Report No. 34582-AF Afghanistan Managing Public Finances for Development (In Five Volumes) Volume V: Improving Public Financial Management in the Security Sector December 22, 2005 Poverty Reduction and Economic Management Sector Unit South Asia Region Document of the World Bank Table of Contents ACKNOWLEDGEMENTS IV EXECUTIVE SUMMARY ........................................................................................................................ ...................................................................................................................... V Introduction .......................................................................................................................................... v Conceptual Framework......................................................................................................................... v Afghanistan's Security Sector and Recent Progress.......................................................................... vii K e y Challenges................................................................................................................................. viii ... Toward a Fiscally Sustainable Security Sector ................................................................................... ix Budget Execution for Effective and Efficient UtilizationofFunds .................................................... Strategic Prioritization and Sequencing................................................................................................ x xi Summary of Recommendations........................................................................................................ xiii ... CHAPTER1 SECURITY,DEVELOPMENT,AND PFMPERSPECTIVE . ........................................ A.INTRODUCTION 1 ..................................................................................................................................... 1 Background and Rationale.................................................................................................................... 1 Objectives and Chapter Content ........................................................................................................... 2 B.DefinitionANDDEVELOPMENT ofthe Security Sector .......................................................................................................... 3 SECURITY ............................................................................................................ c.KEYPFMPRINCIPLES 4 7 Aggregate Budgetary Outcomes........................................................................................................... AND APPLICATION TO THE SECURITYSECTOR................................................. 8 Strategic Allocation of Resources Across Sectors and Programs......................................................... 9 D.EffectiveDELIVERY and Efficient Utilization of Budgetary Resources .............................................................. 10 SERVICE CHAPTER2 AFGHANISTAN'S SECURITYSECTORAND KEYDEVELOPMENTISSUES 15 . PERSPECTIVE...................................................................................................... 11 ... A.HISTORICAL BACKGROUND RECENTPROGRESS........................................................................ 15 AND Historical Background........................................................................................................................ 15 Progress in Security Sector Strengthening and Ref0rm...................................................................... 17 B.OVERVIEWOFAFGHANISTAN'S SECURITYSECTOR........................................................................... 19 Legal Foundation................................................................................................................................ 19 Main Components and Size................................................................................................................ 19 c.INSTITUTIONALSTRUCTUREANDCOORDINATION............................................................................ 23 Salient Features................................................................................................................................... 21 Institutional and Administrative Reforms .......................................................................................... 23 D.CoordinationinISSUES PERSPECTIVE.......................................................... the Security Sector.................................................................................................... 26 KEYSECURITY FROMA DEVELOPMENT 29 CHAPTER3 PUBLICFINANCEMANAGEMENTAND AFGHANISTAN'SSECURITY SECTOR .................................................................................................................................................... . A.IMPLICATIONS FORPFMPRINCIPLESFORAFGHANISTAN ................................................................. 33 33 B.THESECURITYSECTORAND THEBUDGET PROCESS ......................................................................... 36 Budget Formulation............................................................................................................................ 36 Budget FundingArrangements........................................................................................................... 39 CHAPTER4 TOWARDA FISCALLYSUSTAINABLESECURITYSECTOR . ............................. 42 A.RECENT TRENDSINSECURITY SECTOREXPENDITURES .................................................................... 42 B.FUTUREEXPENDITURE DRIVERS PROGRAMMEDEXPANSION AND .................................................... 43 c.PROJECTED COSTS. FISCALBURDEN. AND LACK SUSTAINABILITY OF .............................................. 46 D.PROGRESSINGTOWARDA SUSTAINABLESECURITYBUDGET 47 CHAPTER5 STRATEGICPRIORITIZATION AND SEQUENCING . ........................................................... ............................................ A.THESTRUCTUREOF SECURITY SECTOREXPENDITURES 50 ................................................................... 50 B.IMBALANCESANDANOMALIES.SEQUENCINGISSUES. ..................... 51 Apparent Imbalances and Anomalies ................................................................................................. AND CONTRIBUTINGFACTORS 52 i Sequencing Issues............................................................................................................................... 53 ContributingFactors........................................................................................................................... 54 c.USINGTHE CHAPTER6 BUDGETEXECUTIONFOREFFECTIVEAND EFFICIENT USEOFFUNDS 60 . BUDGET AS A PRIORITIZATION TOOL ............................................................................... 55.... A.KEYPFMPRINCIPLES FORBUDGET EXECUTION ............................................................................... 60 B.STRENGTHENING BUDGET EXECUTION .............................................................................................. 61 c.TOWARD 64 CHAPTER7 CHALLENGES,PRIORITIES,AND RECOMMENDATIONS .PERFORMANCE MEASUREMENT SECURITYSECTOR ............................................... INTHE ................................. 67 A.SUMMARY OFKEYCHALLENGES....................................................................................................... B.MAINRECOMMENDATIONS................................................................................................................ 67 67 c.ROLEOF DONORS REFERENCES .......................................................................................................................................... ............................................................................................................................... 69 70 BOXES Chapter 1 Box 1. 1:Definition o fthe Security Sector ................................................................................................... 3 Box 1 2: Pnnciples of SoundBudgeting and Financial Management.......................................................... . 8 Chapter 2 Box 2 1:MDGGoal 9 Enhancing Security................................................................................................ . 18 Box 2. 2: LegalFoundationof Afghanistan's Security Sector.................................................................... 19 Box 2 3:Provincial Reconstruction Teams (PRTs) ................................................................................... 21 Box 2 4: Security Policy Coordination at Different Levels ....................................................................... .. 28 B o x 3. 1:Overview of the L a w and Order Trust Fundfor Afghanistan (LOTFA)..................................... Chapter 3 40 Box 3.2: Findings of a LOTFAMonitoringMission................................................................................. 40 Box 3 3:Fundingfor Counter Narcotics Law Enforcement ...................................................................... . 41 B o x 5. 1:The Role of the Budget inDeterminingPriorities inDifferent Sectors ...................................... Chapter 5 56 Box 5.2: H o w to Prioritize?........................................................................................................................ 57 B o x 5.4: Equippingthe Police Force.......................................................................................................... Box 5.3: Socio-Economic Cost-Benefit Analysis o fLandmine Clearance................................................ 58 59 Chapter 6 Box 6 1:Contracting for Landmine Clearance........................................................................................... . 63 FIGURES Chapter 1 Figure 1. 1: Security. Rule of Law. and the InformalEquilibrium............................................................... 6 Figure 1.2: Process of Conducting Defense Reviews/Developing Defense Policy Frameworks................9 Figure 1.3: Service Delivery Framework ................................................................................................... 12 Chapter 2 27 Figure 2.2: Afghanistan's Perceived Performance in Six Dimensions of Governance .............................. Figure 2. 1: Security Sector Policy Coordination........................................................................................ 32 Figure 3. 1: Pay Scales ($/month) in Security and other Sectors ................................................................ Chapter 3 35 Figure 4. 1: Security Spending as a Share of GDP...................................................................................... Chapter 4 Figure 4.2: Forecast Ordinary and Development Budget Spending Against Forecast Revenues ..............43 47 Figure 5. 1: Composition of Security Expenditures. 2003/04-2004/05 ....................................................... Chapter 5 50 .. 11 Figure 5.2: Economic Classification of Security Sector Spending. 2004/05 ............................................. 52 TABLES Chapter 2 Table 2. 1: Security Sector Parameters ....................................................................................................... 17 Table 2.2: Approved Staffing Establishment inthe Security Sector .......................................................... Table 2. 3: M a i n National Entities Involved in Security, their Roles, Gaps and Duplications ...................22 24 Table 2.4: Current Status o f Security Sector PRR Applications ................................................................ 25 Table 3. 1: Security Sector FundingArrangements ................................................................................... Chapter 3 39 Chapter 4 Table 4. 1: Summary o f Total Security Sector Expenditures 2003/04 -2005/06 (US$million)................42 Table 4.2: Security Sector Expenditures by Program (US $ million) ........................................................ 43 Table 4. 3: Likely Dnvers of Security Sector Expenditures ....................................................................... 46 Chapter 5 Table 5. 1: Functional Breakdown of Expenditures for the Security Sector (2002/03 to 2004/05) ...........51 Chapter 6 Table 6. 1:Deviation Analysis of Security Sector Budget.......................................................................... 66 ... 111 ACKNOWLEDGEMENTS As part o f the overall Public Finance Management Review (see Acknowledgements for Volume I), this volume was prepared by Peter Middlebrook (SASPR), Nicole Ball (Consultant, DFID), William Byrd (SASPR), and Christopher Ward (Consultant). Contributions to this work were also made by Sttphane Guimbert (SASPR) and, on restructuring and reform o f civilian management agencies, by Satendra Prasad (SASPR) and Jalpa Pate1(SASPR). Comments fkom Daud Yaqub and Mark Sedra are gratefully acknowledged. This volume was processedby Juliet Teodosio (SASPR). The participation and support o f DFID made an invaluable contribution to this work and i s gratehlly acknowledged. The process o f carrying out the PFMReview involved extensive interactions with the Ministry o f Finance and other Government agencies, whose valuable cooperation and support was indispensable for completing the review and also greatly benefitedthis report. iv EXECUTIVESUMMARY INTRODUCTION 1. Security, an essential public good, i s a central issue for Afghanistan's reconstruction, development, and public finances. It impacts on the full gamut of development issues faced by Afghanistan, ranging from state building and capacity development to revenue collection, service delivery, and encouraging private sector-led growth. For the purpose o f its National Development Strategy (ANDs,currently under preparation), the Government o f Afghanistan innovatively has added an MDG on enhancing security to the internationally agreed list o f MDGs. Afghanistan's external partners (including the World Bank) have consistently highlighted the importance o f security as a key enabling factor for Afghanistan's reconstruction and development. And most important, the Afghan people themselves emphasize the importance o f security intheir daily lives and for buildingtheir livelihoods and communities. In addition to its importance from a development perspective, the security sector accounts for a very highproportion o f the country's total budgetary expenditures (around 40%) as well as external assistance, and raises major public finance management issues. In view o f these considerations, the Government o f Afghanistan requested that the security sector be included in the World Bank's Public Finance Management (PFM) Review, and this volume i s the result. ii. Themainobjectivesofthisreviewofsecuritysectorexpenditures, intheoverallcontextofthe PFMReview andfrom a PFManddevelopment perspective, are as follows: To gain as complete an understanding as possible o f the current level and structure o f security expenditures, recent trends, and likely future expenditure requirements based on current plans. To assess the extent to which coherent strategies, for the security sector and the major parts o f the security sector, guide public expenditure allocations. To review processes for determining funding levels, expenditure allocations, budget execution, and post-execution functions, and assess the extent to which they follow sound PFMprinciples. To describe the institutional framework for the security sector and ascertain the extent to which it embeds or i s conducive to sound public financial management. To assess the strategic coherence and consistency and the fiscal sustainability o f security sector expenditures. CONCEPTUAL FRAMEWORK iii. Definition of thesecuritysector. Thisvolumeusesabroaddefinitionofthesecuritysectorbut focuses on those components which are of major importance from a PFM perspective. The budget categories and programs reviewed include: defense, police and other internal security services, intelligence, counter-narcotics, the judiciary and penal system, mine clearance, and the DDR program. The main security institutions in Afghanistan include: (i)security management and oversight bodies: National Security Council and National Security Advisor, Ministry o f Defense, Ministry o f Interior, Ministry o f Counter Narcotics, Ministry of Border and Tribal Affairs, Ministry o f Justice, Ministry o f Finance, and the Auditor General's Office; (ii) security services: Afghan National Army (ANA), Afghan National Police (ANP), National Border Guards, Counter-Narcotics forces, National Directorate o f Security (NDS - responsible for intelligence), Customs General Directorate under MoF, remnants o f earlier military and police forces (the former subjected to DDR); and (iii) justice and legal institutions: Supreme Court, Judiciary, Attorney General's Office and prosecutorial offices (saranwal). Excluded from detailed analysis in this volume are the foreign and international armed forces operating in Afghanistan, and also non-statutory security bodies including remaining unofficial militias and armed factional groups, privatebody-guards, and security companies. V iv. Security and development. As discussed in Chapter 1, the close linkages between security and development have become increasingly recognized in recent years. It i s unfortunately where security i s absent - most notably in countries inconflict -that the role o f security as a crucial enabling condition for development becomes glaringly obvious. The breakdown o f security and law and order that occurs duringprotracted conflict cames enormous human, social, physical, financial, andresource costs. Also o f particular importance are the damaging consequences for the state and the adverse and distortionary effects on private sector development. Another important aspect is the development costs o f criminality, which is fostered by conflict but can be a major problem even in the absence o f conflict, and in post- conflict situations where social norms and law enforcement have eroded. More specifically, crime can weaken social and human capital, worsen the business climate and discourage investment, and undermine the state and its ability to promote development. v. Application of PFMprinciples to the security sector. It has become widely accepted, including by internationalorganizations like the OECD, that sound principles o fpublic finance management should be applied to the security sector. This i s in the interest o f the security sector itself, and also very important for the PFM system as a whole interms o f achieving desirable PFM outcomes with respect to (i) budgetary aggregates and overall fiscal discipline; (ii) strategic allocation o f resources across sectors and programs; and (iii) efficiently managing the use of budgetary resources. These in turn contribute to positive development outcomes. Appropriate modifications in PFM practices can be made to take into account confidentiality considerations where necessary, but not in a blanket manner throughout the security sector or in a way that undermines the fundamental principles o f accountability to elected civil authorities. Key implications o f application o f PFMprinciples to the security sector (outlined in Chapter 1and discussedfurther inChapter 3) include the following: e First, the security budget and the budgets of its main components (defense, interior, justice, etc.) should beprepared against clear strategies at overall national security and sectoral levels. Specifically in the case of defense, the strategy needs to be based on a transparent "threat assessment" fully owned by the government. e Second, security sector policies and expenditures must be affordable, both in the short run (annual budget) and in terms of implications for spending in future years. This requires full incorporation of the security sector inthe annual budget formulation process, subject to aggregate fiscal constraints and sector ceilings like any other sector, and it also means that the security sector must be fully incorporated inmedium-term fiscal projectionsand planning. e Third, once an overall resource envelope is agreed, resources should be allocated in line with strategic priorities, both within the security sector and across all sectors (including security). This means that, in addition to the need for sound strategies as emphasized above, key PFM principles like comprehensiveness and transparency shouldbe appliedto the security sector. e Fourth, resources appropriatedfor the security sector need to be used efficiently and effectively, which implies full application o f PFM instruments and processes in the budget execution sphere (discussed in detail in Volume I,Chapter 7 and in Volume 11). Of particular importance are the control framework, good information flows, procurement, and monitoring and evaluation. vi. Service deliveryperspective. Since security is an essential public service, usinga service delivery framework (see Volume I, Chapter 8) makes good sense and provides guidance for improvingPFMinthe security sector. The distinctive features o f security as a public service - e.g. the inherent threaduse o f force, difficulties in measuring and monitoring performance, etc. - need to be taken into account. Some key implications o f a service delivery approach to the security sector (focused on institutions, incentives, and accountability relationships) include the following: e Effective oversight by legitimate civil authorities i s essential; for national security forces this is the primary mode of accountability, and for both national and local security forces such accountability is essential to prevent abuse o fpower by security forces. vi 0 Where local public security i s involved, there i s scope to enhance the accountability o f service providers (i.e. police) directly to the local population that they serve. 0 Inthiscontextitmakessensetoconsider some formofde-concentrationofgovernment oversight even in an overall unitary state structure, as long as the state i s finctional at the local level and not "captured" by regional and local powerbrokers, who would misuse access to security forces and associated power. 0 Donors need to reinforce and not inadvertently distort or undermine the domestic accountability relationshipsinthe security sector. AFGHANISTAN'S SECURITY SECTORAND RECENT PROGRESS vii. Nearly a quarter-century o f protracted conflict in Afghanistan, starting in the late 1970s, led to pervasive insecurity throughout the country and destroyed or profoundly damaged the institutionsthat had been responsible for providing security in the pre-war period -both formal governmental institutions and traditional mechanisms (Chapter 2). Government security forces were misused in ways that reduced rather than enhanced security, e.g. during the Soviet occupation o f the 1980s, and they became increasingly "captured" by factions in Afghanistan's civil war during the 1990s. The emergence o f numerous well-armed militia groups, in shifting loose alliances, meant that the ability to use large-scale violence in pursuit o f objectives was no longer the monopoly o f the government but rather was widely contested, greatly undermining security. The judicial system was similarly undermined, and traditional mechanisms for providing justice often captured and used for their own ends by commanders and warlords. Women suffered greatly from the conflict as a result o f violence against them, and also due to limited mobility and lack o f access to education and health services, culminating in the Taliban's severe social restrictions. Neighboring countries and other powers repeatedly and competitively intervened militarily - either directly or indirectly by providing arms and other support to different factions - contributing in a major way to insecurity in Afghanistan. In sum, the long period o f conflict and associated insecurity had profoundly damaging economic, social, and political consequences for the country and its people. viii. The critical importance o f enhancing security and reforming the security sector in Afghanistan has been widely recognized. While this was signaled in general terms in the Bonn Agreement, which provided for the establishment o f an International Security Assistance Force ( I S M ) , two subsequent donor meetings on security set targets and assigned donor responsibilities. Considerable progress has been achieved in strengthening and in some respects reforming the security sector, particularly inthe last two years. Major successes include the avoidance of large-scale fighting since late 2001, the steady shift away from a pattern o f endemic conflict toward a situation where the people and society become progressively more accustomed to peace, and the gradual extension o f the writ o fthe national government beyond Kabul. However, great gaps remainboth interms o f areas that the Bonn Agreement didnot cover and in achieving agreed security-related targets. Moreover, although very substantial, international security forces' presence inAfghanistan has been limited as compared to other post-conflict situations. ix. Salient features o fAfghanistan's security sector at present include the following: The historical legacy of conflict wasfragmentation into regional and local militias, capture of policing and military functions by non-legitimate actors, and erosion of both formal and traditional justice systems, also with capture by non-legitimate authorities. Building the state since late 2001 hence has entailed reconstituting and reforming the security forces under legitimateoversight by the civil authorities. Immediate security threats faced by Afghanistan are variegated and disputed by different stakeholders. Some observers see the continuing insurgency in the south and east and conflict with TalibadAl Qaeda forces as the most important security problem. Others are most concerned vii about criminality and lack o f rule o f law at the local level. Still others are concerned about the limited management capacity o f civil authorities and inadequate oversight`accountability o f security forces. The drug industryis widely seen as a major cause and consequence o f insecurity. The development o f an agreed national security strategy, currently under preparation, will facilitate reaching consensus on the different security threats and their relative priority. The overall size of the security sector, even with recent andplanned expansion, does not seem unreasonably large by international standards. These include a planned size o f the ANA o f 70,000 (not all o f them combat troops) and a planned size o f the Afghan National Police (ANP) of 62,000. However, the appropriateness o f the size and balance between different security forces can be questioned inthe context o f Afghanistan's particular situation and security needs. Afghanistan's security sector neverthelessis costly and may be unaffordable. Over the medium term, it is extremely doubtful whether the sustaining costs o f Afghanistan's security sector at planned force levels can be absorbed by Afghanistan's national budget. This implies that difficult trade-offs will need to be made in prioritizing expenditures, means o f achieving efficiencies in security-related expenditure will have to be identified, and the possibility o f multi-year donor financing commitments for the security sector will needto be explored. Progress in developingdifferentparts of the security sector has been uneven. After a slow start initially, the formation and expansion o f the ANA have moved forwqrd rapidly. Progress in forming the new ANP has been much slower, and reform and capacity building in the justice sector have lagged far behind, with adverse implications for legal reforms