review of their mandates, goals and objectives. The next step would be to zoom in on the key institutions responsible for implementing the spending under review and assess whether their internal structure is well suited to efficient and effective service delivery. Comparisons with similar structures in the country or in other countries can be useful. Decision trees can be helpful in analyzing which functions could potentially be reduced in volume or dropped and whether there are any overlapping functions which could be outsourced/(de)centralized/optimized/eliminated. Gap analysis of missing or misplaced functions can be conducted as well. Annex 5 provides examples of decision trees which can be used when performing analysis of functional coherence. 60. Data requirements: the primary sources of data for this analysis are legislative documents establishing the mandates and organization of the reviewed institutions, internal rules of organization and organizational charts, employment registries, job characteristics, and in-depth interviews with managers. A survey might also be used to collect more detailed and disaggregated data. The data collection effort should focus on collecting the following data: functions by classification (e.g. policy functions, co-ordination, supervision and performance monitoring functions, service delivery functions, support functions, and regulatory functions) and spending levels per function; the products/outputs of each function; staffing levels per function and per institution; employment types and contracts; unit labor costs; managerial span of control and whether managerial responsibilities are balanced and sustainable; management systems and culture for policy coordination and decision-making; staff workloads and qualifications; challenges related to recruitment, retention, and motivation of staff; performance incentives; arrangements for internal control, audit and M&E functions. B. Developing and analyzing savings/reallocation options 61. Developing options for savings or reallocations is at the heart of the spending review process. This subsection provides guidance on the process of developing such options. Actors 62. In most countries, the development of options for savings or reallocations is primarily undertaken by the spending ministries and the MoF, while external consultants and advisers play a supporting role. It is broadly accepted that government officials should play a central role in the conduct of spending reviews. Although the use of outside experts is very widespread, these outside experts in most cases serve as advisors to (or sometimes members of) review teams which are themselves directed by, and predominantly composed of, government officials. Using government officials makes particular sense in order to accumulate knowledge and sustain the in- house capacity to conduct spending reviews. Nevertheless, it should be recognized that private sector experts can potentially play a major role in the search for efficiency savings. Business process or system reviews – e.g. of procurement processes or IT systems – is an area where spending review work is often rather generic, and where private sector expertise can add value. 63. In the case of Bulgaria, the savings/reallocation options will be developed by the 35 Working Groups. In line with good international practice, the Working Groups are composed of ministry staff and MoF is responsible for conducting the analysis. External experts might also be included. The identified savings options will be presented for decision to the CoM (see next section). 64. It is also recommended to validate savings/reallocation options by conducting field visits, and consulting with operational staff and the general public. Consulting operational staff limits the risk of a disconnect between recommendations and the actual practice, which often deviates from legal and procedural provisions. By collecting real-life evidence, field visits can be beneficial to the quality and credibility of the reform options put forward. For example, a review of healthcare polices and spending might benefit from testing reform ideas with the directors of the largest hospitals. Following the global financial crisis, some countries have even decided to consult end-users to generate ideas for savings options or to give end-users a chance to voice their preferences between different proposals. This could be done through online surveys, dedicated websites or through more focused questionnaires targeted to an informed audience, such as the target population or officials in charge of the spending. As an example, the UK government invited input from the public for the 2010 Comprehensive spending review through an open invitation labelled the “Spending Challenge�. Process 65. An explicit set of review criteria should be developed to guide the development of savings/reallocation options. This is particularly important for strategic savings options as they involve determining which services or transfer payments can be eliminated or scaled back. Particularly in countries where the spending review process is based on bottom-up reviews (see next section); it has proven to be useful to make these criteria completely explicit at the outset of the spending review process. It is also important to ensure political endorsement of the criteria. 