and a pro-private sector business climate. Security andfiscal issues are closely interlinked. For example, low domestic revenue is both a consequence and a cause o f insecurity. Reconstruction costs tend to be higher due to insecurity, while on the other hand the security sector i s a source o f considerable fiscal pressure. Pay and grading decisions with respect to the security services, made outside national budget processes, can have important fiscal implications. KEYCHALLENGES x. From a PFMperspective, key challenges facing the security sector include: The needfor an integrated security sector strategy andpolicy framework, without which sound programmatic - e.g. on force sizes, equipment - and public expenditure decisions cannot be made. Preparation o f the National Security Strategy is at an advanced stage. Very high donor-executed spending through the external budget, which renders achieving coherence o f spending across and within subsectors much more difficult, and also raises questions about the fiscal sustainability o f the security sector in the medium term when donor funding may decline. DifJiculties in coordinating andprioritizing security sector expenditures and actions, reflecting the above problems and fragmentation o f decisionmaking across donors. These factors make it more difficult for the National Security Council (NSC) to fully carry out its mandated strategic, leadership, and coordination functions, although the situation inthis regard has improved in2004 and 2005. Concerns about the growing security sector wage bill and pressuresfor fiscally unaffordable salary increases, as well as thefiscal sustainability of security sector staffing levels. The ANA salary structure, determined apparently without reference to fiscal constraints or pay elsewhere in the civil service, has set a precedent which the police and other sectors aspire to and which will be fiscally costly. There are also doubts about whether recruitment practices are being regulated by staffingestablishment size limits. Slow implementation of priority public administration reforms and capacity building in security sector management and oversight institutions, which have lagged behind the vlll ... Medium-Term Fiscal Framework (MTFF) into which it would feed. The sustainability of planned expenditures needs to be assessed within the context o f the overall rolling fiscal plan, growth o f domestic revenues, and possible medium-term commitments from donors to fund recurrent costs. Revisit strategic objectives and trade-offs in the light of fiscal constraints. This would ideally be done as part o f the exercise to finalize the national security strategy, which itself should be consistent with and supportive to the national development strategy (ANDs). Take actions to improve the situation. The Government may consider (1) avoiding any further wage increases inthe security sector; (ii) reconsidering programmedforce sizes, especially for the ANA, in the light o f fiscal constraints; (iii)exploring possibilities for containing non-salary recurrent costs in a manner that does not adversely affect readiness and operational capabilities; and (iv) similarly reviewing construction plans and equipment costs. Explore with donorsprospects for predictable medium-term externalfunding for the security sector, in line with agreed security sector objectives and a clear road-map for phasing out such fundingover anumber o fyears as Afghanistanprogresses toward a sustainable fiscal position. Ensure that the fiscal sustainability "lens" is applied to future decisions, with a view to preventing decisions or actions from worsening o f the fiscal situation o f the security sector. STRATEGICPRIORITIZATIONAND SEQUENCING xiii. As can be seen from Figure 2 below (discussed in Chapter 5), security sector expenditures have been dominated by Defense (primarily ANA) and Interior (mainly police), which together account for close to 90% o f the total. Counter-narcotics expenditures are variable and inpart may be double-counted with other spending inside and outside the security sector, but they also have become quite large in the last two years. Figure 2: Composition of Security Expenditures, 2003/04-2004/05 National Police and Law Enforcement 28% -Justice 3% Afghan National ,MineActio Army - 6% 60% -DDR 3% I Source: MoF budget documentsandjiscal reports, staff estimates. xiv. The low spending on the justice sector is striking and i s reflective o f the lack o f progress on improvements and reforms in the justice system. The national police also until recently appears to have been underhnded, although this i s being corrected. Thus the overall allocation of security sector expenditures seems somewhat disconnected from the concerns voiced by large numbers o f Afghans about security and crime. X xv. As in the case of the national budget as a whole, there are no easy answers or technical fixes for allocating public spending, either between security and other sectors or within the security sector. Clearly, the very large share o f security spending in the External Budget (nearly 80% in 2004/05), and fragmented donor decisionmaking on expenditures, make the task o f the Office o f the National Security Council (ONSC), the Ministry o f Finance (MoF), and other oversight bodies especially difficult in this regard. From a more positive perspective, some points with respect to intersectoral and intrasectoral prioritization are outlined below (see Volume I, Chapter 5 for further discussion o f prioritization). xvi. Prioritization across sectors: Sophisticated technical approaches may require a great deal o f resources and even so are highly imperfect and can result in misleading conclusions. Simpler rules o f thumb may work better, andjudgment is called for. Intersectoralallocationdecisions need to be anchored inthe overall national security strategy. Identifying and mitigating gross anomalies in intersectoral allocations is very important (e.g. the serious underfunding o f thejustice sector, discussed earlier). International patterns o f spending and outcomes (for example for police) may provide some guidance on where funds should be allocated. Identifying bottlenecks and directing expenditures to alleviate them may be a useful short-run approach. Some linkage between sectoral allocations and performance may make sense but needs to be watched carefully and corrective actions taken to ensure that large imbalances between sectors do not emerge over time. Finally, it i s essential that intersectoral prioritization be comprehensive - no sector (e.g. Defense) shouldbe considered sacrosanct or immune from scrutiny. xvii. Prioritization within sectors: Technically, prioritization within sectors is more manageable than intersectoral prioritization because the alternatives tend to be more comparable. Intrasectoral allocation decisions need to be anchored ina sound sector strategy. Technical analysis (e.g. cost-effectiveness analysis) can be used to evaluate individual project proposals and also to make more generic prioritization decisions (socio-economic cost-benefit analysis o f landmine clearance i s a good example). Ruthless culling o f projects with lower benefits andor which do not have good prospects for rapid completion will the improve timeliness, efficiency, and effectiveness o f sector programs. Complementarities need to be factored into prioritization decisions (e.g. there i s no point in having a police force that i s not adequately armed, clothed, equipped, and with vehicles and fuel). Correcting geographical and other disparities (for example in terms o f police coverage) may help guide prioritization. BUDGET EXECUTION EFFECTIVE EFFICIENTUTILIZATIONOF FUNDS FOR AND xviii. The effective and efficient utilization o f funds to achieve stipulated purposes i s a key element of the PFM system, including in the security sector (Chapter 6). A widely accepted general principle i s that all the elements o f a sound budget execution process should be applied to the security sector, inthe same way as to the rest o f the budget. Security expenditures within the Core Budget are essentially subject to similar control mechanisms, accounting and reporting requirements, and procurement arrangements as in the rest of the Core Budget. The main difference i s the absence o f a strong internal audit function - a role that in the case o f the civilian budget i s to a large extent played by the Monitoring Agent o f the ARTF (see Box 6.2, Volume I). absence o f such a mechanism - albeit extraordinary (and temporary) - in The the security sector means a weaker control framework despite the application o f other elements of the budget execution process for the Core Budget. xi xix. Security spending inthe External Budget occurs completely outside these Government processes. Although this spending i s subject to financial controls exercised by donors, the associated disadvantages are potentially very important with respect to the quality and consistency o f information flows, and procurement. Another major problem i s the wide variety o f salary levels, payment channels, and top-up arrangements in the security sector, which risk fragmenting the accountability and loyalties o f security sector staff. xx. With respect to procurement, it is possible to follow sound procurement procedures despite the actual or perceived obstacles in defense or other security sectors, and good procurement practices will reduce vulnerability to corruption. Inparticular, competitive procurement with adequate transparency and oversight i s critical. With the recent passage o f a new Procurement Law that meets international standards, donors funding security through the External Budget should move toward using sound internationally acceptable procurement procedures, either those provided for in Afghanistan's Procurement L a w or similar procedures. This is also important in the case o f procurement o f supplies, security services, etc. by international militaryforces inAfghanistan. xxi. Developing and monitoring performance indicators tends to be even more difficult in the security sector than inother sectors, partly because the outcomes are not easy to measure precisely, and also due to the difficulties faced in identifying indicators that provide the right incentives. Options will vary across sectors and programs. 0 In the case of defense, indicators of preparedness (although they are more ahn to output indicatorsrather than to outcomes) are the most appropriate performance indicator inpeacetime. 0 For thepolice, and for the internal security system more generally, the ultimate outcome o f local security and rule o f law is difficult to measure. Official crime statistics are hard to use in monitoring performance over time and especially across countries. Information from household surveys on the numbersand types o f crime encountered by households can provide a useful check on the validity o f official statistics. The effectiveness of the police and prosecutorial services can be measured by the percentage o f different categories o f cases resulting in convictions, based on which type o f evidence (confessions, physical evidence, etc.). The availability o f female police officers and separate detention facilities for women are indicative o f mitigating the risks women face duringdetention. 0 The performance o f the justice system can be measured by various indicators like numbers o f cases handled, timeliness o f processing court cases, and households' perceptions o f the justice system and their indications o f willingness (or otherwise) to use formal justice channels when seeking redress. The existence o fFamily Courts and Juvenile Courts and their functioning is also indicative o f how sensitively and even-handedly the justice system deals with women and children. 0 In the case of mine action, outcomes of mine clearance activities (land and areas free of landmines - with clear linkage to ultimate development benefits) can be specified and measured reasonably well, and compared with the costs o fmine clearance. 0 For counter-narcotics, the appropriate outcome indicator at the national level i s progressive reduction over time and eventual elimination o f the opium economy. However, measuring the performance o f counter-narcotics activities i s difficult, and linking such performance to progress toward achieving the strategic objective o f reducing opium production and trade i s even more difficult. Given the variety o f activities encompassed under the counter-narcotics rubric, performance indicators for the different components will need to be appropriate to their varying characteristics. 0 More generally, inall parts o f the security sector some basic indicators of inputs can be useful in measuring performance, for example number and ratio o f staff on the job (confirmed by spot- inspections), percentage o f staff in security services properly equipped, proportion of staff who xii have received training, percentage o f budget resources spent (both Core and External Budget to the extent possible), etc. Expenditure tracking surveys can be a useful tool inthis regard. 0 Unit costs, although not an outcome indicator, nevertheless can be useful performance indicators of the efficiency with which inputs for service delivery are beingprovided. xxii. All this leads to a rather formidable agenda for generating and effectively utilizing monitoring information. This can start with small steps and make progress incrementally over time. One promising option i s to generate and regularly monitor basic financial information on security sector expenditures in the Core Budget, and encourage donor partners to provide similar information on as consistent a basis as possible for External Budget expenditures inthe security sector. Staffing data, which have suffered from inconsistencies, need to be improved and monitored. These can be augmented as proposed above with indicators o f numbers or proportions o f staff on duty and available to carry out duties. Subsequently, simple indicators o f output/performance can be compiled and used inmonitoring. It is more important to collect and effectively utilize simple information than to devise highly sophisticated data and monitoring systems which are difficult to implement and become a burden on scarce Government capacity. Finally, the institutions concerned need to take on the role o f monitoring and evaluation and treat it as a serious part of their work, and make meaningful use o f performance indicators in decision making. As indicated above, an important gap inthis regard i s the internal audit function. SUMMARY OFRECOMMENDATIONS xxiii. Chapter 7 concludes that the most important message emergmg from this analysis o f security sector expenditures i s that the security sector needs to be appropriately integrated into all aspects of Afghanistan's PFM system: medium-term fiscal programming (MTFF), annual budget formulation, staffing and payroll management, financial management and controls, and external scrutiny, among others. There i s no justification for treating the security sector as separate or sacrosanct, and not subjecting it to budgetary and fiduciary processes. Specific recommendations include the following. xxiv. Review the Fiscal Implications of Ongoing Security Sector Reforms: Quickly enhance the information base on security sector expenditures, and conduct an integrated security sector-wide fiscal review to forecast fiscal implications o f security sector development and reforms over the medium- and longer-term. This would serve as a basis for revisiting security sector force levels, notably for the ANA, and for discussions with donors on longer-term financing. It would also facilitate policy- and program- basedbudgetingo f security sector expenditures with a medium-term perspective. xxv. Further Develop Policy Making Capacity in the NSC and Finalize the National Security Strategy: The National Security Advisor and the NSC are tasked with coordinating policy development and overseeing integrated strategic planning across all security institutions. Adequate capacity i s required for these purposes. An integrated security sector strategy and policy i s needed to determine acceptable longer-term costs o f security institutions and appropriately prioritize and sequence spending across and within subsectors (e.g. justice and police versus ANA). As in the case o f the ANDs as a whole (discussed below), the National Security Strategy i s being developed by Afghan stakeholders, with full Government ownership (across the ministries and agencies involved), and will need to gain wide buy-in within andoutside Government. This will require a meaningful and substantive consultationprocess. xxvi. Treat Security as an Integral Part of the National Development Strategy Currently under Preparation: The ANDs can be used as an opportunity to ensure that the security sector both i s accorded its due importance as a development issue for Afghanistan and that the broader developmental and fiscal perspective is injected into the security sector strategy. xiii xxvii. Implement Priority Administrative Reforms and Develop Security Sector Management and Oversight Capacity: Support reforms enhancing human resource capabilities in the Ministries o f Defense, Interior, Justice, Counter Narcotics, and in the National Security Directorate. A particular priority is developing solid financial management capacity and practices in these line ministries. Strengthening the control framework, notably with the development o f an internal audit function, i s a priority. Following the example o f the civilian budget, an option would be to buy the services o f a reputed firm. But, as this might prove to be sensitive, an alternative option is to focus the newly developing Internal Audit Department o f the Ministry o f Finance initially on the security sector. The vocation o f this department, as per the new Public Finance and ExpenditureManagement Law, is to cover the whole o f Government. Given that a mechanism (the ARTF Monitoring Agent) i s temporarily inplace for most o f the civilian recurrent budget, this department could initially prioritize its work program toward the security sector. xxviii. Rigorously Enforce Staffing Size, and Set Salary Increases Based on Fiscal Absorption Capacity: It i s vital that'the agreed formal staffing establishment sizes be enforced. Moreover salaries, like non salary costs, must be subject to the normal rigors o f public expenditure management and fiscal discipline -the security sector wage bill mustremain affordable. xxix. Role ofdonors. Donorsplay an even greater role inAfghanistan's security sector than they do in other sectors. Consequently they also play a very important role in public finance management in the security sector. Whereas donor practices inadvertently can hinder the Government's efforts to make the national budget the central instrument o f policy, reform, and resource allocation for the security sector, donors can also make a major contribution to improvements in this regard. Specifically, donors can contribute to improving public finance management inthe security sector by: (i) supporting development o f an agreed national security strategy and corresponding strategies and policies for the individual sectors (defense, police, justice, etc.); (ii) reassessing their views about force size and cost requirements in the light o f fiscal constraints; (iii) helping enhance the capacity o f oversight actors (such as the Auditor General, legislature) and economic managers (Ministry o f Finance Budget Office) to address issues relating to financial management in the security sector; (iv) channeling more o f their assistance to the security sector through the Core Budget; (v) for resources not channeled through the Core Budget, ensuring that timely and accurate information i s made available to the budget authorities, and maximizing use o f internationally acceptable procedures e.g. for procurement; (v) helping key security ministries strengthen their budget formulation and budget execution processes; and (vi) discussing medium- and longer-term availability of external resources for the security sector. xxx. Implementing these recommendations and other initiatives to improve security in Afghanistan, which is a necessary condition for the country's development, undoubtedly will take time. However, the needto get a handle on the fiscal sustainability issues inthe security sector i s urgent, as there i s a risk that decisions on force levels, pay, non-salary spending, equipment, construction, etc. that are being made or implemented today will be unaffordable for Afghanistan in the future. Dialogue with donors, including on possible medium-term financial support to Afghanistan's security sector, also is urgently needed. xiv CHAPTER 1. SECURITY, DEVELOPMENT,AND PFM PERSPECTIVE A. Introduction BACKGROUND RATIONALE AND 1.1 Security, an essential public good, i s widely seen as a central issue for Afghanistan's reconstruction and development.' Security impacts on the whole range of development issues faced by Afghanistan, ranging from state building and capacity development to revenue collection, service delivery, and encouraging private sector-led growth. For the purpose o f Afghanistan's National Development Strategy (ANDs), currently under preparation, the Government has added a Millennium Development Goal (MDG) on enhancing security to the internationally agreed list o f M D G s (see Afghanistan Government, 2005b). This demonstrates the importance assigned by the Government to security and rule of law as a critical development issue. Afghanistan's external partners (including the World Bank) have consistently highlighted the importance o f security as a key enabling factor for Afghanistan's reconstruction and development And most important, the Afghan people themselves emphasize the importance o f security and the protection o f property rights in their daily lives and for buildingtheir livelihoods and communities. More specifically: e Security and rule o f law as public goods have been grossly underprovidedin Afghanistan during the past quarter-century, with state functions inthis regard usurpedby factions, regional and local powerbrokers, and external actors. Remedying this situationi s an urgent, high-priority task. e Improving security i s also a core component o f the state-building agenda, seen by most Afghans as an essential function o f an effective, accountable government. e Moreover, an adequate level o f security is a critical prerequisite for effective delivery o f other public services. e Politically, security improvements have been and will continue to be essential for the success o f Afghanistan's transition toward political normalcy. e Economically, security and law and order, along with the protection o f property rights, are required for robust private sector development (including emergence of a productive, competitive formal sector - see World Bank, 2005), which will be the only way to achieve sustained rapid economic growth over the mediumterm. e Insecurity and narcotics are closely linked- insecurity i s the environment inwhich Afghanistan's opium economy thnves, and the drug industryi s a major source of instability and insecurity. e And finally, from the individual, household, and community standpoints, security andrule o f law are extremely important components and enabling factors in the quality o f life, as is almost universally recognized by Afghans. 1.2 Inaddition to its importance as a central development issue for Afghanistan, the security sector accounts for a very high proportion o f the country's total public expenditures as well as external assistance and raises major public expenditure management issues including risks to fiscal sustainability and possible crowding out o f spending on other basic services. Therefore the security sector is also very important from a public finance management perspective. In view of these considerations, the Government o f Afghanistan requested that the security sector be included in the World Bank's Public Finance Management (PFM) Review, and this volume i s the result. ~ This is emphasized in past Government strategic documents like the National Development Framework and Securing Afghanistan s Future. Security also forms a core pillar o f the Afghanistan National Development Strategy currently under preparation by the Government. 1.3 There are in addition a number o f specific rationales for the World Bank's engagement in analytical work on Afghanistan's security sector as part o f the PFM Review and from a development perspective. First, as indicated above, security i s of central importance for Afghanistan's development prospects, including in particular the state-building agenda. This means it has to be incorporated in the country's development strategy and planning (which the Government i s doing), and that development partners needto be engaged inareas o f their competence. 1.4 Second, given the great importance o f security spending in the national budget (accounting for around 40% o f total Core and External Budget expenditures), the security sector must be incorporated in the analysis o f expenditure composition, budget prioritization and trade-offs, downstream expenditure implications o f decisions and investments, fiscal sustainability, etc. These issues lie at the heart o f the PFMReview. 1.5 Third, a weak PFM system inthe security sector (and particularly if the security sector gets left behind while improvements in the PFM system are occurring elsewhere) would adversely affect the overall progress o f PFMreform and institutional strengthening, distort choices made by the Government, reduce the effectiveness and efficiency o f public spending, and increase vulnerability to corruption. All of these areas are o f concern to the Bank. 1.6 And fourth, security sector expenditures are largely donor-financed, raising important issues o f aid management, aid harmonization, and aid effectiveness, as well as capacity development and effectiveness o f technical assistance. These issues comprise an important part o f the overall PFM agenda and hence need inthe context o fthe security sector to be included inthe PFMReview. OBJECTIVESAND CHAPTER CONTENT 1.7 The main objectives o f this review o f security sector expenditures in the context o f the PFM Review and talunga development perspective are as follows: To gain as complete an understanding as possible o f the current level and structure o f security related expenditures, recent trends, and likely future expenditure requirements based on current plans. 0 To assess the extent to which coherent strategies, for the security sector and the major parts o f the security sector, guide public expenditure allocations. 0 To review the processes for determining funding levels, expenditure allocations, budget execution, and post-execution functions, with a view to assessing the extent to which they follow sound PFMprinciples. 0 To describe the institutional framework for the security sector and ascertain the extent to which it embeds or i s conducive to sound public financial management. 0 To assess the strategic coherence and consistency and the fiscal sustainability o f security sector expenditures. 1.8 This chapter sets forth the conceptual framework for analysis o f security sector expenditures from a development and public finance management perspective. It first provides a working definition o f the security sector. The chapter goes on to outline key linkages between security and development (based on recent international thinkingon this subject). It then discusses key public finance management principles and their application to the security sector. The final section briefly summarizes the service delivery perspective (developed inVolume I, Chapter 8) and its application to the security sector. 1.9 Chapter 2 first provides some background on the historicallegacy from the longperiod of conflict in Afghanistan, and summarizes the progress made inbuilding the security sector since 2001. It then 2 overviews Afghanistan's security sector at present - its legal basis, size, institutions, salient characteristics, and coordination - and at the end outlines the key security issues faced by Afghanistan from a development perspective. Chapter 3 discusses key PFM principles and their application to the security sector, the implications for Afghanistan including gaps and constraints, and the Afghan budget process for the security sector. Chapter 4 discusses the crucial issue o fthe fiscal sustainability o f security sector decisions and investments, which i s an important dnver o f the fiscal sustainability o f Afghanistan's public finances as a whole (see Volume I,Chapter 3). Building on the analysis o f fiscal sustainability, Chapter 5 looks at strategic prioritization and sequencing issues in the security sector and how these can be managed inparticular through the nationalbudget process. Chapter 6 focuses on budget execution for effective and efficient use o f funds provided to the security sector. Chapter 7 concludes this volume by outlining the key challenges and priorities for action from a PFMperspective. DEFINITIONOFTHE SECURITYSECTOR 1.10 The security "system" i s defined broadly in OECD (2005b) as including the following main components: core security actors, security management and oversight bodies, justice and law enforcement institutions, andnon-statutory security bodies (see Box 1.1 for details). Box 1.1: Definition of the SecuritySector The OECD's Development Advisory Committee (DAC) defines security as a development objective as follows: "Security is increasingly viewed as an all-encompassing condition in which people and communities live in freedom, peace and safety, participate fully inthe governance o f their counties, enjoy the protection o f fundamental rights, have access to resources and the basic necessities o f life, and inhabit an environment which is not detrimental to their health and well-being. The security o f people and the security o f states are mutually reinforcing. A wide range o f state institutions and other entities may be responsible for ensuring some aspect o f security." This definition is consistent with broad concept o f"human security" usedby the UN. The security system includes the following actors: Core security actors: armed forces, police, gendarmeries, paramilitary forces, presidential guards, intelligence and security services (both military and civilian), coast guards, border guards, customs authorities, reserve or local security units (civil defense forces, national guards, statutory militias). Security management and oversight bodies: the Executive, national security advisory bodies, legislature and legislative select committees, ministry o f defense, ministry of internal affairs, ministry o f foreign affairs, finance ministry (as well as budget offices, financial audit and planning units), and civil society organizations (civilian review boards and public complaints commissions). Justice and law enforcement institutions: judiciary, justice ministry, prisons, criminal investigation and prosecution services, human rights commissions and ombudsmen, and customary and traditional justice systems. Non-statutory security bodies: liberation armies, guerrilla armies, private body-guard units, private security companies, political partymilitias. These actors, their roles, responsibilities, and actions, crucially affect security outcomes as defined above. Source: Adaptedfrom OECD(2005b, pp. 20-21). 1.11 A broad definition o f the security sector along these lines makes the most sense from a development and P F M perspective, as all different parts o f the security sector have significant development implications as well as varying budgetary implications (for both expenditures and revenues)? However, some components clearly do not apply to Afghanistan (e.g. coast guards), while other institutions like Parliaments have not been functioning and hence are not relevant for analysis o f recent developments and the current situation (although the role o f the legislature will be important startinginthe very near future). Moreover, non-statutory security bodies, although they have been major actors in Afghanistan (and mirror the breakdown o f the formal security system during the protracted conflict), for the most part do not have direct budgetary implications and would require a very different mode o f analysis. Therefore they are not looked at closely in this volume except in terms o f their * World Bank(2003a) takes a similarlybroad approachinlooking at conflict anddevelopment policy. immediate budgetary implications- notably the Disarmament, Demobilization, and Reintegration (DDR) program - although their development implications and impact on the formal security sector will be kept inmind. 1.12 Insum, consistent with the overall objectives and approach ofthe PFMReview and inlinewith the increasingly recognized development implications o f the security sector, this volume uses a broad definition o f the security sector but focuses on those components w h c h are o f major importance from a PFMperspective. Interms o fbudget categories and programs, those that form part of the security sector and which are analyzed inthe PFM Review include: defense, police and other internal security services, intelligence, counter-narcotics, thejudiciary and penal system, mine clearance, and the DDRprogram. 1.13 The main security institutions inAfghanistan which will be looked at inthe PFMReview include the following: (i) government security management and oversight bodies: National Security Council and National Security Advisor, Ministry o f Defense, Ministry o f Interior, Ministry of Counter Narcotics, MinistryofBorder andTribal Affairs, MinistryofJustice, MinistryofFinance, and the Auditor General's Office; (ii)security services: Afghan National h y (ANA), Afghan National Police (ANP),National Border Guards, Counter-Narcotics forces, National Directorate o f Security (NDS - responsible for intelligence), Customs General Directorate under MoF, remnants o f earlier military and police forces (the former subjected to DDR); and (iii) justice and legal institutions: Supreme Court, Judiciary, Attorney General's Office and prosecutorial offices (saranwal). 1.14 Excluded from detailed analysis in this volume are the non-statutory security bodies, including remaining unofficial militias and armed factional groups, private body-guards, and security companies. Foreign and internationalarmed forces operating inAfghanistan also will not be analyzed indetail, except to some extent the Provincial Reconstruction Team (PRTs) which are intended to help strengthen governance and carry out reconstruction activities. However, the overall size and cost o f these forces will be considered and their impact on the fiscal and PFM aspects o f the formal security system will be taken into account. The absence o f a Parliament means that its oversight role with respect to security can be incorporatedin terms of future options and recommendations, but not inrelation to the analysis o f recent developments and the current situation. Independent and civil society institutions which play a monitoring and advocacy role with respect to security, such as the Afghanistan Independent Human RightsCommission(AIHRC) are extremely important butare not analyzedindetail inthis volume. B. Security andDevelopment 1.15 Costs of insecurity and conflict. The close linkages between security, an essential public good, and development have become increasingly recognized and closely studied in recent years (see, for example, UK DFID, 1999 and 2005), although knowledge gaps remain. The crucial enabling role o f security in virtually all dimensions o f development tends to be invisible in countries where there is a reasonable level o f security. As noted more broadly inChalmers (2005, p. 7), "The relationshipsbetween socio-economic governance, conflict, and growth are multi-directional, with poor governance and conflict reinforcing each other, andbothphenomena undermining the possibilities for economic development." 1.16 Inparticular, itiswhere security isabsent-mostnotably incountries inconflict-that itsrolein development becomes glaringly obvious. Insecurity, and inparticular conflict, carry major costs from an economic and social development perspective as outlined below (see World Bank, 2003a, Chapter 1): Cost of resources devoted to conflict anrVor combating insecurity, which otherwise could have been used for other, more developmentally productive purposes. Moreover, these resources impose a double loss, since they inflict damage during conflict as enumeratedbelow. 4 e Human costs, including deaths, wounds, displacement, and flight, as well as loss o f private property and psychological damage. Inaddition, disruption o f public services like education and health due to conflict or insecurity results in lower levels o f human capital (due to disruption o f education) and higher rates o f mortality and morbidity (aggravated by the direct adverse impact of conflict-related phenomena like displacement on health). Moreover, it should be emphasized that modem warfare, and especially internal conflicts like civil wars and insurgencies, tend to generate highcasualties and other costs among the civilian population. e Social costs, including weakening o f national, community, and family links, shorter time horizons, and weaker constraints on criminal and opportunistic behavior. e Loss of physical assets, directly by destruction during conflict, and indirectly due to lack o f maintenance, falling into disuse, etc. e Loss offinancial assets, through capital flight, can be a major cost of conflict or generalized insecurity. e Costs to government, including government revenues lostlforegone due to conflict, and destruction or erosion o f the state and state assets as it becomes weakened and contested by different factions duringconflict. Costs to theprivate sector, including disruption o f trade and markets, theft of business property by armed groups, "protection costs" frequently imposed, shifting incentives away from formal- sector activity infavor o fthe informal sector (see World Bank, 2005, Chapter l), andthe like. 1.17 Unfortunately, the adverse implications of conflictlinsecurity for development continue, and in some respects may increase, after the end o f outright conflict. So-called "legacy" costs o f conflict, based on cross-country analysis reported in World Bank (2003a, pp. 19-31), include in particular (i) higher residual military expenditure; (ii) flight (which tends to continue or even increase after conflict); capital (iii) ofsocialcapital-whichcantakealongtimetorebuildafterconflict;(iv)governmentpolicies, loss which on average are negatively affected after conflict; (v) deterioration o f political institutions which takes time to reverse; and (vi) destruction o f the revenue apparatus and loss o f revenue culture, which tends to persist after conflict. 1.18 Insecurity in a conflict or post-conflict situation also can be part o f a set o f mutually reinforcing tendencies that promote perpetuation o f insecurity and associated development costs, including in particular state weakness. For example in the case o f Afghanistan, World Bank (2005) emphasizes that lack o f rule o f law and continuing existence o f non-state powers (warlords and commanders and their forces) can lead to a vicious circle whereby insecurity, l o w revenues, and a weak, ineffective state are perpetuated (see Figure 1.1). 1.19 Another important aspect i s the costs to national development o f ~riminality,~which as noted above i s fostered by conflict but can be a major problem even in the absence o f conflict and in a post- conflict situation where social norms and law enforcement have been eroded. When pervasive, crime may also fuel wider conflict. Although the direct damage caused by crime i s far less than that o f major conflict, it can have major adverse effects on development, especially if a high level o f criminality becomes the norm. More specifically: e Crime erodes social and human capital: Loss o f trust in communities and countries due to widespread criminality can weaken the social fabric. Human capital can be weakened by the direct effects o f crime on victims, and by the associated atmosphere o f fear that may encourage others (particularly better-off and skilled people) to leave. By limiting movements o f people, crime may impede access to employment and educational opportunities (particularly in the case o f females). This paragraph draws from UNODC (2005, Executive Summary) which discusses these issues in the context of crime and development inAfrica. Also see Frank (2005) for a discussion of the crime issue ina post-conflict context. 5 0 Crime worsens the business climate and discouragesinvestment: Criminal activities are often directed against businesses and hence directly inflict losses on them, but more damaging i s the harm crime and perceptions o f it do to business investments. Foreign investors may be discouraged, domestic investors may not start activities, and more generally financial capital may flee from high-crime areas and countries. Corruption (see below) also can adversely affect the business climate ina major way. Crime undermines the state and its ability topromote development: Finally, crime can become closely linked to corruption (itself a form o f criminal activity), with very harmful consequences for development and public finance management (see Volume I,Chapter 7). Trust between people and the state i s destroyed andor i s prevented from being built. Corruption and other forms o f crime can erode the revenue base o fthe state and thereby its effective functioning. Figure 1.1: Security,RuleofLaw, and the InformalEquilibrium I Source: WorldBank (2005, p.10). 1.20 Criminality, and corruption inparticular, can drive business activity into the informal sector, and keep most businesses small (so as to avoid visibility vis-&-vis criminals andor corrupt officials). This has adverse consequences for investment, technology, productivity growth, and more generally the competitiveness o f the economy, reducing prospects for sustained rapid economic growth which i s the only way countries can develop and reduce poverty. 1.21 The adverse effects o f conflict and insecurity, and often o f crime as well, tend to be disproportionately felt by the poor, minorities, and other vulnerable groups. This i s partly because as in the case of many other public goods, the better-off can use their own resources to insulate themselves at least to some extent from insecurity (private security, physical protection through fences etc., mobility), while minorities and vulnerable groups are at higher risk of being targeted in situations o f conflict or breakdown o f law and order when the state no longer is able to protect its citizens. However, where insecurity i s generalized and long-lasting, as was the case in Afghanistan in the 1980s and 199Os, the effects reach all levels of society, and many assets o f the rich may get lost or devalued. Nevertheless, the poor are invariably more vulnerable to conflict and insecurity as they have fewer assets and weaker coping mechanisms to start with. Even if they lose many of their assets, better-off people have a greater chance to escape conflict and tend to be ina better position interms o f livelihoods where they flee to. 6 1.22 Conflict and insecurity can have an especially damaging effect from a gender perspective, notably in the form o f escalated violence against women. Rape and sexual assault against women of opposing groups in community-based conflicts are frequently used as an ultimate means o f dishonoring entire communities and reducing people's capacities to resist military advances, causing serious social and psychological disruptions. Any children resulting from such encounters will often be rejected by society and lead a pariah existence. Inaddition to violence directly against women, the environment o f insecurity associated with conflict or a post-conflict situation can have a disproportionate adverse effect on women, as women's higher vulnerability in conflict may lead to imposition o f stricter social norms resulting in limited or no access to education and basic health care due to their reduced mobility. Prolonged periods o f conflict, high levels o f social violence, and lack o f security can also be reflected in increased levels o f domestic violence, in which women and children are the primary victims. Conflict and insecurity have profoundly damaging effects on children. The loss o f one or both parents, homelessness, the need to shoulder economic responsibilities inchildhood, exposure to violence and abuse, disruption o f elementary education, and lack of access to health services and immunization are examples o f conflict-induced consequences which have profound and life-long impacts on children. 1.23 Positive synergies between security and development. From a more positive perspective, "The security o fpersons, property and assets, and the protection o f humanrights are fundamental to sustainable development and a pre-condition for people to improve their lives, particularly the poor." (OECD 2001, p. 37). Assets should be broadly interpreted to include public goods such as roads, common water points, and public infrastructure as well as social infrastructure and private assets such as property. With the holistic definition of the security sector (Box 1.1), the positive linkages between security and development are many and closely interlinked, mirroring the negative linkages outlined above! C.KeyPFMPrinciplesandApplicationto the Security Sector 1.24 As seen inVolume I(Chapter 2), the outcomes o f a country's PFM system and national budget process can be divided into three main categories, namely (i) budgetary aggregates and overall fiscal discipline; (ii) strategic allocation o fresources across sectors and programs; and (iii) efficiently managing the use o f budgetary resources. The six core dimensions o f PFMperformance (credibility o f the budget; comprehensiveness and transparency; policy-based budgeting; predictability and control in budget execution; accounting, recording, and reporting; and external scrutiny and audit) all affect these budgetary outcomes. 1.25 The World Bank's Public Expenditure Management Handbook lists 10 generally accepted principles o f sound budgeting and financial management, shown in Box 1.2. The overall PFM performance framework and principles o f sound budgeting and financial management are widely consideredto be hlly applicable to the security budget and its main components including defense, with a view to ensuring that security sector expenditures like those in other sectors promote the three categories o f good budgetary outcomes. Consolidating and summarizing these points around the three main budgetary outcomes, the applicability o f PFM practices and principles to the security sector becomes clear, as outlinedbelow. Most characterizations of security from a development perspective define it as the absence (or more realistically low levels) o f insecurity, rather than givingit a specific positive definition. See for example Stewart (2004, p. 3). 7 Box 1.2: Principlesof SoundBudgetingandFinancialManagement The World Bank Public ExpenditureManagementHandbook identifies ten widely accepted principles of sound budgetingand financialmanagement. b Comprehensiveness: The budget must encompass all fiscal operations ofgovernment. b Discipline: Decision making must be restrained by resource realities over the medium term; the budget should absorb only those resources necessary to implement government policies; and budget allocations shouldbe adhered to. b Legitimacy: Policy makerswho can change policies during implementationmust take part in and agree to the original policy. Flexibility: Decisionsshouldbepushedto the point where all relevantinformationis available. Predictability: There mustbe stabilityinmacro andstrategicpolicy and infunding of existingpolicy. Contestability: All sectors must compete on an equal footing for funding during budget planning and formulation. b Honesty: The budgetmust be derivedfromunbiasedprojectionsofrevenueandexpenditure. b Information: A medium-termaggregate expenditurebaselineagainstwhich the budgetary impact ofpolicy changescanbe measuredandaccurate informationon costs, outputsandoutcomes shouldbe available. b Transparency: Decision makers should have all relevant issues and information before them when they makedecisions andthesedecisionsand their basisshouldbe communicatedto the public. b Accountability: Decisionmakersare responsiblefor the exerciseof the authorityprovidedto them. These principles are argued by Ball and Holmes (2002) to be fully applicable to the defense sector, and by extension to other parts of the security sector. While considerations of confidentiality may dictate some departuresinterms ofbudget practices, there isno grounds for departing from these basic principles. I Source: World Bank (1998), cited in Ball and Holmes (2002, p. 6). AGGREGATEBUDGETARYOUTCOMES 1.26 First and foremost, and applicable to prioritization o f spending inthe security sector (see below) as well as aggregate security sector expenditures, the security budget and the budgets o f its main components (defense, interior, justice, etc.) shouldbe prepared against clear strategies - at overall, sector, and subsector levels - and corresponding policies. The security strategy in turn needs to be an integral part o f the broader national fiscal and development strategy. These principles are applicable to all parts o f the public sector, and security should be no exception. Specifically in the case o f defense, the strategy needs to be based on a transparent "threat assessment" fully owned by the government. 1.27 Figure 1.2 presents a generic process for defense policy and organizational reviews, which i s broadly applicable to other parts o f a country's security sector as well.. While the full complexity shown neednot be followed inevery case, the main elements needto be present. It i s important that these policy development and organizational review processes are guided from the beginning by an indicative fiscal framework to promote sustainability. These review processes should ideally be as transparent and inclusive as possible. Insofar as they are based on broad consultations among the relevant stakeholders, and if the final product i s made public, for example through the publication o f a policy paper, their legitimacy will be strengthened. Ofkey importancehere is the economic framework, and its linkage with setting the defence policy framework, assessing different fiscal options, and ensuring oversight by relevant executive and legislativebodies. 1.28 Second, as for the budget as a whole and as with any other sector, security sector policies and expenditures must be affordable, both inthe short run(annual budget) and interms o f the implications for spending in future years (e.g. decisions on staff levels, associated non-salary costs, procurement o f equipment, andother sustaining costs). This implies that the security sector needs to be fully incorporated inthe annual budget formulation process, subject to aggregate fiscal constraints and sector ceilings like any other sector. It also means that the security sector must be fully incorporated in any medium-term 8 fiscal projections and planning, for example the Medium-Term Fiscal Framework (MTFF) that has been developed by the Ministryo f Finance to underpinthe ANDs (see Volume I, Chapter 3). Figure 1.2: Process of Conducting Defense Reviews/Developing Defense Policy Frameworks Source: Ball and Holmes (2002, p. 55). 1.29 Thus the principles o f comprehensiveness and discipline (Box 1.2) clearly need to be applied to the security sector. This means that all government transactions in the security sector should be on- budget, including revenues earned directly by security sector entities. Equally important i s the need for transparency and information, without which decision makers cannot make informed choices with respect to the budget. Legitimate issues o f confidentiality may constrain public dissemination o f some information, but this should not prevent sufficient relevant information being made available to decision makers, through devices like closed budget hearings and the like. The principle o f flexibility requires such availability o f information. Inclusion o f the security sector inthe PFM system i s essential to achieve the principles o f honesty and predictability, i.e. basing the budget on unbiased projections o f revenue and expenditure, and stability inpolicies and funding. All in all, application o f these principles and inclusion of the security sector inthe PFMframework will lead to a predictable budget. STRATEGICALLOCATION OFRESOURCESACROSS SECTORS AND PROGRAMS 1.30 Within the agreed overall resource envelope, good public finance management principles imply that resources are allocated according to strategic priorities both within the security sector and across all sectors (including security). Sector strategies are critical in this regard, as are the broad priorities indicatedinthe overallnational strategy for development. 1.31 It is extremely important for strategic prioritization that the principles o f comprehensiveness (see above) and contestability are applied. The former principle means that the security sector i s fully incorporated in the budget and inthe budget process, the latter that all sectors (including security) must compete on a level playing field for funding during budget planning and formulation. This does not 9 imply a mechanical or "equity" based allocation o f resources, but rather that different sectors contest for resources during the budget formulation process on an equal footing based on their role in the national strategy and (for sub-sectors within the corresponding sector strategy). Even where security requirements change unexpectedly, the resource implications should be managed through the budget process. 1.32 Another key aspect o f prioritization is allocation o f resources over time. A classic misallocation o f fbnds found in many countries i s building physical assets like infrastructure or service delivery networks like education and health systems, without planning for or even taking into account the associated recurrent expenditure requirements. The same can occur in the security sector, where lethal and non-lethal equipment i s procured, facilities are constructed, or security forces are expanded, without consideration o f their full life-cycle costs. While in annual budgets this will be manifested as an imbalance between capital and recurrent expenditures, particularly non-wage recurrent spending, from a medium-term perspective the problem i s one o f prioritizing present and hture expenditures between the two categories. Thus a medium-term fiscal perspective i s an essential tool for prioritization as well as for assessing overall fiscal affordability . EFFECTIVE AND EFFICIENT UTILIZATION OFBUDGETARY RESOURCES 1.33 Finally, the resources appropriated, both for the security sector and for other sectors, must be used efficiently and effectively. This requires application o f a whole range o f instruments and processes inthe budget execution sphere (see Volume I,Chapter 7, as well as Volume 11). The principle o f comprehensiveness applies here as well, in the sense that the budget execution process for the security sector should include the same instruments, mechanisms, controls, and monitoring that are used for execution o f the rest o f the national budget. 