66. The criteria should be based on the problem areas defined for the spending review in question and should emphasize feasibility of implementation. The choice of criteria will be derived from the specific problem areas addressed in a spending review, i.e. allocative and technical efficiency, effectiveness and functional coherence. In addition, criteria should include a consideration for the feasibility of implementing the proposed reform options. Feasibility refers to multiple factors, such as the one-off cost of implementation (e.g. initial investment in IT, staff training), the risk (administrative or social resistance etc.), the time horizon (including the political timing), the pre-requisites (e.g. amending a law), the negative spillovers on other expenditure lines, etc. The capacity of an administration to simultaneously implement multiple reforms – even if each of them rates well on feasibility – has to be taken into account as well. Given the timeframe of a spending review is often tight, implementation of proposals usually requires further work to be undertaken beyond the review process. 67. Once criteria have been defined, options for savings or reallocations for each individual spending review can be rated and mapped on a matrix. The matrix could be structured by saving and feasibility, and potentially additional dimensions. In Figure 10 below, the 36 size of the bubble reflects the estimated consequence for the quality of service – red is negative, green is neutral or positive. Each mix leads to a specific conclusion: stop implementation if both saving expectations and feasibility are low, prioritize implementation if feasibility is high, even with limited savings, etc. For some individual spending reviews, none of the options will be appealing, leading to the conclusion not to pursue these reforms. Figure 8: Prioritization matrix Source: European Commission (2014). 68. When formulating reform options, the potential performance of alternative expenditure should be assessed. One approach to this is to perform sensitivity analysis to simulate the impact of a lower level of spending and/or of an alternative allocation of spending on policy objectives. Another approach is to make one major cost driver vary and deduct the overall savings. Yet another option is to start by redefining the policy (e.g. by removing one redundant step in the process, or pooling support functions) and then estimate the new spending level (from scratch or by difference). In any case, savings should be net of new recurrent costs potentially generated by the new target process – after the one-off implementation costs. 69. Once the spending review savings or reallocation options have been developed, it is important to lock them into fixed multi-year budgets for the ministries concerned. Many of the savings identified as a result of the review can only materialize be realized over a number of years, and some will require upfront investment or other costs (e.g. redundancy payments) in order to be released. If formal budget decision-making and management remains on a purely annual basis, then there is no guarantee that money given to ministries in the first year will yield reductions in spending in the second and third year. Therefore, the only effective means of ensuring that savings are realized over the coming years is to reflect savings (and any initial spending necessary to realize them) in the medium-term budget targets for the ministry concerned. It is important that the annual budget for each fiscal year is derived from a roll-forward of the medium-term budget target from the previous year, so that savings agreed in previous years are always reflected in the 37 baseline spending for a given fiscal year. The multi-year financial predictability given to the line ministry will also help them plan ahead and take the decisions needed to deliver the savings. Integrating savings into medium term budget plans will also require the Budget Department of the MoF to engage in the spending review process and gain an early understanding of the budgetary implications of different reform options. 38 VI. DECISIONS ON SPENDING REVIEWS A. Process of selection of savings or reallocations and approval of spending reviews 70. The final decision on which savings or reallocation measures to implement must, in general, be made by the CoM. The responsibility for the selection of savings or reallocation options should be given to the CoM to ensure there is sufficient weight to overcome resistance during the implementation phase. The locus of political decision-making with respect to savings measures also depends to some extent upon the nature of spending reviews. If the focus is upon strategic savings, the political sensitivity of decisions to cut or scale back programs means that such decisions will almost always be taken at the very top. By contrast, for a spending review leading to efficiency savings only, the final decision could be left to, say, the Minister of Finance. 