1.34 Of particular importance in promoting effective and efficient resource utilization are predictability, flexibility, information, transparency, and accountability. In this context monitoring and evaluation, with an effective feedback loop to decisionmalung, will help ensure improving utilization o f budgetary resources in the security sector over time. The need for external scrutiny and audit (with appropriate provisions in line with confidentiality concerns as needed) i s just as great in the security sector as inother sectors. 1.35 The control framework, and also the accounting categories and reporting process, are very important for effective and efficient utilization o f budgetary resources. Hence good practice calls for standardizing these important components o fthe P F M system inall sectors, including the security sector. 1.36 More specifically, perhaps the two most important supporting systems for good budget execution and therefore for effective and efficient utilization o f budgetary resources are sound government financial information and reporting systems and procurement. In line with the principles o f comprehensiveness and consistency o f treatment o f different sectors, the security sector should be covered by the govemment-wide treasury system - otherwise financial information for the security sector will not be available on a comparable basis with other sectors throughout the budget cycle. Related to this, classification systems used for security (including defense) need to be fully consistent with those used in the rest o fthe budget. 1.37 One area where different procedures are followed in many countries in the security sector, especially for defense, is procurement. The overriding objective o f a national public procurement system i s to deliver efficiency and value for money in the use o f public funds, while adhering to fundamental principles o f non-discrimination, equal treatment, and transparency. Procurement i s therefore at the core of the PFM system and contributes greatly to several of its objectives, including efficiency, transparency, and accountability. 10 1.38 Defense, security, and budgeting specialists generally take the position that security sector procurement and acquisition practices should adhere to the same principles that guide procurement in non-security areas of the public sector: fairness, impartiality, transparency, cost-effectivenesdefficiency, and openness to competition. Additionally, it i s essential that there be systematic evaluation o f all major procurement transactions for all forms o fpublic sector procurement. 1.39 Defense budgetingspecialists have also noted that - with the exception o f procurement o f works and commodities (such as construction, clothing, food, fuel, office equipment, general vehicles, and consultancy services) - defense acquisition in particular (and also to a lesser extent other security acquisitions) does exhibit some distinctive characteristics. These relate to: (i)the relative importance o f cost in determining which bid i s accepted; (ii) confidentiality related to national security considerations; (iii) frame formajorweaponsprocurement; (iv) complexity ofdefenseprocurement; and(v)the time existence o f international arms control treaty regimes and national legislation governing arms procurement. However, these distinctive characteristics are deviations in scale rather than principle. Adequate levels o f confidentiality can be maintained without violating basic public expenditure management principles. There certainly should be skepticism about any claims that procurement o f relatively standard works, services, and commodities for defense or any other part o f the security sector should be subject to different rules. 1.40 More specifically, while there are valid considerations of confidentiality that are relevant for some procurement transactions particularly in the case o f defense, this should not prevent the key principles o f good procurement practice from being fully applied throughout the security sector. Not doing so runs the risk o f limitingcompetition, raising costs, and exacerbating vulnerability to cormption (which can be a significant risk particularly in the case o f larger defense procurements). While as noted, confidentiality is a consideration for certain types o f procurement in the security sector (e.g. advanced defense equipment), much o f the procurement for the security forces (food, uniforms, transport vehicles, small arms and ammunition, etc.) does not entail overriding confidentiality considerations for the most part. In any case, it i s important that sound public procurement principles and procedures be applied to the security sector, with appropriate provisions to ensure confidentiality where necessary. 1.41 To summarize the points made above with respect to aggregate budgetary outcomes, prioritization, and effective and efficient utilization o f budgetary resources, it is clear that principles o f good public finance management can and should be applied to the security sector, with appropriate modifications to take into account confidentiality considerations where necessary (but not in a blanket manner or in a way that undermines the fundamental principles o f accountability to elected civil authorities). PFM in the security sector in fact can be subject to the same performance assessment and ratings as developed for the PFM system as a whole (Volume I, Chapter 2 and Volume 11,Part 1). Thus it should be incorporated inthe overall PFMperformance assessment framework. D.ServiceDelivery Perspective . 1.42 As noted earlier, security is an essential public service. Thus taking a service delivery approach makes good sense, and provides some guidance for improving public finance management inthe security sector, particularly with respect to institutions, incentives, and accountability. The distinctive features o f security as a public service needto be taken into account, however. 1.43 Conceptualframework A simple model illustrates the institutions, accountability relationships, and financial flows associated with service delivery and their implications (see Volume I, Chapter 8, and World Bank, 2003b). There are three main actors: the people (as citizens and consumers/clients o f 11 services), the state (as political body and policymaker), and the serviceproviders. The interactions by which these three actors influence and are accountable to each other form an "accountability triangle" (see Figure 1.3a), and there are correspondingly financial flows among them (Figure 1.3b). The people can affect service delivery by influencing policy makers (e.g. by voting or through the advocacy role of civil society), who in turn exert influence on the service provider (this is the so-called "long route" o f accountability). Citizens can also directly influence service providers (the "short route" o f accountability) by selecting the provider (when there is competition), using their voice (e.g. complaints), or making financial contributions (when there i s cost recovery). Another important set o f actors is the donor community, which exerts influence through its dialogue with policymakers, its financing, and its direct contracts with service providers. Figure 1.3: Service Delivery Framework a. Accountability b. Financing Assess S.rvlC. Source: Adaptedfrom World Bank (20036). 1.44 One o f the main messages emerging from this conceptual framework i s that it is often possible to improve service delivery by strengthening the short route o f accountability, i.e. through relationships and accountability mechanisms directly between the service providers and the beneficiaries o fthe service (and their communities). Another key message is the need for donors to reinforce accountability o f service providers rather than inadvertently undermining accountability by making direct payments to them (in cash, salary top-ups, in-kmd contributions, etc.). The conceptual framework can also shed light on financing issues related to service delivery, specifically that there i s scope for cost recovery (a financial flow from beneficiary to service provider) to be an instrument o f accountability as well as for covering the cost o f services. 1.45 Distinctive features of security. Inapplying this framework to the security sector, it needs to be kept in mind that security as a public service has some distinctive characteristics that affect the way it i s deliveredand the options for accountability and financing arrangements. 1.46 First, a state of security of the public from violence o f all kinds comes close to being a pure public good, meaning that everyone in the locality (inthe case o f internal security) or in the country (in the case o f external security i.e. defense) benefits from a situation with a general level o f security. This implies that the set o f actual or potential beneficiaries is n o different from the population of the locality or the country, and that cost recovery from beneficiaries i s not possible. However, access to public security services (ranging from the police to courts) may be far from equitable for all, and moreover individuals may be able to augment their own personal security through expending resources on private security guards, stronger physical protection o ftheir property, locks and alarm systems, etc. 1.47 Second, security in a fundamental sense involves the threat andor use o f force against internal or external threats to security. This capacity to use force can be misused, as seen in Afghanistan's recent history as well as in many other countries affected by.civil conflict. Security forces easily can become a source o f insecurity rather than security for the populace. Moreover, monitoring the performance o f 12 security service providers can be difficult, as security forces (such as police) are armed and potentially can threaten civilian monitors. (See World Bank, 2003b, Box 3.3, p. 54, for a discussion on creating conditions o f accountability for the police.) 1.48 Third, andrelatedto the other two characteristics, effective provisiono fpublic security involves a monopoly in the provision o f the service, i.e. the legitimate exercise o f force, which is a fundamental attribute o f a functioning state (see World Bank, 2005, Box 4.1, p. 45). Competition in the realm o f security results in insecurity rather than improved security. However, competitive tendering can and should be used incontracting for certain security-related services like landmine clearance (as discussed in Box 6.1). 1.49 Fourth, measurement o f outcomes and performance in terms o f security i s not easy, as the presence o f security is essentially the absence o f insecurity and associated adverse outcomes. Hence analyses o f security, for example o f defense as a public good, often use the level o f fundingo f defense (an input indicator) as a measure o f the amount o f defense services provided, which can be misleading and may have little or norelation to actual external security (i.e. being at peace) as an outcome. 1.50 These features apply in varying degrees to different components of the security sector. Defense against external security threats i s essentially a pure public good at the national level, to which all these characteristics apply in full. Internal public security, supported by the police and justice system, comes close to being a pure local public good for the locality concerned (although access may well be inequitable). Justice i s also a public good inthat its existence benefits everyone. However, any particular justice transaction benefits (or harms) individuals, and moreover, in many countries like Afghanistan people in need of recourse may have a considerable degree o f choice as to whether to go through the formal justice system or instead rely on traditional informal mechanisms for this purpose. Landmine clearance has elements o f both a public and a private service; clearly the private benefits are high if the land de-mined is under private ownership, whereas the public benefits are high if it i s publicly ownedused land (e.g. roads or irrigation canals). 1.5 1 Implications of applying the service deliveryframework. One extremely important implication i s that effective oversight by the civil authorities is essential. Particularly in the case of the army and other national security forces, the long route o f accountability i s the only viable one, and this requires that service providers (security forces) are accountable to the state (which i s in turn accountable to the population for ensuring the provision o f security). Legitimate oversight by the civil authorities is also important, and equally with respect to local police, because o f the risk that otherwise power would be abused. 1.52 Second, in the case o f local security forces providing a localized public good, there is scope to enhance the accountability of the serviceproviders directly to the local population they are serving. Police i s a good example. The importance o f good police-community relations has been widely recognized and reflects the importance o f strengthening the short route o f accountability in the case o f local security service^.^ Inthis context, effective complaints channels and other short-route accountability mechanisms are very important, and women and other vulnerable groups inparticular need to have access to them. See, for example, the Police Accountability Newsletter (htto:/lwww.~oliceaccountabilitv.co.z~ewsletter/backIssues.as~), which focuses on Africa. 13 1.53 Third, as far as local security services are concerned (police, and mucho fthejudiciary), it makes sense to consider some form of de-concentration of government oversight even in an overall unitary state structure, as long as the state i s functional at the local level and not "captured" by regional or local powerbrokers. 1.54 Fourth, the importance o f donors reinforcing and not inadvertently distorting or undermining the accountability relationships i s at least as great inthe security sector as inother sectors. Ths implies that donors should as much as possible work through budget channels and not build up parallel structures, avoid undermining the service providers' accountability relationshipwith the state (which can happen for example if they provide resources, e.g. in lund or salary top-ups, directly to the providers), resist "flagging" projects they support to the detriment o f the Government's visibility, and encourage cost containment for example by supporting competitiveprocurement. 1.55 And finally, as discussed in Chapter 6, despite major difficulties it is critically important to develop monitoringframeworks andperformance indicators so that oversight bodies, and ultimately the people, can assess performance with regardto security. 14 CHAPTER 2. AFGHANISTAN'S SECURITY SECTORAND KEY DEVELOPMENTISSUES 2.1 Afghanistan's security sector, reflecting its historicallegacy, recent improvements, andremaining problems and gaps, consists o f several different security forces and institutions. This chapter first briefly summarizes the historical background and outlines the progress made by the security sector since 2001. It then provides an overview o f the present, rapidly evolving security sector in Afghanistan, including the legal framework, its main components, overall size, and salient features. The chapter then discusses the institutional structure, institutional issues, and coordination arrangements inthe security sector. The final section summarizes the key security issues faced by Afghanistan from a development perspective. A. HistoricalBackground and Recent Progress HISTORICAL BACKGROUND 2.2 Afghanistan's history over the past several centuries has been marked by frequent conflicts (external and internal) and political instability, with considerable adverse impacts on people's security at various times. The four decades prior to the 1970s, however, were an exception, during which Afghanistan remained at peace (with external policies o f neutrality and non-alignment, respectively, during World War I1and the Cold War), major internal conflicts were avoided, and some progress was made inbuildingbasic Government security institutions. While the reach o f the national government and its security forces was not great, they were not themselves a source o f conflict until the army coup o f 1978 (despite a tradition o f ethnic bias in recruitment and promotions), and they operated alongside functioning traditional mechanisms for providing security andjustice. 2.3 Nearly a quarter-century o f protracted conflict in Afghanistan, starting in the late 1970s, led to pervasive insecurity throughout the country and destroyed or profoundly damaged the institutions that had been responsible for providing security inthe pre-war period -both formal governmental institutions and traditional mechanisms. Government security forces were misused in ways that reduced rather than enhanced security, e.g. during the Soviet occupation o f the 1980s, and they became increasingly "captured" by factions in Afghanistan's civil war during the 1990s. The emergence o f numerous well- armed militia groups, in shifting loose alliances, meant that the ability to use large-scale violence in pursuit o f objectives was no longer the monopoly o f the government but rather was widely contested, greatly undermining security. The judicial system was similarly undermined, and the imbalance o f relations between political/military powers and the judiciary particularly in rural districts adversely affected and continues to threaten the population's access to justice (see Nojumi, Mazurana, and Stites, 2004). Traditional mechanisms for providingjustice were often captured and used for their own ends by commanders and warlords. And neighboring countries and other powers repeatedly and competitively intervened militarily - either directly or indirectly by providing arms and other support to different factions - contributing ina major way to insecurity inAfghanistan. 2.4 Pervasive insecurity, beyond the damaging effects o f the conflict itself, has been profoundly harmful to Afghanistan's development. Consistent with the discussion on security and development in Chapter 1, key adverse impacts o f the conflict on Afghanistan can be summarized as follows: 0 Erosion o f the authority and capacity o f the state, and in particular the loss o f its monopoly on organizedegitimate use o f force, meant that internal conflict and insecuritybecame the norm. Related, the state no longer provided public goods, notably security, and traditional mechanisms for this purpose were also underminedby exercise o f force by regional and local powerbrokers. 0 The breakdowno fthe judicial system resulted ina pervasive culture o f impunity. 15 e Other public services (even though not widely provided in the pre-war period) were severely affected - virtually stopped in most cases. For example, Kabul University stopped turning out high-quality graduates to work in the public or private sector; combined with the flight o f talent from conflict, this left Afghanistan with very limitedin-country human capital resources. Inthe meantime several generations o f Afghan children got little or nobasic education. e With respectto infrastructure, muchwas destroyed byconflict. Assets such as bridges, dams, and power lines were frequently targeted and sabotaged by opposing groups. But even more significant damage was caused by facilities not being maintainedand falling into disuse. e Conflict also resulted in economic distortions, incentives to shift into the informal economy and illicit activities, use of the latter to finance conflict, and the progressively emerging synergstic relationship between drugs, arms, and conflict. e A s inother conflict-affected countries, women and girls became victims o f sexual assault, as rape and sexual violence were resorted to by armed groups. Forced marriages to combatants were common. Boys likewise fell prey to abduction and sexual assault. The general atmosphere o f fear and intimidation disproportionately affected women's mobility, ability to participate in economic life, and access to basic services. e Moreover, during the conflict gender issues became politicized to an unprecedented extent, undoing the limited progress on gender equity achieved over the previous decades, and culminating in the Taliban policies o f excluding women from any public role in society, with dramatic adverse consequences for their legalprotection and ability to access services. e The effects o f conflict, insecurity, and economic disruption spread beyond Afghanistan's borders. While competition among regional and other powers was a source of insecurity for Afghanistan, smugglingand narcotics trafficking emanating from the country had major regional implications, as didthe wholesale displacement o f Afghans as refugees inneighboring countries. e Finally, the laying o f millions o f landmines during the conflict and the existence o f enormous amounts o f unexploded ordnance resulted inheavy human and economic costs (see Box 5.3). 2.5 While the conflict resulted in enormous social and economic costs in terms o f displacement o f people, disruption of normal life, loss o f livelihoods, etc., Afghanistan's social indicators before the war had been among the worst in the world, and they did not deteriorate in absolute terms. However, the country did not participate inthe major improvements insocial indicators that occurred worldwide during the latter part o f the 20* century, including in low-income countries. Moreover, although the conflict distorted and "transformed" social capital inAfghanistan (see Colletta and Cullen, 2000, for a discussion o f changes in social capital during conflicts, with case studies o f several countries), traditional forms o f social capital remained and helped the population cope. Family, village, and tribal ties continued to be strong, even when group members became separated due to displacement or migration. Traditional social capital and coping mechanisms continue to be very important for most Afghans today. 2.6 Inaddition, some ofthe "legacy costs" ofconflict (see Chapter 1)areevidentinAfghanistan, for example the difficulties faced inresurrecting a functional revenue apparatus, which has had to be built up from an extremely low level. Another legacy o f conflict has been tremendous shortages o f professional, managerial, and technical human resources, which will persist for a while due to the time required for such skills to be built up through education. Landmines, which can cause injury or death and render land unusable for many years, can be considered yet another legacy cost o f conflict, especially severe in the case of Afghanistan which i s one o f the most mine-affected countries inthe world (see B o x 5.3). 2.7 Inconclusion, at the time whenmajor conflict ceasedinlate 2001,Afghanistan faced aprecarious security situation with great uncertainties. State security forces had ceased to function or had been captured by factions. The formal justice system was moribund. Numerous armed groups exercised control in different parts o f the country. The downfall o f the Taliban left a vacuum which inthe absence 16 o f an effective state was quickIy filled by reaonal and local powerbrokers. Opiumproduction was poised to rebound, and criminal behavior, often engaged in or abetted by armed groups, was common. PROGRESSINSECURITY SECTOR STRENGTHENINGAND REFORM 2.8 Enhancing security and reforming the security sector have all along been seen as critically important for Afghanistan's reconstruction agenda, although these topics were dealt with only in general terms in the Bonn Agreement o f December 2001. At two subsequent donor meetings, key targets were set for the security sector, as shown in Table 2.1, but these were fixed without reference to fiscal affordability or a threat analysis and instead reflected complex political judgments. Nevertheless, progress has been achieved in strengthening and in some respects reforming the security sector, particularly inthe last two years. Major successes include the avoidance o f large-scale fighting since late 2001, the steady shift away from a pattern of endemic conflict toward a situation where the people and society become progressively more accustomed to peace, and the gradual extension of the writ of the national government beyond Kabul. However, great gaps remain both in terms of areas that the Bonn Agreement did not cover and in achieving agreed security-related targets. Moreover, although very substantial, international security forces' presence in Afghanistan has been limited as compared to other post-conflict situations. Sector Size Lead Donor Army 70,000 USA Air Force 8,000 USA Border Control Security Force 12,000 USA PoliceForce 70,000 Germany n/a Italy Counter-Narcotics ~~ n/a UK Demobilizationo f militias 100,000 Japan 2.9 Over the past year or so, important gains have been made in terms o f operationalizing both the national army and the national police forces. Moreover, some progress has been recorded in terms of policy and strategy development, as o f mid-2005. More specifically: The Afghan National Army (ANA) now stands reportedly at around 26,000 fully trained and equipped troops, although the projected expansion to 45,000 by the end o f 2005/06 i s unlikely to be achieved.6 ANA interventions in several instances have already made positive impacts in improving internal security and in helping to extend the reach o f the national government into areas previously controlled or disputed by major warlords. There have been gains in police training, with more than 40,000 police trained. A new salary scale for the police i s currently under consideration, but it has yet to be formally adopted pending discussion with the Ministryo f Finance. The Afghanistan New Beginnings Programme (ANBP, the DDR program) has successfully completed disarmament o f more than 63,000 Afghan Militia Forces, with nearly all o f them demobilized and most provided some form o f re-integration support - although re-integration, essentially for the sustainability o f DDR, will take time. (The number o f militia forces to be disarmed turned out to be considerably lower than the original estimate o f 100,000.) In total, 36,571 weapons have been collected with 11,004 heavy weapons, as o f early October 2005. 0 The focus is now shiftingto other (illegal) militias, and a program for Disbandment of Illegally Armed Groups (DIAG) has been initiated, although it i s at an early stage (and requires a different approach as unlike DDRit is not necessarily on a voluntary basis). Reportedly slower than expected expansion o fthe ANA i s due inpart to departures of soldiers, but also to a slowdown in training due to logistical bottlenecks inensuring that troops trained are properly equipped on a timely basis. 17 Twenty-two Provincial Reconstruction Teams (PRTs) are now operational across much of Afghanistan, intended to support efforts to improve provincial security arrangements, create an environment supportive o f relief and development work, and even become directly involved in such work (e.g. building schools, clinics, and irrigation structures). 0 Over 500 judges have been trained through an International Development L a w Organization program, although there are concerns that the amount and quality o ftraining are insufficient. 0 A national threat assessment was prepared in March 2004, and a revised, updated version was completed in September 2005. Interestingly, the threat assessment introduces concern for the fiscal sustainability o f security sector institutions, given the state's limited revenue base and increasing operating costs o f defense and other security sectors. Preparation o f a national security strategy i s at an advanced stage, and a national military strategy has been prepared. Security has been included in Afghanistan's Millennium Development Goals (MDGs) as an additional MDG (see Box 2.1) and is forming a central component o f the Interim Afghanistan National Development Strategy (I-ANDs). Box 2.1: MDGGoal 9 EnhancingSecurity f i e Government's MDG report (Afghanistan Government, 2005b) summarizes the situation, issues, and constraints with respect :o Afghanistan's MDG Goal 9, enhancing security. Y - Security is critical for Afghanistan's reconstruction. Without an adequate level o f security, not only will the country fail to achieve the Millennium Development Goals, but also progress achieved during the last four years will be reversed, increasing the likelihood o fresumption o f large-scale conflict. Considerable progress has been made by the Ministryo f Defense, supported by the UnitedStates, to recruit, train, equip and field the Afghan National Army (ANA). The ANA has established nationwide presence since 2003 and fielded a Brigade into each Regional Command. This has greatly facilitated the expansion o f central Government authority and the rule o f law, supported DDR and will support revenue collection and counter-narcotics efforts. Establishing effective control over the use o f armed forces i s the foundation for all other progress in Afghanistan. Challenges to the creation o f the new ANA include the training, equipping, fielding and paying o f the Army. An estimated 17% o f the GDP was allocated in the 2004/05 national budget for the ANA. Over the next 15 years this is targeted to come down to 3-5%, comparable to other developing countries. ANBP has been quite successful in the disarmament and demobilization components of the DDR process. But in a fragile security environment, it will be impractical to aim for complete disarmament. Forcible disarmament o f commanders and other powerful individuals may lead to increased insecurity. With the passing of a new gun law, a license is now required to possess a firearm. The licensing process offers an opportunity to reduce the incidence o f firearms misuse and gun crime. The human resource requirements for the ANP are substantial. In 2002, less than 10% o f police officers were properly equipped and more than 80% o f the infrastructure had been destroyed during the years o f conflict. At the current rate, the training programs are expected to meet the target o f 62,000 trained personnel by late 2006. There are serious concerns about the fiscal sustainability o f the proposed reforms, especially given the shortfall for LOTFA during its recent second phase, when only 40% o f requestedfunds were received. The major donors are beginningto signal a shift away from completely underwriting the full recurrent budgetary expenses o f the ANP. Furthermore, the pace o f proposed reforms inthejustice sector has to be aligned with attempts to reform and restructure the police. Landmines and unexploded ordnance (UXO) currently affect some four million Afghans. Only two provinces are wholly free o f the problem. The unit cost o f clearing the remaining identified and surveyed minefields in Afghanistan has been estimated at $1 per sq. m. In a situation where there i s a large movement o f population, and development and reconstruction projects impeded by mines/UXO, it is necessary to expand the de-mining program and make faster progress toward eliminating the problem o f landmines. Opiumdominates the national economy. It accounts for one-third of estimated total GDP and 50% of legal GDP. The total areaunder poppy cultivation increasedby nearly two-thirds during 2003/04. Even as the Government attempts to reduce and eventually eliminate poppy cultivation, there is a risk that targeting poor farmers and landless laborers involved in poppy cultivation with eradication and law enforcement measures may increase poverty in the short-term, until genuine and- sustainable alternative livelihoods can be developed. The cost o f securing Afghanistan from terrorism, insurgency and tribal conflicts, organized crime, narcotics production and trafficking i s high. The Government needs to follow a two-pronged approach o f sustaining high levels o f economic growth while increasing domestic revenue to ensure long-term fiscal sustainability o f the security-sector reforms. 2.10 In summary, there has been significant progress in building up Afghanistan's security sector, broadly in line with the directions set in the Bonn Agreement and at subsequent security meetings. 18 However, there is still a very long way to go, and as will be seen later in this chapter, numerous issues remain relatedto the security sector and the important security problems affecting the lives o f Afghans. B. Overview ofAfghanistan's Security Sector LEGAL FOUNDATION 2.1 1 The 1382 (2004) Constitution provides the legal foundation for Afghanistan's security sector. As noted inBox 2.2, the Constitution does not provide much inthe way o f specifics and hence there i s a need for legislation for this purpose. More generally with respect to the legal and justice system, although there has been considerable progress in finalizing and promulgating legislation on various subjects in recent months, a large backlog remains o f 100-plus laws and regulations at various stages o f preparation and approval. Capacity in the Ministry o f Justice to review and process draft laws and regulations is an important constraint, and the recent election o f Parliament means that all new legislation will require Parliamentary approval, which may result infurther slowing down the passage o fnew legislation. Box 2.2: Legal Foundation of Afghanistan's Security Sector The 1382 Constitution, promulgated in January 2004, provides general guidance on the nature and objectives o f the security sector institutions. Article 5 states (according to unofficial translation) that: "implementation o f the provisions o f this constitution and other laws, defending independence, national sovereignty, territorial integrity, and ensuring the security and defense capability o fthe country, are the basic duties o f the state." Article 64 vests substantial power in the hands o f the President, who is the Commander-in-Chief o f the Afghan National Army. The President is also responsible for: Declaring war and ceasefire with the confirmation o f the National Assembly; Taking the required decisions to defend the territorial integrity and protect the independence o f Afghanistan; I Sending contingents o f the armed forces to foreign countries with the confirmation o f the National Assembly; Appointing Ministers, the Attorney General and Head o f the National Security Directorate; Appointing the head and members o f the Supreme Court with the confirmation o f the Wolesi Jirga; and Appointing, retiring, and accepting the resignation o f and dismissing judges, officers o f the armed forces, police, national security, and high-ranking officials in accordance with the law. Other sections o f the Constitution deal with the judiciary, the rights and responsibilities o f citizens, and the role o f the Government in providing security, the role o f the National Assembly, the declaration o f states o f emergency, and the protection o fhumanrights. Internationally, constitutions increasingly specify (a) lines o f authority between major security sector actors (civil authorities and security services); (b) the basic responsibilities o f each o f these actors; and (c) broad democratic principles to which the security-sector institutions should adhere. Afghanistan's Constitution has very little detail o f this nature. While the President i s named as commander-in-chief o f the national army and the Constitution contains an entire section on the judiciary, it i s essentially silent on other parts of the security sector. This means that considerable subsidiary legislationwill be required. Although the 1382 Constitution does not specifically direct the President or the Government to develop laws governing the security sector, the need for legislation for at least some portions o f the security sector has been recognized. The National Security Law, which regulates the National Directorate o f Security, was approved by the President in March 2004. M o I i s drafting a police law and i s working with the Ministry o f Justice to redraft the existing counter-narcotics law. The UNOffice on Drugs and Crime is drafting a Penitentiary Law for the Ministry o f Justice, while the latter indicates that it is reviewing a law on armed forces officers and a prisons and detention centers law. In view o f the emphasis that has been given to rebuildingthe country's security services, the approval o f laws governing these bodies should be accorded similar priority. MAINCOMPONENTSANDSIZE 2.12 Inlinewiththe definition of the security sector put forward in Chapter 1, the main components o f Afghanistan's security sector are briefly describedbelow. 2.13 Securityforces. These include the ANA (now around 26,000 fully trained and equipped troops); the ANP (roughly 70,000 police, only some o f whom are trained); elements o f the National Directorate o f Security (doing intelligence work); several specialized counter-narcotics forces; border guards; and penal staff. As mentioned earlier, decisions about the numbers o f personnel in the various national security 19 forces in Afghanistan were determined early on, inthe absence o f an agreed threat assessment, a national security strategy, sectoral policies, or consideration o f how much it would cost to maintain a security establishment o f a particular size. Targeted staffing levels instead represented political compromises related at least inpart to plans at the time to accommodate existing armed groups (after DDR) inthe new security services. 2.14 Management and oversight bodies. The main management and oversight bodies in Afghanistan's security sector include the National Security Council, National Security Advisor, Ministry o f Defense, Ministry of Interior, Ministry o f Justice, Ministry o f Border and Tribal Affairs, the recently established Ministry of Counter Narcotics, the Anti-Corruption Office, and (with respect to enforcement actions by customs agents) the Ministryo f Finance. Many o f these entities have corresponding provincial departments. There are no obvious anomalies in the constellation o f security sector oversight bodies, although most o f them suffer from limited capacity. The oversight and coordination o f counter narcotics activities has been limited and fragmented in the past, and the creation o f the Ministry o f Counter Narcotics i s intended to remedy that. Once Parliament i s established (in December 2005), there will be civilian legislative oversight o f the security sector and nationalbudget. 2.15 Justice and law enforcement agencies. These includethe Supreme Court, lower-level courts, the Attorney General's Office and sub-national prosecutorial staff, and prisons, as well as customary and traditional justice bodies. Another significant body has been the Judicial Reform Commission, although its role has been diminishing and it did not evolve, as intended by the Bonn Agreement, into the main driver o f the judicial reform process. 2.16 Non-statutory security actors. Although this part o f the security sector i s not discussed in detail inthe PFMReview, the substantial size and rapid growth ofnon-statutory providers of security services has potentially important implications and deserves to be emphasized. In addition to the "official" Afghan Militia Forces (AMF), which have now been eliminated through DDR, and unofficial armed groups still in existence, various private and non-Government security entities have grown rapidly in response to demand. These include international contractors, domestic security firms, de-mining agencies (organized as NGOs), armed groups providing security under contract, and individual guards. Private security services are contracted for a variety o f purposes, including among others (i) security for foreign and international agencies' premises (including CoalitiodSAF in some cases); (ii)security for reconstruction projects, including security services for international contractors carrying out such projects; (iii) security for road missions o f aid agencies; and (iv) bodyguard and close protection services. Furthermore, during and after the war to dislodge the Taliban, non-statutory armed groups such as local militias inmany cases were provided fundingthrough payments from external forces. 2.17 Inaddition, government security forcesare oftencontractedandpaidfor similartypes ofsecurity services by the international community (guarding o f premises and provision of security for road missions are two common examples). Thus there i s a grey area and a blurringo f distinctions between public and private provision of security, with potential for confksion o f roles, distorted incentives (e.g. for government security forces to in relative terms neglect their public responsibilities while pursuingwork for which they are paid), and possible irregularities. 2.18 While no estimates are available o f the number o f private and unofficial security personnel and companies, and the volume o f security services they provide, given the size o f the foreign presence and ongoing reconstruction efforts, it i s clear that a great deal o fprivate security provision i s occurring. 2.19 International security forces. International security forces have a substantial presence in Afghanistan, and their cost exceeds that o f all Afghan security forces combined by 10-foldor more. They can be divided into three broad categories: 20 The US-led Coalitionforces, now numbering approximately 20,000, o f which U S military forces account for the vast bulk. Estimated annual cost i s inthe range o f $12 billion per year. The International Security Assistance Force (ISAF), now under NATO command, numbering some 5,000 military personnel from numerous countries. Estimated cost is roughly $3 billion per year. Leadership o f I S M , which i s concentrated in Kabul, has been rotating among several o f the main providers o ftroops for I S M . The 22 PRTs that have been established in provincial capitals, with a mandate to help improve governance and support reconstruction as well as to facilitate improved security in the localities concerned (see Box 2.3). Some PRTs are part of and report to the Coalition, others are under I S M . The current PRT structure currently allows the 22 unitsto cover all 34 provinces. integrated within the national budget framework; and (viii) apparently highunit costs for certain activities. Importantly, lack of sub-national capacity in the Government often left highly-qualified PRT units without a clear development counterpart at the provincial level, undermining coordination o f efforts and integration with other ongoing reconstruction activities. Currently efforts are underway to consolidate PRTs and provide greater clarity for their missions and management structure. 2.20 Size and staffing. Available information on the number o f staff indifferent parts o f the security sector i s provided in Table 2.2. Staffing information is subject to inaccuracies and fluctuations, and regular payroll data has not been easily available for some parts o f the security sector (e.g. MOD,MoJ, and NDS). Nevertheless, the data do provide a clear indication that Afghanistan's security forces are staffing up toward targeted levels. Also, inflation innumbers o f civilian staff inoversight bodies does not appear to be occurring. The sharp decline in authorized total for MODreflects the disbandment o f the existing AMF which has now been completed, and its progressive replacement by the ANA currently under expansion. SALIENT FEATURES 2.21 Based on the legacy o f conflict and recent developments, Afghanistan's security sector has some important characteristics with significant fiscal and development implications. 2.22 First, as discussed earlier, the historical legacy o f conflict was fragmentation into regional and local militias, capture o f policing and military functions by non-legitimate actors, and erosion of both formal and traditionaljustice systems, also with capture by non-legitimate authorities. Buildingthe state since 2001 hence has entailed reconstituting and reforming the security forces under legitimate oversight 21 by the civil authorities. While significant progress has been made in this regard, rule o f law is still lacking in many respects and in many parts o f the country, and insecurity remains a major constraint to state building,public service delivery, and private sector development. Table 2.2: Approved StaffingEstablishmentin the Security Sector MinistryofIntenor MinistryofBorder and Tribal Affairs MinistryofCounterNarcotics Note: Average numbers onpayroll refer to civilian employees only, i.e. the civilian employees of oversight bodies. Data in other columns refer to authorized levels of staflng Source: ARTF Monitoring Agent (2005). 2.23 Second, the immediate security threats faced by Afghanistan are variegated and disputed by different stakeholders. Some observers see the continuing insurgency in the south and east and conflict with TalibadAl Qaeda forces as the most important security problem. Others are most concerned about criminality and lack o f rule o f law at the local level, with associated capture o f policing and justice functions by warlords and commanders. Still others are concerned about the lack o f civil authorities' management capacity and inadequate oversight/accountability o f the Government's security forces. The drug industry is widely seen as a major cause and consequence o f insecurity. The development o f an agreed national security strategy, currently under preparation, will facilitate reaching consensus on the different security threats and their relative priority. 2.24 Third, the overall size o f the security sector, even with recent and planned expansion, does not seem unreasonably large by international standards. These include a planned size o f the ANA o f 70,000 (not all o f them combat troops) and a planned size o f the Afghan National Police (ANP) o f 62,000. However, the appropriateness o f the size and balance between different security forces can be questioned inthe context ofAfghanistan's particular situationand security needs. For example, the role o fthe ANA in meeting different external and internal threats would need to be clarified in order to assess the appropriateness o f its targeted size. 2.25 Fourth, Afghanistan's security sector nevertheless i s costly and could well be unaffordable. The lack o f short-run budget discipline (and lack o f policies and programs to form the basis for sound budgeting), and the involvement o f different "lead donors" and line ministries in decision-making on force levels and expenditures, have resulted in limited efforts to control costs. Over the medium term, it i s extremely doubtful whether the sustaining costs o f Afghanistan's security sector at planned force levels can be absorbed by Afghanistan's national budget (as will be discussed in detail in Chapter 4). This implies that difficult trade-offs will have to be made in prioritizing expenditures, means o f achieving efficiencies in security-related expenditures will need to be identified, and the possibility o f multi-year donor financing commitments for the security sector should be explored. 22 2.26 Fifth, the progress in developing different parts o f the security sector has been uneven. After a slow start initially, the formation and expansion o f the ANA have moved forward rapidly over the past two years. Progress in forming the new ANP has been much slower, andreform and capacity buildingin the justice sector have lagged far behind, with adverse implications for legal reforms and a pro-private sector business climate. Moreover, within sectors there has been a disproportionate focus on buildingup security forces and less attention to strengthening the key management and oversight institutions, most notably the Ministryo f Defense and especially the Ministry o f Interior. 2.27 Sixth, security and fiscal issues are closely interlinked. For example, l o w domestic revenue i s both a consequence and cause o f insecurity. Reconstruction costs tend to be higher due to insecurity, while on the other hand the security sector i s a source o f considerable fiscal pressure. And pay and grading decisions withrespect to the security services can have important fiscal implications. C. InstitutionalStructure and Coordination 2.28 Security sector institutions, and coordination among them, are critical for achieving national security objectives. Table 2.3 summarizes institutional roles in the security sector, and points out gaps and areas o f duplication. A strilung feature i s the many different entities responsible for various aspects o f counter-narcotics activities, and the bifurcation o f oversight between the newly-formed Ministry o f Counter-Narcotics (MCN) and the wing o f M o I responsible for counter-narcotics law enforcement. Moreover, some counter-narcotics forces and activities are closely supervised (and directly funded) by donors, which further fragments management and administration o f these functions arrangements. The M C N i s intended to provide overall coordination but as a new ministryrequires time to buildup capacity for this purpose. The capacity and effectiveness o f institutions in the justice sector remain a serious concern. And although as discussed later in this section there has been major progress in developing security sector coordination mechanisms under the leadership of the NSC and NSA, there is room to further enhance their effectiveness, particularly inthe context o f the national security strategy. INSTITUTIONAL AND ADMINISTRATIVE REFORMS 2.29 The pace o f institutional and administrative reforms inthe security sector has been slow, although the Ministry o f Defense and National Directorate for Security have been exceptions. With the focus on restructuring, training, and equipping the ANA, the ANP, and the various counter-narcotics forces, reforms o f the parent ministries have received less attention. In most cases, insufficient emphasis has been placed on developing sustainable institutions. Lack o f reform inM o I in particular, under which all provincial and district administrationoffices are accountable, has been a major disappointment. 2.30 Restructuring and capacity building support have not been well-coordinated, have been driven by project financing and, by and large, have neglected the need for performance-based recruitment, upgrading o f administrative and financial management, and unifying duplicate functions. Gains in the performance o f some security institutions have been made outside the Government's agreed administrative and civil service reform framework, but many o f the Core Budgetary and other institutional functions have yet to be strengthened. 23 Table 2.3: M a i n National Entities Involved in Security, their Roles, Gaps and Duplications I Entitv I Roles I Gaos/DuDlications Ministry o f Defense (MOD) Provide overall strategic direction, develop policies Afghan National Army (ANA) Defense o f state and Islam against external aggression, Overlap ininternal security support political process, disarm illegal groups, counter- with ANP terrorism activities. IAfghan Militia Forces (AMF), DDRofAMF completedbutreintegration continuing; Illegal Armed Groups (IAG) IDisbandment o f IllegalArmed Groups (DIAG) underway 1 Law Enforcement Ministry o f Interior (MOI) IOverall responsibility for law enforcement Afghan National Police (ANP) IHandles policing functions nationwide 1Uncertain role on counter- further clarification. 2.3 1 The Priority Reform and Restructuring(PRR) tool has not been usedto support the restructuring of key security-related institutions. The only exception i s the PRR application o fMoJ, which hasrecently been approved by IARCSC (see Table 2.4). In 2004/05, the IARCSC commenced reforms o f Provincial Governor's Offices, which have security, modest service delivery, and development coordination 24 functions. A new institutional structure i s beingpiloted inBalkh, and the recruitment o f 21 qualified staff there i s a major step forward. This pilot reform i s also being implementedin Samangan and Jawzjan. Table 2.4: Current Status of Securit; Sector PRR Applications PRR Stage 1 PRR Stage 2 Security Body 4. UnderConsideration A. Under Consideration Ministryof 3ffice o f Governor in the following Office o f the Governor and ProvincialGovernment o f [nterior provinces: Nangahar, Kandahar, Kunduz, Samangan Herat, Paktia, Faryab, and Badakhshan. B. Approved: B.Approved SamanganProvince Office o f the Governor & Provincial Administration Jawzjan Province of Balkh Office of the Governor & Provincial Administration o f Kabul Province MoI (Proper) Office o f the Minister Local Governments Department (Province, MunicipalityAIistricts) Population Registration Department Finance and Administration Department Justice B. Approved Justice Ministry except Directorate o fPrisons and Directorate o f Juvenile Prisons. The departments and numbers o fpersonnel involved are: 2. Office o f Secretary General - 9 (new) 1.Office o fthe Minister - 10(from 42) PRR Task Force - 5 3. Legislative Department - 59 ( from 93) 4. Coordination, Evaluation & Registrationof PoliticalParties and Social Organizations - 15 (from 33) 5. Auditing and Internal Inspection- 24 (new) 6. HuquqDirectorate - 22 at central: 264 in 33 provinces and 748 in374 District Huquqs 7. Directorate o f Publications- 17 (from 27) 8. Government Cases - 26 (from 44) Note: N o recruitment is happening A. Under Consideration A. Under Consideration Attorney Attorney General's Office PRR Stage 2 under consideration but General's Office cooperatiodworki s slow A. Under Consideration A. Under Consideration Ministry of PRR Proposal Stages 1 and 2 for entire This i s a new ministry with the following proposed Counter Ministryisundertechnical evaluation. offices/departments: Narcotics 1. Office o f the Minister 2. Internal Audit Office 3. Office o f Secretary General including Task Force 4. Department o f Eradication 5. Department o f Law Enforcement and Interdiction 6. Department o f Counter Narcotics Policies and Operation Planning 7.Provincial Counter Narcotics Operations Offices 8. Directorate o f Finance, HR and Administration. 2.32 As o f February 2005, o f the 34 Government agencies that had attained PRR status, not a single one had a direct mandate for managing provision o f security or promoting law and order. Moreover, providing security, justice, and rule o f law country-wide requires both provincial and district level reforms, and progress at the sub-national levels has also been slow. T o take one example, unless the Department for Local Government inMoIi s strengthened through the PRR process, it will be difficult, if not impossible, to implement effective reforms in the Provincial Governors' offices and in the District 25 Administrations. Progress inreforming this department has been limited, and given that elected Provincial Councils will soon be established, weak Provincial Governor's offices are likely to be a significant constraint impeding the Provincial Councils' effectiveness inmonitoring and enhancing governance. 