71. Irrespective of the institutional set up for spending reviews, the challenge function of the MoF is of fundamental importance in getting the spending ministries to collaborate. Twelve of the fourteen countries participating in a 2012 OECD survey on spending reviews reported that the MoF was the (or one of the) main institutions responsible for the supervision and review of reports prepared during the spending review process. It is essential, therefore, that the MoF actively analyzes the merits or calls for modifications of savings options put forward by the spending ministries. In a joint review process, such as the one proposed for Bulgaria, it will be the MoF members of the working groups and steering committee, who bear prime responsibility for carrying out the challenge function. This makes it vital that the MoF commits substantial resources to the work of the working groups and steering committee. In some countries, like Ireland and Slovakia, there is a separate permanent unit established within the MoF structure charged with carrying out spending reviews. 72. According to the proposed institutional set up for spending reviews in Bulgaria (see Annex 3) the process of deciding on savings options and approving the spending review report is as follows: Working Groups prepare a menu of reform options and identify the savings and reallocations. The MoF participates actively in conducting the analysis and developing the savings options. The Steering Committee provides strategic guidance in this process, and once it endorses the Working Group’s recommendations and savings options, it submits them to the CoM for consideration and approval. The CoM, having examined the full package of spending review documents, makes final decisions on savings and reallocation options and approves the spending review report. B. Information required for ministers to make an informed decision 73. To aide decision makers in weighing the various trade-offs, each policy proposal for savings should include the following information: i) the fiscal impact for the budget and three forward years, with costings made using consistent methodologies and assumptions across programs; ii) the public policy rationale for the proposal; iii) the impact of the policy proposal on affected groups and any social, regional or economic consequences; iv) the legislative requirements, implementation arrangements (including whether the measure could be 39 implemented as part of the current or next year’s budget or requires further development), and the process for consultation, if required; and v) a roadmap with a calendar and the leader of the implementation phase. An example of a template for deciding on saving options is presented in Table 4 below. Table 4: Savings decision making template Program Budget Impact ($m) 2017 2018 2019 2020 2021 Total Recommendations of Baseline Budget the Steering Committee cost (-) / saving (+) FY1 FY2 FY3 for CoM’s consideration Option A – Summary of policy change Option B- Summary of policy change Option C – Summary of policy change Memo: Total resourcing Description Policy Description: [Describe policy proposal] Public rationale: [Provide public policy rationale for policy proposal] Who will be affected: [Detail impact of policy change: who will be impacted; the number impacted; the financial impact on those affected; any regional or sectoral impacts.] Other sensitivities: [Identify other sensitivities e.g. Are there any adverse social / economic consequences of the change? Is the policy change consistent with previous government policy commitments? Are there consequences for other programs?] Implementation: [Legislative change will / will not be required. When will legislation need to be implemented by? Does the policy change require consultation with third parties prior to implementation? Who will lead implementation and in what timeframe?] Likely third-party reaction: [Detail stakeholder reaction: are there any lobby groups/think tanks/institutions who will advocate for/strongly oppose the change] Source: IMF (2014). 40 VII. REPORTING AND COMMUNICATION A. Presentation of data, analysis and recommendations 74. It is recommended that the Government agrees on a common structure and format of the spending review reports. The reports should be as concise as possible (see Annex 6 for an example of a spending review report outline). As a minimum, each report should contain an executive summary, table of contents, a chapter with an overview of the sector, methodology and key questions to be answered by the review; a chapter focused on analyzing the key questions; and a chapter with conclusions, policy options and recommendations. The annexes should include a list of references, the spending review TOR and the composition of the working group. Additional communication products for different audiences might be created as well (see next section). 75. A well-written spending review report has the following characteristics: • a good structure, reflecting the logical order in which the reviewer has clustered the analysis; • no gaps, i.e. the research questions should be worked through in a logical order of presentation; • short sentences, with one idea per paragraph, using topic sentences (i.e. capturing the main message of the paragraph) to begin each paragraph; • simple and unambiguous wording; and • coherence between analysis and conclusions. 