2.33 Currently, a number o f security-related agencies have attained both PRR Stage 1 and Stage 2 applications (see Table 2.4), but the sequence o freforms does not reflect the logic o f the process, starting at the center, moving to the provinces, and then reaching to the district level. While there may be some exceptions to this principle, reforming Provincial Governors' offices prior to reforming the Administrative Affairs Department in M o I to which they report could lead to unnecessary complications and may not achieve the desired results. Furthermore, in line with the requirement for greater inter- ministerial coordination o f the security sector reform program, it i s vital that the NSC and the ONSC be directly involved inoversight o fthe reformprocess, both incentral Ministries and outside Kabul. 2.34 There has been a tendency in the security sector to develop and implement administrative reform plans outside the PRR process. Such plans have been put inplace for the ANA and ANP without having been agreed by IARCSC, undermining the transparency o f the reform process. There has also been lack of progress in reorganizing M o I and MoJ. It would be useful if the structuring o f security sector management ministries conforms to the formal PRR process, which has been revised and offers a reasonable framework for drivingreforms inthe security sector ina coherent manner. 2.35 The lack of external pressure for institutionalreforms probably reflects two factors: (i) the actors most heavily involved in the security sector give relatively limited attention to institutional reform o f ministries (with some exceptions) and very little attention to reform o f management and oversight bodies, and (ii) the development donors, which do focus more on institutional reform o f ministries and oversight bodies, have had a limitedrole inthe security sector inthe past. However, it i s interesting to note that the OECD Development Assistance Committee recently clarified the definition o f Official Development Assistance (ODA), and technical assistance for the "management o f security expenditure through improved civilian oversight and democratic control o f budgeting, management, accountability and auditing of security expenditure" i s now eligible to be counted as ODA (OECD, 2005a). 2.36 Currently, MODi s being restructured without recourse to the PRR fkamework. The Counter Narcotics Ministry on the other hand i s aligning its reforms to the PRR process. It has sufficient international technical assistance to make some progress on this and may become the first example o f systematic andcomprehensive reforms ina security sector body. COORDINATION INTHE SECURITYSECTOR 2.37 The process of developing and implementing a national security strategy and sectoral security policies, in the context of strengthening public finance management, will be greatly facilitated to the extent that the different parts o f the security sector work together. In a country such as Afghanistan, where international actors are directly engaged in security-related activities and provide the bulk o f the resources for restructuring and reforming the security sector, effective coordination between national actors and international partners, as well as within the international community, also i s very important. 2.38 National coordination. The responsibility for coordinating security policy inthe Government of Afghanistan i s located in the National Security Council, an inter-ministerial body, and the National Security Advisor (NSA) who reports to the President, supported by the Office o f the National Security Council (ONSC) which advises the President on national security issues. The ONSC has prepared the national threat assessment and a draft National Security Strategy. However, although the NSC i s mandated to handle inter-ministerial coordination, at the outset this was often difficult, since some o f the factionalized security ministries were powerful and tended to act independently. Furthermore, in many 26 areas (particularly defense), Afghan ownership o f policy malung was limiteddue to the heavy presence o f external military institutions and advisors. There have been major improvements in the situation over time, as line ministries have become somewhat less factionalized and as the role o f the NSC and ONSC has developed. Since early 2004, the NSC, NSA, and ONSC have increasingly taken on their mandated roles o f strategic leadership and coordination, as i s now fully recognizedby ministries and external actors. 2.39 The Government's paper submitted to the 2005 Afghan Development Forumnoted that "in order to ensure that SSR activities are coherent at the national level, two strong strategic-level coordinating committees have been established that have direct responsibilities for the management o f the SSR Strategy and National Security Strategy; respectively called the SSR Coordination Committee (SSRCC) and the National Security Coordination Forum (NSCF). They also have the responsibility, on behalf o f the NSC, o f ensuring de-confliction and that their two strategies are mutually reinforcing to the maximum extent possible.'' A Deputies Committee is also being established to develop inputs for the NSCF and SSRC and to ensure communication and implementation o f the decisions made by these bodies. A diagrammatic representation o f the National Security Policy Coordination mechanisms i s provided in Figure 2.1. Box 2.4 provides further detail on the three levels o f coordination now inplace. Ofparticular note in this regard i s the creation o f an SSR Coordinator in the ONSC. This coordination structure has allowed greater coherence to emerge while strengthening Government ownership. Figure 2.1: Security Sector Policy Coordination Note 1 - JointSecretariat from ONSC. CFC -A and UNAMA Note 2- Election Security? Source: Afghanistan Government (2005a. Theme 8: "Security, Justice and Law and Order"). 2 10 International coordination. Formally established in April 2002 at a security donors conference in Geneva, Switzerland, Afghanistan's security sector reform agenda consists o f five pillars, each supported by a different lead donor: 0 Military reform (ANA, Border Guards, Air Force) -U S lead; 0 Police reform- Germany lead; 0 Disarmament, Demobilization, and Reintegration (DDR) -UNlead funded by Japan; 0 Judicial reform-Italy lead; 0 Counter-narcotics -UK lead. 2.41 Although intelligence, inthe form o f the National Directorate o f Security (NDS), does not have a formal pillar to support it, the N D S receives financial and other support from Germany, the United Kingdom, and the United States. 27 Box 2.4: Security Policy Coordination at Different Levels The introduction o f a management level between the NSC and the various subordinate steering groups and working groups is designed to enhance both horizontal and vertical coherence of all SSR and security related issues. There are three levels o f management as indicated in Figure 2.1: level 1 = political or grand strategic; level 2 = strategic; and level 3 = operational (policy). In addition to chairing the NSC at level 1, the President chairs the SSR Coordinating Committee (SSRCC), at level 2. This ensuresthat the President remains accessible and closely engaged with the steering groups at level 3. The presence o f the NSA as chair o f the National Security Co-ordination Forum (NSCF) and as Deputy Chair o f the SSRCC will promote coherence between the activities o f the SSRCC and the NSCF. SSR Coordination Committee: The purpose o f the SSRCC is to coordinate all SSR related activities and to develop strategies and issue directives to ensure effective vertical and horizontal coherence. The President as chair o f the SSRCC is assisted by the appointment o f two subordinate co-directors: the NSA and the SRSG. The commanders o f Combined Forces Command- Afghanistan (CFC-A) and ISAF also attend as principals. The role o f the Commander o f I S M may increase as I S M assumes more security responsibilities from CFC-A. Membership o f the SSRCC will include the co-directors (co-chairs), o f the five SSR Steering groups. As per the NSCF, the secretariat of the SSRCC is jointly provided by the Office o f the National Security Council (ONSC) and CFC-A and will perform similar functions to thejoint secretariat o fthe NSCF. Deputies Committee: Chaired by the DeputyNational Security Advisor and includingthe deputy ministers o f the ministries represented on the NSCF and SSRCC, the DC identifies policy gaps and develops inputs for the NSCF and SSRC, and also ensures communicationandimplementationofthe decisions made by these bodies. SSR Steering Groups: SSR Pillar Groups have established their own coordination mechanisms and working groups. The purpose o f the SSR Steering Groups is to coordinate all activities within their SSR areas o f responsibility, to develop strategies, and to issue directives to ensure effective vertical and horizontal coherence and the achievement of strategic goals as directed by the SSRCC. SSR Steering Groups decide upon their own composition and whether it is necessary to create their own architecture o f supporting sub-groups. They also provide their own secretariats and decide on the frequency o f their meetings, but meetings should be at least one per month in order to prepare for the monthly meetings o f the SSRCC. SSR Working Group: An SSR Working Group has been established below the Principals' level o f the SSRCC involved in the five SSR pillars from the International Community. The purpose o f the SSR Working Group (SSRWG) is twofold: first to discuss and propose the agenda to the National Security Advisor (NSA) for the main SSRCC and second to ensure that Afghanistan's international partners are fully informed, engaged, and able to brieftheir principals. SSR Coordinator: InNovember 2004 the post o f SSR Coordinator was established inthe ONSC. On behalf of the Afghan Government and the NSA, the role o f the SSR Coordinator i s to effect strategic coordination between the SSR activities o f the International Community and the responsible Afghan line ministries. Source: Governmentof Afghanistan (2005a, Theme8: "Security,Justice and Law and Order"). 2.42 This structure places a premium on the ability o f international actors to coordinate both among themselves and with the Government ifresources are to be used as effectively and efficiently as possible inthe security sector, andifthe security bodies created are to be sustainable over the longterm. To date, coordination among the lead donors has often been inadequate, with donors to a considerable extent acting bilaterally. For example, the U S has played an increasingly central role in police training, first through the Department o f State and more recently through the Department o f Defense, due to U S concerns about the slow pace o fpolice training efforts. 2.43 It is difficult for the Government to impose coherence and to harmonize the priorities of individual donors with the national strategy and priorities. This i s well illustrated inthe case o f counter- narcotics. Despite the existence o f a cross-cutting agency for counter narcotics (MCN, previously CND), an agreed national strategy (the NDCS), and a designated strong lead donor (the UK), there are multiple national agencies and a number o f different donors involved. One donor has commented that "legislation.. .policy development.. .money.. .they are all going in different ways". Another donor observed that "up to now there has been n o coordinated implementation plan for NDCS, so it i s not surprisingthat there is little connectionbetween strategy andprograms." 2.44 Inthe face of these challenges, a significant effort hasbeenmade to set upcoordination structures for counter narcotics, as discussed earlier. In 2004 counter narcotics was treated for the first time as a "cross cutting theme". A Counter Narcotics Advisory Group in the consultative group structure was 28 proposed by the Ministry of Finance, but t h s was not implemented. Instead a Counter Narcotics Steering Group (CNSG) was established in the spring o f 2004 and met regularly for information sharing, overall coordination, and progress monitoring with respect to the full range o f counter-narcotics policies and actions. It was chairedjointly by the Government (National Security Council) and the UK as lead donor. The group has been useful for sharing information but did not play a role in coordinating strategic decisions or resource allocation. More recently, the coordination structure has been changed again. A Counter-Narcotics Subcommittee o f the Cabinet has been formed, including the Ministries o f Counter- Narcotics, Interior, Finance, and Foreign Affairs, and also with, most unusually, key international actors (UK, US, UNODC, etc.) joining the meetings of this Cabinet Subcommittee. A larger group including other donors and international agencies i s convened fairly infrequently, mainly for briefings and information sharing. 2.45 Without effective Government-led coordination, the various investment projects in the security sector risk being little more than the sum o f individual projects. Furthermore, projectization does not facilitate a more rationale debate on key policy issues such as fiscal sustainability (which requires serious thought on strategic prioritization within the sector as a whole, as well as on pay and grading reforms) or institutionalreform within the security ministries. 2.46 The Government's efforts to strengthen Afghan ownership o f the security sector reformprocess, working through a restructured SSR Coordinating Committee where the Government i s in the lead and Afghan stakeholders are well represented, offers an opportunity for constructive dialogue in the Government on security priorities that can be linked to the budgeting process. In particular, this could help shift expenditure decisions inthe security sector from a supply-driven to a demand-driven approach. Additionally, financing Government security priorities in a sustainable manner will be facilitated by increasing the amount o f donor resources for the security sector that flow through the Core Budget. D.Key Security Issues from a Development Perspective 2.47 Security i s widely considered the Number One issue for Afghanistan's reconstruction and development agenda (Afghanistan Government, 2004, p. 78). Dealing with the legacy o fnearly a quarter- century o f protracted conflict, and with the difficult security issues currently facing the country, has been and will continue to be a major challenge. The main security issues in Afghanistan from a development perspective (not necessarily in order o f importance) are outlined briefly below. Needless to say, many of them are closely interrelated with each other. 2.48 Continuing conflict/insurgency/terrorisrn and associated insecurity. Although outright insurgency and significant military conflict are concentrated in areas inthe south and east o f the country (where much terrorist activity i s occurs as well), this conflict has been continuing on a significant scale albeit with seasonal and other fluctuations, and i s adversely affecting security in these areas with important development implications (along the lines of those described in Chapter 1, Section B). Elsewhere, sporadic acts of terrorism have occurred from time to time, but not on a widespread scale or in a sustained manner. These have a disproportionate impact on perceptions o f the security situation particularly from outside the country, and hence deter foreign investment. 2.49 Another adverse implication i s movement restrictions on aid agencies and higher security costs for development activities, as well as for private business. For example, highway reconstructionprojects reported that their direct spending on security added 3-15% to project costs (see Volume IV, Chapter 5), and a recent survey o f private business firms found that their reported spending on security infrastructure accounted for close to 15% o f sales revenue, and that an additional 3% o f sales was reportedly spent on protectionpayments to commanders, government officials, or organized criminals (World Bank, 2006). 29 2.50 Lack of rule of law and poorly functioning judicial system. This i s a fundamental problem, seriously affecting Afghans intheir daily lives, which i s both a manifestation o f underlying insecurity and itself contributes to continuing insecurity. Rule o f law and the judiciary were weak during most o f the past quarter-century, and restoring let alone enhancing these critical components o f the security system will be a major challenge. 2.51 More specifically the police, weakened and largely captured by factional forces during the conflict (see below), has often been more a source o f insecurity than o f security for most Afghans. Anecdotal evidence indicates that District officials are unable to reliably resolve local disputes in their official capacity, unless they have personal prestige through their traditional connections to do so. Few police officers in rural areas have professional training, and many are former mujahedin or militia members who became police officers when their commanders joined the Government after 2001. Police forces continue to operate largely independently from a centralized command structure under the Ministry o f Interior, and many remain loyal to their former commanders. The police force i s overwhelmingly male, and many districts lack detention facilities for women. The lack o f women police officers to handle the arrest, detention, or transport o f female detainees places any woman arrested at heightened risk o f abuse (Nojumi, Mazurana, and Stites, 2004). 2.52 The formal judicial system was devastated duringthe conflict, and its rebuilding since 2001 has been slow. In addition to the lack o f recourse for Afghans and inability to effectively punish serious crimes (including those committed during the conflict), the weakness o f the formal judicial system i s a deterrent to business activity in the formal sector and to foreign investment. Although historically Afghanistan has relied much more on customary and traditional practices and adjudication mechanisms, these also were significantly eroded in many respects duringthe conflict, and/or "captured" by warlords and commanders, with adverse effects on local security as well as on business activities and livelihoods. 2.53 An important manifestation o f these problems i s the widespread uncertainties and disputes over land, leading to ambiguities and insecurity related to land tenure as well as theft and illegal appropriation o f land (see Wily, 2003). Confirming other information, recent fieldwork in Nangarhar and Laghman provinces found that conflict over agricultural land (and related water supplies) was the most common source o f violence (Koehler and Zurcher, 2005). Access to land i s the second most widely cited constraint on business activity ina recent survey o f firms (World Bank, 2006). 2.54 Finally, there i s a long tradition o f the Judiciary being an exclusively male domain, and only around 3% o f judges currently are women. With the exception o f the heads of the juvenile and family courts in Kabul, women are not found in key positions within the judiciary. Where women do serve as judges, reports indicate that they do not perform the same functions as their male counterparts. Female judges tend to act in the capacity o fjudicial clerks and are rarely involved in the adjudication o f cases, which severely limits the ability o fAfghan women to approach the courts. 2.55 Dominant drug industty. The opium economy, which is estimated to comprise more than a third o f total economic activity in Afghanistan, i s a direct source o f insecurity (drug-related conflict and violence, financing o f armed groups and warlords and commanders who provide protection). But even more important, the drug industry i s a major force worlung against state-building and good governance, and thereby has a major adverse effect on security indirectly as weL7 The drug industry can thrive in a situation where there i s a weak, corruptible state and security forces, no or ineffective rule o f law, and inadequate sanctions or negative incentives to discourage the whole range of actors - from farmers to drug traders to their sponsors inside and outside o f government as well as regional drug industry actors. The opiumeconomyi s analyzedfrom adevelopmentperspective inWard andByrd (2004) and World Bank(2005a, Chapter 7). 30 Therefore the drug industry comprises a powerful interest group diametrically opposed to the heart o f Afghanistan's reconstruction agenda and thereby promoting insecurity. 2.56 Remaining factional forces. Although as noted earlier the DDR process has been completed, there are numerous "illegal armed combatants" not included inDDR. The new program for Disbandment of Illegal Armed Groups (DIAG) is intended to address this problem; however, this is expected to take several years. Moreover, Afghanistan by tradition and even more during the conflict has had an armed population, and this i s not expected to change very quickly. Widespread ownership of small arms can be a source of insecurity as long as there i s not basic rule o f law and a minimum adequate level o f government functionality inthe security sector. Popular revulsion against warlords and commanders and their armed groups potentially can be a very powerful force, but it has not yet been harnessed. The Government's MDGreport makes strong reference to the urgent need for licensing o f arms.8 2.57 Violent and other crime, including corruption. Although it is difficult with available data to ascertain whether violent crime and other forms o f ordinary crime are more pervasive than in other low- income countries, anecdotal evidence and surveys suggest that crime i s an important issue for Afghans. As noted in Chapter 1, criminality can thrive in a conflict or post-conflict situation where legal, law enforcement, and social checks are weakened. The same is true o f corruption; rough indicators compiled by the World Bank (Figure 2.2), which have substantial margins o f error, suggest that Afghanistan is perceived to be among the countries most severely affected by corruption (see Volume I, 7 for a Chapter discussion of corruption issues in the PFM context). The recent Transparency International index o f perceptions o f corruption similarly indicates that Afghanistan has a serious cormption problem. 2.58 It should be noted that while corruption in government contracting, the tax regime, licenses and permits, etc. may be significant problems in Afghanistan, the enormous role o f the drug industry in promoting corruption, as well as to a lesser extent the role o f other parts of the informalhllicit economy, must be emphasized. The huge sums of money involved - in2004 gross value of opiate production in Afghanistan was roughly estimated at $2.8 billion, o f which $2.2 billion went to traders and processors and only $600 millionto farmers -virtually guarantee massive levels of corruption. Moreover, narcotics- related corruption most likely to a considerable extent takes the form o f "grand corruption" and capture o f parts o f the state apparatus, which is profoundly damaging to the state-building agenda as compared with smaller "petty corruption". Corruption contributes to personal andbusiness insecurity, as the "rules o f the game" are not clear and can change depending on who i s inparticular government positions, and also can take coercive forms like extortion o fpayments. 2.59 Landmines. Afghanistan inherited from its long period o f conflict one of the worst landmine problems in the world, as well as enormous amounts o f unexploded ordnance. While progress has been made in clearingmines inurban areas and some other parts o f the country, enormous numbersremain and continue to kill and maimpeople, harmlivestock, and render many areas unusable. 2.60 Finally, from a developmental perspective the impact o f insecurity on vulnerable groups like women continues to be very serious. Inthe absence o f effective mechanismsto investigate gender related violence, the vast majority o f such violations are not reported to the criminal justice system and almost Encouraging those who currently own weapons to apply for licenses, and identifying and punishing those who fail to comply with the new law, will not be easy. An effective licensing process will be facilitated by the electoral registration-more Afghans have a nationally recognized form o f identification than ever before. A community-level survey o f weapons ownership is also under consideration. However, additional support may be required to build the capacity of MoI to process applications, document the licensing process, and maintain records. The licensing process offers an opportunity to collect data on legal civilian-held weapons. It i s also proposed that licensing o f weapons and statistics on gun crime be monitored as indicators o f a reduction inthe misuse o fweapons. There are no accurate estimates on the number o f civilian-held weapons incirculation, but by all accounts they are very numerous (Afghanistan Government, 2005b) 31 none are subject to investigation or prosecution. The practice o f detention and prosecution o f women for zina crimes (adultery and consensual sex outside marriage) i s common in major cities, while in more conservative rural areas the family might deal with such a case by lulling the girl or woman involved. The predominance o f customary law in traditional jirgdshura decisions by and large works to the disadvantage of women, as customary law tends to grant fewer rights to them than Islamic law or civil law in matters relating to inheritance, property, and marriage, as well as in relation to crimes like so- called "honor killings" for which the perpetrators are generally not punished. Customary law's bias against women is also illustrated when for example murder cases are resolved by the jirga ordering that the alleged perpetrator provide the family o f the alleged victim with a young girl or girls, usually below the legal marriage age, to compensate for the alleged crime. The girl i s then forcibly married to a male member o f the victim's family. Such arrangements are also sometimes used to settle large opium-related debts. A sign of the impact o f conflict, violence, and lack o f access to justice on women i s the reported dramatic upsurge o f female suicides, often to escape unwanted arranged or forced marriages. Figure2.2: Afghanistan's PerceivedPerformancein Six Dimensionsof Governance Political Stability Governnent Effectiveness No data Regulatory Quality No data Rule of Lau Control o f Corruption e 25 58 75 Conparison betueen 2884, 2882, 1996 (top-botton order) Country's Percentile Rank 03-188) Note: the three bars represent,for each dimension of governance, Afghanistan Spercentile rankfor 2004, 2002, and 1996 (top-bottom order). The line represents the standard deviation of this measure (in other words the level of uncertainty of this ranking). Source: Kaufmann, Kraay, and Mastruzzi (2005). 32 CHAPTER 3. PUBLIC FINANCE MANAGEMENTAND AFGHANISTAN'S SECURITY SECTOR 3.1 As seen in Chapter 2, recent thinlung on both public finance management and security increasingly points to the full applicability o f the objectives, practices, institutions, and performance indicators o f the PFM system to the security sector. This chapter first discusses some implications for Afghanistan o f the PFM framework and its application to the security sector, and gaps and issues that arise in this regard. The chapter then analyzes the budget process as it applies to the security sector in Afghanistan along with associated issues. A. Implicationsfor PFMPrinciplesfor Afghanistan 3.2 Budget comprehensiveness. Withrespect to the comprehensiveness o fthe budget, the main issue in Afghanistan is not one of large amounts of government security sector expenditures occurring off- budget as found in many countries. Although there are some exceptions, the bulk o f security sector spending appears inthe Budget. Nor does the security sector appear to be engagedinlarge-scalerevenue- generating economic activities on an official basis.' However, the security sector, like most other parts o f Afghanistan's public sector, faces the issue that the bulk o f foreign-financed public expenditure (which accounted for some 92% o f total public expenditure in Afghanistan in 2004/05) i s executed directly by donors and by entities contractedby them, so it i s not going through the Govemment's Treasury channels and i s not following uniform accounting and reporting practices. More specifically, whereas 74% o f total budgetary expenditure in 2004/05 was in the "External Budget", the proportion o f External Budget spending i s even higher inthe security sector - 78% (Volume I, Table 5.1). This means that nearly four- fifths of total budgetary expenditures inthe security sector do not go through Government Treasury and Core Budget channels. 3.3 The Government has made a laudable effort to incorporate externally-executed spending in its overall budget formulation process through the Consultative Group mechanism, and information about donor-executed programs has been compiled and included in budget documents as part o f the External Budget. Donors are supposed to provide information on their plans, commitments, and disbursements, which can then be used particularly in the budget formulation process. This innovation has been very important in encouraging a holistic approach and enabling assessments and comparisons to be made on the basis o fa more-or-less complete picture o fwhat i sbeing spent onthe public sector inAfghanistan. 3.4 Notwithstanding the significance and usefulness o f this innovative approach, it works only imperfectly in general and especially so in the security sector. There are numerous problems that arise from the domination of the External Budget. While a serious effort is being made to achieve comprehensiveness to in terms o f information and budget documents, this is not possible for budget execution functions, where in the External Budget each donor uses its own procedures. Moreover, the information provided by donors on the External Budget understandably tends not to be fully consistent, either across donors or with the Core Budget. Frequently data on commitments and disbursements i s confused, and information on actual expenditures (i.e. by the executing agency) i s usually not available. Reporting appears to vary among donors in terms o f quality, classifications, and the frequency and regularity with whxh reporting i s done. These problems are aggravated inthe case o f the security sector As noted in Chapter 2, Government security entities are sometimes paid extra to provide security, for example for foreign- financed projects. Such revenues should be shown as incometo the agencies concerned (e.g. MoUpolice) but most probably are not. (They would berecordedas expendituresby the donors concernedinthe ExternalBudget, or by the Government inthe case ofexternally-financedprojectsinthe CoreBudgetthat make such payments.) 33 because funding from the donor countries often i s fragmented among various sources within donor country budgets, frequently outside their development budgets. 3.5 The lack of comprehensiveness insystems and controls for budget execution, as well as problems with information flows, means that application o f sound PFM principles in these areas across all expenditures becomes nearly impossible. For example, the crucial principle o f contestability cannot be applied. While these issues are common to the budget as a whole, they are even more serious in the security sector. 3.6 The "lead donor" system. Another feature which appears to have been in some respects problematic inthe security sector has been the "lead donor" system (see Chapter 2, Section C, and Table 2.1). This system, institutedinthe immediate post-conflictperiod and intended to harness the initiative o f different donors, may have been useful in promoting a division o f labor among donors and clarifying responsibilities for malung progress in different parts o f the security sector, with some accountability on the part of individual donors to take the initiative, mobilize funding, and move forward. Subsequent performance has varied, but clearly many o f the lead donors have taken their responsibilitiesseriously and have made strong efforts to pushforward a reform agenda in"their" sector. 3.7 However, the lead donor system may have made the difficult task of bringing Government leadership and coherence to security sector planning, decisionmakmg, funding, and implementation even more difficult. As it has worked inpractice, the lead donor approach seems to have reinforced, or at least not curbed, tendencies for different sectors to go their own way, pre-empting broader inter-ministerial coordination by the National Security Council and Cabinet, and impeding efforts through an overarching national security strategy to ensure coherence and consistency across sectors. A related point i s that the relative speed and success of progress in different sectors may have been at least to some extent influenced by the relative level o f financial and other resources, pro-activity, and quality technical support and advice brought to bear by different lead donors. This could introduce distortions, as the constellation of donor resources and energy may not coincide with national priorities. And finally, this approach may entrench different practices and standards inthe different security forces and oversight bodies, including inthe PFMsystem, makingharmonizationaround Government systems more difficult to achieve. 3.8 Budget predictability. To a very large extent, security sector policies and spending patterns in Afghanistan have not been determined by the capacity o f national resource to finance them. Targets for military, police, andjudiciary reformwere set outside o f a medium-term fiscal framework, and targets for counter-narcotics efforts have not been linked to the budgeting process. That said, with each successive budget, salary and non salary recurrent costs are increasingly being drawn into the Core Budget. A s a consequence, in future years overall funding levels and allotments will fall more squarely in the Core Budget controlled by the Afghan government. However, the decisions on force sizes and other aspects for virtually all security expenditure have been made outside o f the Core Budget framework and national budget formulation process. The Ministry o f Finance has played a very limited role to date, and the longer-term fiscal implications o f major spending decisions have not been informed by the overall fiscal framework. Furthermore, such decisions have often been made on the basis o f annual rather than medium-term external fundingcommitments, especially for the ANA. Inother cases, such as the creation of the Ministry o f Counter-Narcotics, the Government has solicited a commitment from the UK Government to provide external assistance for a period o f three years. However, predictability of most security sector fundingover the medium term is unknown. 3.9 Salary structure and payment mechanisms. Yet another issue which makes the application o f good PFMpractices inthe security sector more difficult is the variety o f pay scales, not closely related to each other (see Figure 3.1) and the payment o f salaries (not just top-ups, which occurs in other sectors as well) directly by donors to some o f the security forces. While uniformed services having somewhat 34 different salary scales from civil servants i s a common practice internationally, the setting o f pay scales between the security sector and civil service has not been coordinated, and pay scales within the security sector have tended to be set independently across the different security services, resulting in anomalies. Moreover, the ANA salary scale, apparently set without reference to pay levels elsewhere in the Government and seemingly highinrelation to labor market conditions, de facto became the leader and the target to which other parts o f the security sector aspired. As a result, the proposed new ANP salary scale has been set at a level similar to that o f the ANA, with minor differences at the higher levels. Moreover, certain other security forces (not shown inFigure 3.1) are paid directly by donors and may have different salary scales. Setting pay scales without conducting a disciplined analysis and involvingkey stakeholders (including MoF) not only sends the wrong signal in terms o f process, it also risks seriously undermining the fiscal sustainability o f security sector reforms, even if external bilateral support i s provided in the short run(see Chapter 4). Figure 3.1: Pay Scales ($/month) in Security and other Sectors 600 550 500 450 400 350 300 250 200 150 100 50 Civilian PRR Proposed Education Police ANT Army ANA Government New Pay Scale Lower and upper bounds of the scales. This includes the base salary and an estimate of cash allowances. 3.10 Staffing establishment controls. A very important aspect o f the PFM system i s maintaining effective control over staffing levels. By all indications staffing establishment controls in Afghanistan have been working reasonably well as they apply to non-secuhty sector government employees (see Volume I, Chapter 7), supported by external monitoring o f the ARTF. However, inthe security sector the wide variety o f salary, employment, and payment arrangements makes effective exercise o f control over staffing levels more difficult. The old army (Afghan Militia Forces, AMF -now defunct) appears to have been riddled with nonexistent or "ghost" soldiers. Differing data on actual numbers of employees in various parts o f the security sector give grounds for concern. There appears to be little correspondence between the agreed staffing establishment and the actual numbers o f regular (karmund) and temporary (ugir) staff inthe Ministry o f Interior and the Ministry o f Justice inparticular. Moreover, there i s often no agreement on the actual numbers o f staff. For instance, based on different staffing numbers provided by MoF, MoI, and LOTFA, M o Ishows a wide range o fpossible staffing structures. The Justice Ministry would appear to be claiming nearly 70% more staff than currently agreed by the Office o f Administrative Affairs. A notable example o f head-count volatility i s the (now-defunct) AMF, where estimates o f the total numbers of militia to be demobilized varied widely and turned out to be much lower than originally estimated. On the other hand, the National Directorate o f Security several years ago undertook a serious examination o f its staffing level and removed several tens o fthousands o f ghost employees. 35 3.11 Capacity. The lack o f capacity in security sector management and oversight institutionsremains a serious problem. As seen in Chapter 2, implementation o f priority administrative reforms in security sector institutions has been slow and continues to undermine capacity for civil authorities' management and oversight o f the security sector, whose crucial importance was emphasized in Chapter 1, Section D. Inthe MinistryofInterior, for example, bilateralfundingtobuildpolicingcapacity hasnotbeenmatched by investments to undertake administrative reforms in the Ministry itself. Recent efforts to bring MoI under the Priority Reform and Restructuring (PRR) program have yet to be concretized. Although there have been important changes in the Ministry o f Defense in support o f reform, primarily in terms o f personnel, efforts appear not to have been made so far to bring MODunder the PRR program. The Afghanistan Stabilization Program (ASP) has so far failed to bring MoI structures formally under the PRRprocess andhas instead focused largely onphysical infrastructure development at the district level. 3.12 Inmany cases investment priorities have been poorly sequenced due to limited capacity inline ministries and for overall security sector coordination. Inmost cases, the Public Investment Programs (PIPS)that have been prepared by security sector ministries are o f poor quality, and in some cases PIPS have not been prepared. Furthermore, only MOD,MoI, MoJ, and the Supreme Court are actually coordinated within the PIP framework. Other institutions, such as the Attorney General's Office and MoBTA, remain disconnected from the main External Budget framework. B.The Security Sector andthe Budget Process BUDGETFORMULATION 3.13 Ideally, budget formulation i s framed by the following process: (i) setting wider security policy goals, (ii) setting annual policy priorities, (iii) conducting integrated planning both within and between different security institutions, (iv) translating annual planning into public investment programs (which should integrate operating and capital spending), and (v) formally setting expenditure levels and allotments based on top-down budget constraints and aggregate discipline. 3.14 Setting broad national security goals: As discussed earlier, during the early stages from late 2001 onward, there was an absence o f agreement among key Afghan stakeholders on the country's broad national security goals. Moreover, Afghanistan's international security partners tended to play a driving role in determining the direction o f Afghanistan's security investments. While the Government has strengthened its capacity to develop and coordinate security policy through national instruments and institutions, this takes time and requires not only understanding but pro-active support on the part o f donors, who will need to be ready to progressively open up space for the Government to take national leadership with respect to the security sector. 3.15 Medium-term budget planning and costing: Whereas the Government does have a three-year rolling budget including large parts o f the security sector, this appears to be to a large extent mechanical and not linked to resource availability. While these issues also arise in the civilian budget, the size and momentum o f expansion targets in the security sector (see Chapter 4) make these weaknesses more problematic. 3.16 Inmany cases, spending forecasts for the development budget do not appear to be linkedto long- term planning. For example, under the 2004/05 development budget MoI's development budget spending was forecast at zero for 2007/08 (the third year o f the rolling three-year PIP horizon). This was not because there was expected to be no development expenditures on MoI in 2007/08, but rather due to the lack o f medium-tern budget planning. 36 3.17 Inthis regard, a major area of concern relates to the longer-term sustaining costs of the ANA, ANP, and the counter-narcotics police forces, which have yet to be fully costed. Informal consultations with concerned donors suggest that at full strength, the sustaining costs for the ANA alone may be inthe range o f U S $300 million-plus per year, which clearly cannot be covered by Government for some time to come given current fiscal conditions. The lack of long-term planning and related costing thus runs the risk o f undermining the fiscal sustainability o f the current security structure and effective allocation o f budgetary resources, on the one hand, and the ability o f the security services to maximize the security they will be able to provide, on the other hand. 3.18 Setting annualpolicypriorities: Inthe absence o f clear strategies and policies, it i s very difficult to develop a formal process for setting annual budget allocations for the security sector as a whole or its key parts. Budget formulation tends to be conducted at the end o f the fiscal year inreaction to the budget circular from the MoF, rather than beginning mid-year or soon thereafter and reflecting the policy priorities approved by the NSC. Budget formulation i s not addressed as an integrated part o f the whole policy makmg-planning -execution process. Instead, each ministry, and often the lead external agency, sets targets based on assumed levels o f external assistance. Verifiable performance indicators against which performance and cost efficiency can be measured are generally not developed (see Chapter 6 for some suggestions). 3.19 For the most part, however, the annual budget formulation process for Core Budget operating expenditures works reasonably well, albeit without much strategic content. Proposals are developed in line with the M o F budget circular, and staffing establishments and organizational charts are formally agreed by the Office o f Administrative Affairs (OAA). 3.20 Prior to 2005/06, annual priorities laid out in the sectoral PIPSfor the development budget were poorly documented. For example, in 2004105 the PIP for MODwas two pages long and incomplete despite the existence of an agreed budget. In 2005/06, budget targets for ANA and ANT set by MoD/OMC-A and MoUGermany respectively were well thought through, demonstrating that a learning process i s taking place. For DDR, annual performance targets were coherently set within the Performance Measurement Framework by the Afghanistan New Beginnings Program (ANBP - now completed), althoughthe nature o fthe process required flexibility inexecution. 3.21 The Ministry o f Martyrs and Disabled, which has direct relations with tribal militia and border security, does not formally set policy targets. Furthermore, its budget i s determined through the Livelihoods and Social Protection (LSP) program. This i s a policy ambiguity created because the Ministryalso works withthe nomadicpopulation. 3.22 In2004/05, counter narcotics was treated as a cross-cutting theme (along with gender), and a separate memorandum listing was produced for the External Budget, based on the boxes checked for counter narcotics inProject Summary Sheets. The resulting list had a mix o f alternative livelihoods and law enforcement projects. However, some projects which are directly for counter narcotics purposes are left out, and the connection o f some projects listed as counter narcotics to this agenda is not clear (e.g. National ID cards, Support to Paralympics). Most counter narcotics resources flow through the External Budget. MCN i s supposed to play a coordinating role in the budget preparation and mid-term review process. However, MCN's predecessor (the Counter-Narcotics Department, CND) was obliged to refuse to participate in the 2004/05 mid-term review process as they lacked the necessary capacity. It seems desirable for the Government and donors to work together to develop a comprehensive list o f resources and activities for counter-narcotics. It will be important for MCN to develop the capacity to play a significant role inbudget formulation and follow-up. 37 3.23 Budgetplanning: Currently, annual planning for the development budget tends to focus around the funding cycles o f the lead donor agencies and not around.the Government's fiscal year. That said, planning is conducted as a three-year rolling process for defense, law and order, and judicial reform, and not as a one-off planning exercise. The quality o f planning i s directly related to the level o f resources allocated, and as a result planning capacities around the MoD/ANA are probably the most advanced, thanks largely to remarkable efforts by OMC-A and the U S Government. Planning for the A N P L a w and Order program has not fed coherently into the budget process. For counter narcotics programs, performance targets interms o f area to be eradicated and legal reforms have been clearly set in agreement with the relevant national authorities. However, as resources for counter-narcotics have not been consolidated, setting planning targets with an unknown resourcing framework has been difficult. The Counter-Narcotics Trust Fund i s intended to enhance the cohesion and consistency o f counter-narcotics activities by providing an element o f coordinated financing (which has worked well in the case of the civilian sector through ARTF, but not so well inthe case o fpolice through LOTFA- see below). 3.24 There have also been considerable improvements inplanning support for the Ministry o f Justice, the Supreme Court, and the Attorney General's Office. However, planning capacities in the Supreme Court and Attorney General's Office remain very weak. 3.25 Preparation of PublicInvestment Programs: Inall cases PIPs do little more than reflect the sum total of development budget funding currently committed by donors. While PIPs allow M o F to compile an aggregate o fnational budget spending, they tend to be highly projectizedand frequently represent little more than the sum o f the individual projects. The ANA and DDR appear to be the only exceptions to this rule, as funding i s consolidated. In areas like M o I reform or counter narcotics activities such as alternative livelihoods, support i s so highly projectieed that even documenting the existing projects i s a complicatedtask. Furthermore, once completed by the relevant budget entity, PIPStend to be shelvedand not used as a reference document for monitoring progress. 3.26 Setting budget expenditure levels and allotments: A major constraint to effective budget formulation i s the lack of guidance from M o F on the fiscal envelope within which prioritization and fiscal discipline are delivered. Without information on how much expenditure is allowed and without a formal appropriation process for the External Budget, sectoral investments remain supply-driven and not demand-dnven. This has direct relevance to reform o f the ANA and ANP in particular, as pay and grading structures are being developed without prior reference to the Government's medium-term fiscal envelope, thus making them potentially unsustainable. A s with other sectors, MoF does not communicate a budget ceiling to sector Ministries prior to setting the investment program, against which prioritization could take place. Combined with projectized funding, this often means that some programs are over- funded, whereas others of equal ifnot higher importance remain unfunded year after year. For the operating budget, a new object code classification was introduced for the 2005/06 operating budget, against which the Government can clearly apportion planned expenditures by (i) wages and salaries, (ii) goods and services, and (iii) acquisition o f assets. Each classification has detailed budget line items allowing basic expenditure tracking to take place. 3.27 Insummary, themainweaknesses withregardtobudgetformulationinthe security sector include (i) dominantshareofresourcesflowingthroughtheExternalBudgetandhencenotsubjecttothefull the rigor o f the national budget formulation process; (ii) lack hitherto o f a clear and agreed national the security strategy and sector strategies for defense, law and order or judicial reform; (iii) absence o f a medium-term framework for the operating and development budgets; (iv) lack o f annual budget ceilings providing top-down constraints in support o f prioritization; (v) limited scrutiny by the budget committee; and (vi) absence o f National Assembly consideration or scrutiny (which i s hoped to be rectified for the 2006/07 budget with the formation o f Parliament). 38 BUDGET FUNDING ARRANGEMENTS 3.28 Funding sources and mechanisms for the security sector are briefly summarized in Table 3.1. Finance for the Government's operating budget for the security sector i s primarily drawn from domestic revenues and the L a w and Order Trust Fund for Afghanistan (LOTFA). In contrast, operating costs for the ANA are covered directly by the U S through the External Budget, and the Government's development budget i s largely drawn from assistance provided directly by bilateral agencies. As a result, the national budget is merely capturing donor commitments rather than setting funding levels and allotments and determining the sources o f finance. It i s difficult to ascertain the precise shares o f funding for the security sector fromdifferent sources, but it i s clear that the USA accounts for the lion's share, perhaps about two- thirds, with domestic revenues, LOTFA, the UK,and Germany being other sizable contributors. Table 3.1: SecuritySector FundingArrangements Security Body Role in Setting Goals ANA SubstantialANA operatingcosts are now met by the Government from domestic revenue. However, it is not clear how muchANA wage costs are paid by the Government. At least some wage costs, mostnon-wagerecurrent costs other than food, andall capital investmentcosts are funded directly by the US from bothregularandsupplementalappropriations. ANP The Government incurs expenses for police salaries and capital inputs (payroll systems, police buildingsetc.), all of which are reimbursedthroughLOTFA, based on verification. The main donors of LOTFA are the EC, US, Norway, Switzerland, and Finland. Other costs for MoI, including the costs of civilian staffing and other operations and maintenance costs, are covered by domestic revenues. The large MoI/ANP development budget is covered by many donors, among which the EC, Germany, UK and the US are the main ones. Once the CNTF is established, multilateral donor support for counter-narcotics activities will be channelled through this route (althoughthe CNTF is expectedto accountfor arelativelysmall share of total counter-narcotics spending). Justice Operating costs (salaries and modest non-salary recurrent costs) are financed directly by the Government,with some costs reimbursed via ARTF. Developmentbudget funding, largely from the Italian Government, is channelled as discrete projects through various UN agency agencies (UNIFEM, UNDP, UNODC), andthere is also EC funding. DDR/DIAG DDR has beenfunded by Japan, Canada, UnitedKingdom, andthe United States, with Japan as lead donor. The Afghanistan New Beginnings Program (ANBP) has been implemented through UNAMARJNDP. Fundingrequirementsandarrangementsfor DIAGare not yet clear. CounterNarcotics Operatingcosts for the new Ministry of Counter Narcotics are largely covered by Government. The lead donor (UK) covers additional costs relatedto salary top ups and field operations. The bulk of funding, however, is provided by the US with more focused CN strategic funding provided by the UK. Alternativelivelihoodsprojectsare fundedbyanumberofdonors, includingtheEC. 