76. Special attention should be given to the presentation of recommendations. Recommendations are the actions identified as a result of the spending review, which are necessary to achieve savings or reallocations. 41 Box 5: Good practice in spending review report writing When writing conclusions and recommendations, the following issues should be considered: • be clear; • be specific; • focus and prioritize the conclusions; • be systematic in presenting the arguments that support a conclusion; • make any assumptions explicit; • the criteria and reference standards for judgments should be explicitly stated; • avoid duplications and contradictions; • check that all recommended actions proposed are in line with applicable rules and Regulations; and • generalizations should be kept to a minimum. Source: Government of Ireland. (2007). Value for Money and Policy Review Initiative Guidance Manual. 77. The spending review itself is only the beginning of the reform process. The actual implementation of the recommendations is the only outcome that qualifies the review as a success. The level of political will (and realism), and the absorptive capacity of the institutions under review to implement the recommendations will be two factors that should be taken into consideration when preparing the final recommendations. In addition, it is highly likely that there will be strong resistance to many of the recommendations among key stakeholders within the institutions affected. This is likely to be the case where insufficient attention has been devoted to a change management approach throughout all phases of the spending review. 78. Recommendations must be constructive and time-bound and should identify who will implement them. A recommendation should ideally be broken down into its logical implementation steps. Recommendations should address the root causes of problems identified by the spending review and their effects. Generic recommendation should be avoided. For further tips on how to write good conclusions and recommendations, see Box 4. 79. Estimating savings, ranking and phasing of recommendations is good practice. Costing of recommendations is a necessary part of a spending review although usually this will only be possible in the form of an approximation. The inclusion of an estimated savings as a result of the proposed recommendation and potential costs is an important part of the prioritization of recommendations. A spending review should not, as a general rule, recommend an increased resource allocation for the programs concerned. Ranking recommendations is critical. A laundry list of valid issues is a list too long to act on. For policymakers, identify a limited number of key issues – three to five per sector – where getting some traction is most critical. “Nesting� recommendations in hierarchies can be helpful. The main recommendation (e.g. “improve equity of social assistance�) can be aimed at policymakers, with the technical specifics aimed at technocrats (e.g. “change the balance between programs A and B, improve program C by taking actions 1, 2 and 3�). Phasing – short, medium and long-term—is another way to “chunk� recommendations into a digestible form. 42 80. Ideally, recommendations should be accompanied by an implementation plan. The implementation plan is an essential output from the spending review. This plan will help the government decide which recommendations will be implemented and will provide a framework for planning for implementation that can be used to secure stakeholder buy in and necessary resources. At the stage of completing the spending review, the implementation plan can only be indicative in nature because it will need to be subjected to a process of political decision making that will pave the way for implementation. 81. Last but not least, it is important to establish arrangements for monitoring the implementation of spending review recommendations. Establishing a framework for monitoring and evaluation is useful to ensure implementation of results, and to link spending reviews to the wider budget and strategic planning process. It also reflects the concept that spending review is an iterative and continuous process. The arrangements for monitoring need to track and report publicly on the realization of “cashable� financial savings from things like workforce reductions, lower input costs, and reductions in overheads. This requires the spending review to provide a manageable number of clear objectives. The MoF would typically take the lead in monitoring as part of the budget process and report to the CoM. Regular, joint MoF-CoM stock-take meetings should be conducted with the ministers concerned to ensure that the implementation is on track. B. Dissemination 82. The MoF, together with the line ministry(-ies) are responsible for communication and dissemination of spending reviews. In line with good international practice, transparency is encouraged throughout the process, starting from when a spending review is initiated, through reporting on progress, to reporting on implementation of savings. Communication and dissemination are also important for securing stakeholders’ support, as successful implementation of spending review recommendations may involve working closely with different stakeholders. Transparency also increases the accountability of government towards the taxpayers. At a minimum, the MoF should announce the initiation of a spending review on its website, publish the Terms of Reference and create a repository of completed spending review reports published on its website.6 Line ministries should also publish reports examining their spending on their websites. Before a report is finalized and published, the MoF should ensure that procedural guidelines have been followed and the technical quality of the report meets minimum standards, and adheres to the terms of reference. It is recommended that regular reports on progress in implementing the recommendations of the spending review are published as well. 83. Communication and dissemination should ideally be based on a strategy that differentiates different audiences and prepares targeted communication for each audience. A good dissemination strategy usually: (i) determines the target audience against the objectives of the work being undertaken (e.g. policy makers, donors, researchers, or other stakeholders), and; (ii) determines what type of policy dissemination activities are needed such as press interviews, 6 For example, the MoF in Slovakia has a dedicated section on its website with a collection of spending review reports and other useful materials: http://www.finance.gov.sk/Default.aspx?CatID=11190. 43 research reports, and policy briefs. Different communication products should be used for different audiences (Table 5). For the experts, the full spending review report and its annexes, should be published. For an interested, informed audience, a shorter, e.g. 10-page report could be published, containing the essential findings, enhanced with color graphics aimed at policy makers and journalists. For the popular press and general public, a 2-page press release containing only the main messages through sound bites or bullet point type information is recommended. This approach appeals to the mass media and high-level policy makers who do not have time to read longer reports. Table 5: Matching publication to audience Type of Publication Audience Research report Expert Policy report Insider Issues brief Interested Press release Layperson Source: adapted from World Bank (2011). 44 ANNEXES Annex 1: References Better Evaluation (2017). Cost-Benefit Analysis, available at: http://www.betterevaluation.org/sq/evaluation-options/CostBenefitAnalysis?page=0%2C2 Better Evaluation (2017). Cost-Utility Analysis, available at: https://www.betterevaluation.org/evaluation-options/cost_utility_analysis Council of Ministers, Bulgaria (2017): Decision on the budgetary procedure for 2018, January 2017. European Commission. 2008. Guide to Cost-Benefit Analysis of Investment Projects, available at: http://ec.europa.eu/regional_policy/sources/docgener/guides/cost/guide2008_en.pdf Government of Ireland (2007). Value for Money and Policy Review Initiative Guidance Manual, available at http://publicspendingcode.per.gov.ie/wp-content/uploads/2011/11/VFM-Guidance- Manual-2007.pdf; IMF (2015). Republic of Slovenia: Technical Assistance Report-Establishing a Spending Review Process, available at: https://www.imf.org/en/Publications/CR/Issues/2016/12/31/Republic-of- Slovenia-Technical-Assistance-Report-Establishing-a-Spending-Review-Process-43289; IMF (2016). Slovak Republic. Spending Reviews of Health, Transport, and Information Technology, available at: http://www.finance.gov.sk/LoadDocument.aspx?categoryId=11154&documentId=14891; Manning, N., Parison, N. (2001). Modernization of the Federal Civil Service: Methodology for Functional Reviews, available at: http://www1.worldbank.org/publicsector/civilservice/ACSRCourse2007/Session%208/RussiaFu nctional_reviews_final_eng.pdf The Ministry of Finance of Bulgaria (2017): Instructions on the preparation and presentation of budgetary forecasts for the period 2018-20, February 2017. OECD (2013). Spending Reviews, available at http://www.oecd.org/officialdocuments/publicdisplaydocumentpdf/?cote=GOV/PGC/SBO(2013) 6&doclanguage=en Robinson, M. (2014), Spending Reviews, available at http://www.oecd- ilibrary.org/governance/spending-reviews_budget-13- 5jz14bz8p2hd;jsessionid=3c7mgwgrlqln9.x-oecd-live-02; Vandierendonck, C. (2014): Public Spending Reviews: Design, Conduct, Implementation; EU Economic Papers 525/2014), available at http://ec.europa.eu/economy_finance/publications/economic_paper/2014/pdf/ecp525_en.pdf; 45 World Bank (2011). Practitioners’ Toolkit for Agriculture Public Expenditure Analysis, available at: http://siteresources.worldbank.org/INTARD/825826- 1111406330465/23065905/AgToolkit11_web_ePDF.pdf 46 Annex 2: Glossary of terms Baseline Estimated expenditure associated with carrying forward government Expenditure programs and activities with no changes in regulation or policies Benefit-Cost The Benefit-Cost Ratio is the ratio of discounted benefits to discounted Ratio (B/C) costs to achieve a certain policy objective. Cost-benefit Comparison of the costs and benefits associated with alternative ways Analysis of achieving a specific objective. Monetary values are assigned to both costs and benefits. Cost-effectiveness A comparison of the costs and benefits associated with alternative ways analysis of providing similar kinds of outputs, but where the benefits are expressed in terms of outputs or outcomes to be achieved. This type of analysis is often used as an alternative to CBA, when benefits cannot be expressed in monetary terms. Cost-utility analysis A comparison of the costs and benefits associated with alternative ways of providing similar kinds of outputs, but where the benefits are expressed in terms of utility or value based on the preferences of individuals. Discount rate A discount rate is applied to anticipated costs and benefits of a project over the duration or ‘life span’ of the project to convert the value of a return in the future into today’s value, since the value of money or goods in the present is viewed as higher than the expected value of goods and financial returns in the future. The IRR is the discount rate which will make the NPV of a proposal equal zero. Effectiveness The extent to which the objectives of an intervention (policy/program/project) are achieved. Efficiency A measure of how economically resources/inputs (funds, expertise, time, etc.) are converted to results Evaluation The systematic and objective assessment of an on-going or completed project, program or policy, its design, implementation and results Impact Primary and secondary long-term effects produced by an intervention, directly or indirectly, intended or unintended Indicator Quantitative or qualitative factor or variable that provides a simple and reliable means to measure achievement, to reflect the changes connected to an intervention, or to help assess the performance of a development actor Inputs The financial, human, and material resources used for the development intervention Internal Rate Maximum rate of interest that a proposal can afford to pay for resources of Return used which allows the proposal to cover its investment and operating (IRR) expenses and still break even. In other words, the IRR is the discount rate which will make the NPV of a proposal equal zero 47 Medium Term Medium Term Budget projection for aggregated budget headings for Budget Forecast the fiscal year plus three out years Net Present NPV is the sum of the discounted benefits less the sum of the Value (NPV) discounted costs. Outcome The likely or achieved short-term and medium-term effects of an intervention’s outputs. Outputs The products, capital goods and services which result from an intervention; may also include changes resulting from the intervention which are relevant to the achievement of outcomes. Savings Measures leading to reductions in expenditure for a given budget heading. Efficiency savings: savings as a result of providing a given output or set of outputs at lower cost Strategic savings: savings as a result of scaling back on amount or quality of outputs produced Spending review The process of the process of identifying and proposing measures for reducing and/or reallocating government spending, based on the systematic scrutiny of baseline expenditure. 48 Annex 3: A Model Terms of Reference for a spending review This annex contains common elements typically included in Terms of Reference for a SR. It is important that the TOR are clearly formulated and there is a shared understanding and agreement on their contents. It is recommended that the Government adopts a common format for the TOR of SRs in order to ensure consistency. The template could be tailored to the specific needs of each individual SR, but would contain the following elements as a minimum: 1. Context 2. Government request 3. Rationale for SR 4. Objectives and key research questions 5. Scope, baseline and targeted savings 6. Methodology 7. Deliverables 8. Institutional set-up, mandates and reporting lines 9. Resources/budget 10. Team composition 11. Timetable 12. Quality control and peer reviewers 13. Annexes 49 Annex 4: Summary of SR roles and responsibilities Council of Ministers: • Adopts the Decision to launch a SR (framework, SR areas, objectives, timeline and target); • Appoints the Steering Committee and gives it and the MOF a mandate for SR; • Decides on quantifiable saving target(s) to be achieved over which period; • Adopts the recommendations and communicates to the general public; • Provides political leadership and communicates publicly; • Makes final decisions on savings and reallocation options. Steering Committee: comprised of senior staff from the MOF (economic policy, budget, expenditure departments), responsible line ministries with the Deputy Minister of Finance responsible for spending reviews as the Chair. • Provides strategic direction and mandate to the Working Groups; • Provides guidance to the working groups throughout the process, including through the potential reform options based on the diagnostic, potential impact and feasibility of the proposed options; • Holds progress meetings (once every month); • Reviews and challenges proposals by the Working Groups; • Ensures line ministries’ participation in the process; • Communicates internally; • Submits the recommendations/policy options to the CoM for considerations and approval. Chair of the Steering Committee: • Coordinates and supervises the work of the Steering Committee and the Working Groups; • Convenes and conducts working meetings of the Steering Committee to ensure timely delivery of reports in accordance with the Decision; • Seeks revisions of the draft reports if the Steering Committee finds it necessary; • Approves Reports of Working Groups, by the prevailing voice if the majority vote has not been secured by the Steering Committee; • Submits and presents the Reports to the CoM for consideration and adoption. Steering Committee members: • Participate as active members in the working sessions of the Steering Committee; • Provide technical and other advice to the Working Groups if there are some doubts about the implementation of their tasks; 50 • Decide on the proposal by individual Working Groups, proactively guiding their work, monitoring early results; • Supervise the work of Working Groups based on weekly notes on the progress of the work and to propose measures to ensure the timely delivery of reports. Working Groups: chaired by the Spending Review Leader (a MOF staff working under the direction of senior MOF official), and involving technical, planning and budgeting staff of the MOF and relevant line ministries (budget analysts, internal auditors, strategic units, etc.), as well as selected group of local government representatives. If relevant expertise exists, external experts (auditors, consultants, academia) are also welcome. It is important that the working groups are non-political and consist of civil servants and independent experts. The Working Groups could be supported by a small secretariat in the MOF. The Steering Committee and Working Groups would have regular coordination meetings to allow the spending review leaders to monitor progress and the Chair to provide guidance. It is recommended that no working group member may “veto� another member’s idea. This ensures that the CoM (and ultimately the wider public) is presented with a range of alternative policy options. • Conducts analysis; • Prepares reform options within the defined target savings; • Prepares quality, social and risk assessment, as well as legal review; • Prepares other qualitative recommendations. Spending Review Leader/Head of the Working Group: • Convenes and conducts meetings of Working Groups to ensure timely delivery of the report to the Steering Committee; • Proposes topics of working meetings; • Plans and allocates tasks, and setting clear and realistic goals for the Working Group; • Directs the members in carrying out the tasks assigned to them; • Supervises collection of data and conducting analysis; • Requests technical and other instructions of the Steering Committee if there are some doubts about the implementation of the tasks; • Seeks rapid access to the required documents, registers and other necessary data for the analysis; • Submits weekly notes on the progress of the work to inform the Steering Committee; • Submits the report to the Steering Committee by the agreed timeline. Working Group members: • Participate actively in the work of the Working Group and the implementation of the assigned tasks in fast and efficient manner; 51 • Collect relevant data from relevant institutions; • Identify employees of state or local administration bodies who can best assist in the work of the Working Group; • Identify documents, already completed analysis, reports and other working documents that can contribute to the work of the Working Group; • Inform the Chair in the event of difficulties in collecting relevant data; • Provide comments on the draft analysis; • Advise on options for reforms; • Advise on potential costs/risks/sequencing; • Participate in finalization of the report. Ministry of Finance • Drafts the Decision for the COM approval (targets, spending review areas, timeline, Steering Committee and Working Groups TORs); • Prepares baseline data for SR areas; • Shares audit or analytical reports, performance indicators; • Provides space and secretariat support. Communication between members of the committees, as a rule, takes place by e-mail. A report on the results of spending review shall be submitted to the Steering Committee by email and in writing. The Head of the Working Group makes note of the progress of the work on a weekly basis and submits them to the Chair. The note contains information on substance of the discussion, the meeting time and place, members present and the main conclusions. It also contains information on the progress of the planned activities. 52 Annex 5: Analytical questions typically raised in spending reviews This annex has two parts: the first part provides examples of general analytical questions raised in SRs in the UK and Slovenia, while the second part provides examples of decision trees which would assist in conducting a functional analysis as part of a SR. Part I. Examples of analytical questions United Kingdom (2010) Slovenia (2015) • Is the activity essential for meeting Alignment: government priorities? • Is the activity/program aligned with the • Is it necessary for the government to fund priorities of government? this activity? Effectiveness: • Does the activity produce a substantial • To what extent does the program achieve economic value? its objectives? • Can the activity be focused on the • Are there other ways to achieve these population groups which need it most? objectives? • How can the outputs be delivered at a Efficiency: lower cost? • Are there measures which can improve • How can the outputs be delivered more the efficiency of the program? effectively? • Is there room to improve service delivery • Can the activity be performed by a non- and simplify administrative arrangements? state provider? • Can the program be better targeted? • Could non-state providers be paid to carry • Is it possible to subcontract or involve out the activity according to the results non-state providers at a lower cost? they achieve? • Are there any overlaps or duplication with other programs? Source: World Bank. Part II. Examples of decision trees for analysis of functional coherence 53 Decision tree 1- necessary functions Source: adapted from Manning and Parison (2001). 54 Decision tree 2- Selecting operating environments Source: adapted from Manning and Parison (2001). 55 Annex 6: Suggested Outline of a Spending Review This annex contains a generic outline of the key elements which should ideally be included in every report: Title or Cover Page Table of Contents Abbreviations/ Acronyms Executive summary • Context • Analysis • Conclusions and recommendations Section 1: Overview This section could include the following subsections: - Context - Legal basis - Policy goals - Programs supporting the achievement of the policy goals and result indicators - Overview of revenues and expenditure in the sector - Key actors, responsibilities and governance - Reforms undertaken so far - Remaining challenges - Methodology and research questions - Structure of SR report In terms of Methodology and research questions, it is important to outline the three key questions to be answered by the SR as well as an explanation of why these three policy issues are important. Answering these questions will help formulate policy recommendations aimed at improving efficiency and effectiveness of the program/policy under review. For example, the paragraph introducing the key questions could be structured in the following way: “Based on the reform progresses made to date and remaining challenges, this Spending Review (SR) examines three key policy questions with an aim to contribute to enhancing efficiency and effectiveness in topic X: 1. X 2. Y 3. Z� 56 For example, in the SR on Public Order and Safety, the three questions to be answered were: 1) What changes in the allocation of labor could increase government effectiveness in public order and safety within the existing fiscal envelope? 2) What changes in the pay policies could increase government effectiveness in public order and safety within the existing fiscal envelope? 3) What changes in the allocation of complementary inputs (vehicles, IT systems, etc.) would increase government effectiveness in public order and safety within the existing fiscal envelope?). Questions asked in SRs on police performed by other countries include, for example: 1) what are the drivers of efficiencies which can provide for additional policing hours within existing resources? 2) what are the metrics required to assess demand for policing services and to measure the results of increased police investment in the context of potential new approaches to resource allocation? The three questions guiding the analysis of the SR on Waste Management were: 1) What are the key inefficiencies in collection, separation/recycling and disposal and how can these be addressed within the existing envelope? 2) What are the pros and cons of outsourcing vs in-house waste management services? 3) How can effectiveness of waste collection, separation and disposal be improved within existing envelope? Section 2: Analysis This section should also be structured around the three key analytical questions the SR will answer. For each question, the analysis could include: 1) presentation of the problem; 2) benchmarking (international and within country); 3) efficiency; 4) effectiveness. This section outlines also the analytical framework and methodology used in the analysis. For example, a SR from the Netherlands on police used the following framework for the purpose of the analysis: Section 3: Conclusions, Policy Options and Recommendations This section should summarize the main conclusions of the SR with a storyline connecting them. This section should also outline the policy options for savings and make recommendations focused on the three key policy issues and group them around each of the three policy issues. Recommendations should be prioritized and should focus on how to implement reforms, who should be involved, and sequenced into short-term versus long-term. The policy options should include budgetary estimates. In general, a spending review provides options for spending reductions or more value for money, and is not meant to present options that will result in structural additional costs. The policy options can be contrary to the current policies and plans of the current government. In some cases, the conclusion section can be merged with the executive summary. 57