3.29 LOTFAhas played an important role inreimbursingpolice salaries and other inputsina situation when MoI's PFM capacities were very weak (see Box 3.1). The trust fund i s managed by UNDP, channeling financial resources from the EC, US, Norway, Switzerland, and Finland in a consolidated manner. The key strength o f LOTFA i s that payments for the ANP were made possible prior to the establishment o f appropriate internal M o I fiduciary mechanisms, due to the involvement o f UNDP in payment verification centrally and at the provincial level. The funds are channeled to the Ministry o f Interior through the Ministry o f Finance. LOTFA issues advances and is reimbursed against certified financial reports received fiom the MoF. All ANP expenditures are incurred directly by MoI but reimbursed through LOTFA. The fund now covers reimbursements inboth Kabul and all the provinces. Provincial payments are made through joint verification missions conducted by MoI/MoF/UNDP. Switzerland and the EC are currently supporting the computerization of the payroll system. However, 39 despite considerable efforts, a visit to LaghmanProvince described inthe 2004/05 LOTFA Annual Report identified a number o f key constraints to enhancing public expenditure management (Box 3.2). Box 3.1: Overview of the Law and Order Trust Fundfor Afghanistan(LOTFA) During Phase I:From November 2002 through March 2004, LOTFA reimbursed MoF for salary payments to police in the amount o f US$44.4 million. By the last quarter o f 2003, LOTFA was covering police remuneration in each of the 32 provinces. LOTFA has placed two public finance management officers in MoI to support the preparation o f timely financial reports, regularization o f the payroll system, and training o f the Ministry's financial staff. Activities were also initiated with EC and Swiss Development Cooperation (SDC) support for computerization o f the payroll system at MoI. Through contributions to the National Development Budget and other funding windows, LOTFA also procured and made available to the police 1,100 vehicles with communications equipment and spare parts valued at US$11.45 million. During Phase I,funding limitations prevented LOTFA from moving substantially beyond its first and second priorities: police salaries and equipment. During Phase 11: From April 2004 through March 2005 (covering Afghan Fiscal Year 1383), LOTFA was expected to: ensure payment o f police salaries in all provinces and improve the payroll system at MoI; provide substantially more police equipment; rehabilitate and rebuild priority police facilities; and support M o I in undertaking institutional reforms (including activities to build financial and project management capacity). In response to donor concerns, and to ensure proper accountability and transparency, special emphasis was to be placed on monitoring o f expenditures and the effectiveness o f payments to the police force. 3.30 LOTFA successfully provided salaries and other allowances to police forces in all 34 provinces for the whole o f 2004/05. During that year a total of 53,400 police - including 17,705 officers, 4,843 sergeants and 30,852 patrolmen-received a total o f US$54 million (US$70 on average per police officer per month). Payments took place through the regular Governmental payroll system, and funds were channeled through the Treasury department o f MoF. As no nominal roll o f police officers existed, the payment process has been complicated. A computerized payroll system is being developed to (i) interface to capture data, (ii) collect personal records, (iii) budget, revenue, and salary expenditure details, include and (iv) record information on attendances, salaries and allowances. On the investment side, procurement of vehicles, rehabilitation o f police and checkpoint facilities and gender sensitization o f the police force are funded through the Treasury. Box 3.2: Findings of aLOTFA MonitoringMission A monitoringmission to Laghman, conducted by a small MoI-UNDP team, identifiedthe following factors that affect P F M practices: Payments were on time according to the formats; US$70 was disbursed per person for national police patrolmen. A civilian has been appointed as head of police in Qarghaie district. Staff files andrecords were well managed. Based on staff reports, the head o f police attends the office irregularly. Organizational and functional charts were not available. Vehicle registration booklets were not properly maintained. Facilities for operational activities in some police units and sections are insufficient. Office space was visibly insufficient. Employment for the head o fpolice and district commissioner are on a quota basis. There is a lack o f coordination between M o I and the local PRT on provision o f equipment, vehicles, and other supplies. Communication with M o I central office is not maintained. A very small amount o ffunds is allocated for fuel, making it difficult to maintain security. Source: LOTFA 1383Annual Report 3.3 1 However, despite significant gains, LOTFA suffered from serious start-up problems initially (see Box 3.3 related to CNPA) and faces a number o f constraints that need to be addressed: LOTFA is under-funded, and there are continuing uncertainties regarding hture commitments from major donors, which have yet to materialize. 0 Donors complain that there i s little transparency with regard to the political decision making process onpay and gradingreform, compared with progress made on other technical issues. 40 There appears to be little awareness regarding the fiscal constraints facing the ANP. There is a general feeling within Government that the international community is somehow obliged to continue meeting security sector budget requirements (including ANP financing) without conditionality. There are some continuing concerns about payroll integrity. Although UNDP management has improved, lack o f technical assistance is undermining the costing and planning o f investments. External scrutiny inthe form o f audited accounts has been subject to long delays. And finally, there is no exit strategy for the trust fund, and the current fiscal situation will not allow the Government to absorb police operating costs within the core recurrent budget for many years to come, not least because increases inANA funding are progressively being drawn into the Core Budget. 3.32 A Counter Narcotics Trust Fundhas been established, also under the management of UNDP, to provide a coordinated funding source for counter-narcotics activities. Its likely modest size means that the CNTF will probably be used for gap-filling and meeting urgent Government priorities in the counter- narcotics field that are not being covered by donors through the External Budget. Box 3.3: Fundingfor Counter NarcoticsLaw Enforcement Since the Counter-NarcoticsPolice of Afghanistan (CNPA) was founded, its salary and operating costs have been financed through LOTFA. In 2003/04, slow payments from LOTFA were a major handicapto the CNPA. Despite generous donations of equipment from the donors, there was virtually no operating budget in late 2003. There was a brand new headquarters in Kabul, but computers initially could not be used as there was no electricity. Inthe field there were vehicles but no fuel. As there were delays in funds coming from LOTFA for salaries, the Ministry of the Interior paidthe staffsome of their wages from MoI's own resources. But evenwhen salariesarrived, they were low under the existing Governmentsalary scales. The payroll of the entire 700-strong CID Kabulbranch of in October 2003 was $16,000, or $23 amonthper person. The situation is expectedto change as a new salary scale has been proposed for the ANP, which presumablyalso would applyto CNPA. In addition to LOTFA funding of CNPA, the counter narcoticssector has a number of special financing arrangements. First, the previous CND's entire operatingcosts were financed on a "project basis" by the UKthroughaprojectincludedinthe ExternalBudget. (The new MinistryofCounterNarcoticsis supposed to be part of the Core Budget; however it can be expected that integration will take time.) In 2005, these expenses will be financed from the UK-financedInstitutional Development Program. Second, the UK has created an ad hoc arrangement for financing of CNPA top-ups for up to six months, pending the effectivenessof the InstitutionalDevelopmentContract. Third, donor financingis supportingall operations ofthe various specialized counter-narcotics forces. 41 CHAPTER 4. TOWARDA FISCALLYSUSTAINABLE SECURITY SECTOR 4.1 The fiscal sustainability o f the security sector i s a central issue for public finance management in Afghanistan (see Volume I,Chapter 3). This chapter first reviews recent trends in security sector expenditures and highlightsthe key factors behind their rapid growth. It then discusses drivers o f future security spending including programmed expansion o f the sector. These dnvers are expected to result in an unmanageable level o f expenditures over the medium term, and some very rough projections of the likely order of magnitude of overall security sector expenditures are put forward. A process for assessing the fiscal sustainability of the security sector and making any necessary adjustments (including possible medium-term external financing for parts o fthe security sector) i s outlined at the end o f the chapter. A. Recent Trends inSecurity Sector Expenditures 4.2 Security sector expenditures have been growing very rapidly during the past three years, and growth i s projected to continue in the current financial year (see Table 4.1). Whether in absolute terms (an estimated $1.3 billion in 2004105 not including most counter-narcotics expenditures) or as a share o f GDP (23%), the size of the security sector i s daunting. As can be seen in Figure 4.1, security sector spending reached a very high share o f GDP by 2003/04 and has remained high, albeit with External Budget expenditures on security declining modestly as a share o f GDP (while continuing to grow robustly inabsolute terms). Ifmaintained, andparticularly with an expected shift from the ExternalBudget to the Core Budget, such a highlevel o f security sector expenditures would carry serious fiscal risks and would threaten the sustainability of the security sector. Table 4.1: Summary of Total Security Sector Expenditures2003/04 -2005/06 (US$ million) 2003104 2004105 2005106 Estimate Estimate Budget Total CoreandExternalSecurityExpenditures 1,105 1,328 1,615 Core Security Expenditures 292 384 - Operating 235 233 275 303 - Development 2 17 81 ExternalSecurity Expenditures 870 1,036 1,23 1 TotalCoreand ExternalBudget Expenditures 3,847 4,213 4,862 Yoof Security inTotal Expenditures 29% 32% 33% TotalDomesticRevenues 208 269 333 %of Securityto DomesticRevenues 533% 494% 485% 42 Figure 4.1: Security Spending as a Share of GDP L" 20 15 10 5 2003104 2004105 2005106 Core Budget External Budget Source: MoF budget documents andfiscal reports, staffestimates. 4.3 The main dnver o f security sector expenditures has been the Afghan National Army (ANA), which after a start-up phase in 2002/03 has been absorbing large amounts o f funds on the order o f $800 million per year during 2003/04 - 2005/06 (see Table 4.2). Equally strilung more recently has been the growth of police expenditures, which are targeted to quintuple in two years to nearly $600 million (budgeted) in 2005/06. Other components, while much smaller in terms of their impact on aggregate security sector spending, have also seen very large increases, notably mine action and (starting from a tiny base) and justice. Not surprisingly in Afghanistan's difficult post-conflict circumstances, the overall pattern has been one o f considerable fluctuations and at times large, order-of-magnitude increases in expenditures on different components, amid continuing rapid growth o f overall security spending. Table 4.2: Security Sector Expenditures by Program (US $ million) 2003104 2004105 2005106 Estimate Estimate Budget Total 1,105 1,328 1,615 AfghanNationalArmy 797 788 830 National Police and Law Enforcement 209 340 597 Justice 16 45 63 MineAction 63 96 77 DDR 21 59 48 B.FutureExpenditureDrivers andProgrammedExpansion 4.4 The rapid growth o f security sector expenditures seen in recent years has not yet reached a plateau, and there are a number o f drivers which can be expected to generate further increases in coming years. The investment component o f security sector spending (construction and equipment) may decline after another couple o fyears, but other dnvers can be expected to further increaserecurrent expenditures. 43 4.5 Continuing investments and equipment purchases: The large one-time investments in setting up ANA base facilities and in equipping the ANA are expected to taper off over the next several years. However, substantial investments inthe ANP have been programmed for 2005/06 and can be expected to continue for a time. Ithas beenreported that the U S i s expectedto commit some $700-800million for the ANP in the coming two years, much o f which would be used for infrastructure development and equipment purchases. Moreover, the costs associated with maintaining the ANA and ANP "capital stock", as well as similar investments in the counter-narcotics field, are not precisely known but can be expected to be substantial. This underlines the importance o f instituting a multi-year planning and budgeting cycle and o f fully costing equipment acquisitions and other capital investments. It also should be kept inmindthat forecasts made by security-related ministries o f investment expenditures over several years are likely to understate the requirements in outer years, as plans and investment programs have not yet been formulated. Moreover, since virtually all security investment costs are currently borne by the External Budget and external financing for the security sector can be expected to decline by at least as much as these investments fall (if not more), there is no net benefit to the Afghan budget from this expected future trend. 4.6 Continuingprogrammed expansion of securityforces: As seen in Chapter 1, target force levels were set based on various considerations (including the expected need to accommodate existing combatants in the new order), without the benefit o f a carefully thought out or agreed threat assessment. Continuing expansion o f employment in the security sector will be required to reach these target force levels, most notably in the case o f the ANA (targeted level 70,000, as compared with around 26,000 combat troops at present), but there will also be additional recruitment into the police. The annual fiscal impact o f increasing the ANA by another 40,000 or so to reach the 70,000 target would be more than $100 million for salaries alone. The total salary bill for the ANA at full strength can be estimated atjust under $200 million annually. All in all, if expansion continues, ANA salaries will be a major dnver o f security sector expenditures (and indeedo f total budgetary expenditures) incoming years. 4.7 Although not as dramatic as inthe case o f the ANA, other security forces also are programmedto expand, including the ANP. Moreover, there has been a proposal to double the number o f border guards from the targeted level o f 12,000 to 24,000. 4.8 High salary scales, being applied to an increasing proportion of security sector staffi Salary and wage levels in the recent past have varied considerably among the different security institutions (see also Figure 3.1). Inthe ANA they range from a l o w o f US$70 to a maximum o f US$850 per month. In contrast, M o I staff (including police) have received between US$30 and US$122 per month. A new pay and grading structure has been proposed for the ANP that will bring their pay structure on par with (and insome grades above) that o f the ANA. The new Counter-narcotics Ministryalso appears to be getting higher salaries andor top-ups. The N D S similarly hopes to at least reach parity with ANA. What i s more, setting new pay scales has implications for longer-term pension liabilities, although this issue was not addressed when the new ANA and ANP salary structures were set." Clearly, security sector personnel require a living wage. At the same time, a wage rate that is driven by individual Ministry requests and funded by donors through the External Budget, without a Government-wide agreement on an overall wage structure that reflects Afghanistan's fiscal realities, runs a serious risk o f not being sustainable should donor support be reduced before adequate local revenue comes on stream. lo A recent Supreme Court ruling provided for defense retirees' pensions to be adjusted retroactively inline with the current salary structure, which would have an estimated one-time extra cost of $45 million, plus higher militarypension costs o f $15 million annually. While the status o f implementation o f this rulingis not yet clear, it illustrates the major fiscal liabilities that can be created by decisions on salary structure and pensions inthe security sector. 44 4.9 It is not clear whether and to what extent there will be further wage increases in the security sector, although civil servants in the Government management bodies (i.e. parent ministries) will be included in any general civil service salary increases. Even if further increases in salary scales, particularly for the ANA and ANP, are avoided, these have been set at relatively high total compensation levels inrelation to local conditions and the rest o f government employees. In addition to the expansion o f the ANA referred to above, the progressive shift o fmost o f the police force from its previous relatively low salary structure to the new ANP pay levels (see Table 2.1) will generate substantial increased salary costs. Even ifthere are no further salary increases duringthe coming several years, rising wage costs are already built in due to the shift from a low pay structure to a much higher one. A progressively larger proportion o f total security sector employees will be earning higher salaries - along the lines o f ANA and ANP -and this is likely to exert upwardpressure on civil service salaries more generally. 4.10 Correspondinggrowth of non-salary expendituresin the security sector: Security forces need a range o f complementary non-salary inputs in order to function effectively and remain at full readiness. Examples include fuel for mobility, uniforms, food, ammunition, etc. And there are also very substantial maintenance costs for military and other security forces' equipment. Inthe case o f the ANA, the non- salary recurrent costs o f an ANA o f 70,000 have been very roughly and conservatively estimated at around $120 million per year. Combined with salary costs this would be a major fiscal burden inview o f Afghanistan's low domestic revenues (only $269 million in 2004/05 and originally budgeted at $333 million in2005/06). 4.1 1 Counter-narcotics expenditures: While uncertain inmagnitude and possibly fluctuating widely, spending on counter-narcotics programs, including both law enforcement and alternative livelihoods, can be expected to be high. Information i s incomplete, and there i s a risk o f double-counting o f alternative livelihoods spendingwith rural development programs. Moreover, the longer-term fiscal and expenditure management implications o f the new institutional arrangements for counter-narcotics (oversight and coordination by MCN) are not yet clear. Pay structures are not yet fully clear but appear to be high at least in some parts o f the counter-narcotics apparatus, with direct payments o f salariedtop-ups by donors playing a prominent role. Spending on counter-narcotics (both ordinary and development budgets) may increase considerably, and there i s a concern that the costs o f counter-narcotics may spiral out of control, or at least overtake spending on the ANA and ANP. It i s very difficult to capture all information regarding counter narcotics funding because (i)many expenditures are not budgeted (and much is provided "in lund") and (ii) little information i s available more than one year ahead (largely because the program i s evolving fast, and i s supported by donors who make only annual commitments). Nevertheless, it seems likely that spending on counter-narcotics activities in coming years will be inthe $500-1,000 million range annually. 4.12 These various drivers o f security sector expenditures and very rough estimates o f their likely future impact (on an annual basis) are outlined inTable 4.3. Taking into account the substantial double- counting that most likely exists between counter-narcotics spending and other categories inside and outside the security budget, it nevertheless seems likely that the overall costs o f Afghanistan's security sector will be inthe range o fUS$900-1,000 million per year over the mediumterm, ifnot more. 4.13 Inconclusion, it is frequently the case ininternational experience that when, often as part of a security sector reform initiative, salaries and wages increase inthe security sector and modern equipment is procured, staffing levels in the security services decline. Afghanistan i s in quite a different position. Due to its problematic historical legacy and the urgent need to buildup security forces, the country faces the prospect for at least some years to come o f simultaneously rising average salaries, staffing levels, non- salary expenditures, and equipmentlmaintenance costs. This, combined with the fragmented character of decisions on key security parameters, will place enormous fiscal pressures on Afghanistan's budget. 45 Counter-narcotics spending i s highly uncertain but i s likely to further augment total security expenditures, with associated issues for fiscal sustainability. Table 4.3: Likely Drivers of Security Sector Expenditures /a Incrementalcosts over current levelsare roughly estimatedat over $100million. /b Veryroughestimate. Total sustainingcostsoftheANA, includingbothwage andnon-wageexpenditures,havebeenroughly estimatedat the rangeof $300-400millionper year. /c Rough estimateincludingtraining costs. /d Very rough estimatebased on assetscreatedby current very high investments. /e Very roughestimate; this is likely to involve double-countingwithbothother security sector categoriesand civilian developmentexpenditures ( e g onruraldevelopment). Source: Afghanistan Pay and Employment Model: staff estimates. 4.14 Moreover, within total security sector spending, important shifts between the External and Core budgets can be expected, with major fiscal implications which could be unsustainable if not well- managed. Currently very large amounts o f recurrent expenditures in the security sector - on wages and purchases o f good and services - are paid directly by donors through the External Budget. Available data do not allow much precision, but it would appear that very roughly on the order o f 60% o f security sector spending inthe ExternalBudget consists o frecurrent expenditures (see Volume I, Figure 6.2). 4.15 In this regard, there are already pressures from key donors for the Core Budget to pay progressively more o f the security sector wage bill over time. If this shift occurs rapidly, it will put enormous pressure on the Government's domestic revenues. Similar pressures are likely for non-wage recurrent expenditures, and already the cost o f food for ANA troops has been shifted from the External to the Core Budget. Just adding up the recurrent drivers listed above and excluding counter-narcotics, the total well exceeds current and forecast domestic revenues. C. ProjectedCosts, FiscalBurden, andLackof Sustainability 4.16 Projecting security sector expenditures forward over time is a challenge given data limitations and poor data quality. However, targeted force sizes for important components o f the security sector (e.g. the ANA and ANP) can be used to generate an idea o f some o f the key future cost components (building on the analysis o f expenditure drivers inthe previous section). Given the very large magnitudes and past and continuing expenditure growth in the security sector, even very rough projections can be useful in providing an overall picture and highlightingthe sustainability issue. 4.17 Based on the current structure o f security spending and national budget rolling forecasts, it can be roughly estimated that total security spending including counter-narcotics will absorb 4550% of total public expenditures, an even higher ratio than inthe recent past. There i s thus a major risk that security spending will crowd out other spending aimed at supporting economic growth and poverty reduction. The expected further increase inthe security budget largely results from the projected growth inspending on counter-narcotics and from increases in operating expenditures, both wage and non wage recurrent costs, primarily for ANA and ANP. 46 4.21 This issue is o f great urgency, since decisions already made or currently being made, and expenditures now being incurred or planned, could have potentially profound implications for the fiscal sustainability o f the security sector. Thus actions are needed, sooner rather than later, to: (i) obtain full information and assess more reliably the magnitude o f the fiscal sustainability problem in Afghanistan's security sector; (ii) bring together all the key stakeholders - across the Government and outside o f Government, including key donors - to discuss and reach greater consensus on this issue; and (iii) take actions to ameliorate the fiscal sustainability problem, ensure that it does not get worse, and put inplace arrangements to ensure that it i s properly managed and does not prevent Afghanistan from progressing toward overall fiscal sustainability (Volume I,Chapter 3). A possible approach along these lines is put forward for consideration. 4.22 Quickly enhance the information base on security sector expenditures, especially with respect to the cost implications o f mandatedplanned expansion o f security forces, and also of associated wage structures and non-wage recurrent expenditure requirements. This will require a short but concentrated effort on the part of the Government with support from donors and other stakeholders. The information gathering work can build on a similar exercise conducted in late 2004 and early 2005 as part o f the PFM Review, during which very good cooperation was provided to the PFM Review team by Government agencies and donors. The information gathered earlier needs to be updated and corrected where necessary, to be made more comprehensive, and to reflect recent decisions on basic features o f the security sector (e.g. on the size of the border police force, size o f ANP, counter-narcotics forces, etc.). 4.23 Review thefiscal implications of ongoing security sector expansion and reforms: Conduct an integrated security sector-wide fiscal review to forecast the fiscal implications o f currently programmed security sector expansion and reforms over the next few years as well as over the medium-term. This exercise should be led by the Government (the NSC and M o F presumably would play very important coordination and oversight roles), with technical inputs and information provided by external partners. The security sector fiscal review would logically be done under the overall umbrella of the MTFF into which it would feed. The sustainability o f the plannedexpenditure increases needs to be examined within the context o f the overall rolling fiscal plan, the growth in domestic revenues, and long-term commitments from the donors to fund recurrent costs. A number o f key issues need to be addressed in this regard: (i)salary levels and structure; (ii)staffing levels; (iii) non-salary recurrent costs (including unitcosts); (iv) integrationo frecurrent andcapitalcosts; and (v) development spending forecasts. 4.24 Revisit strategic objectives and trade-offs in the light of fiscal constraints. This would ideally be done as part o f the overall exercise to prepare/finalize the national security strategy, which itself should be consistent with and supportive to the national development strategy (ANDs). 4.25 Take actions to improve the situation. Options that the Government could consider include (i) avoiding any further wage increases inthe security sector (it would be difficult to roll back increases that have already been implemented); (ii) reconsidering programmed force sizes, notably for the ANA, inthe light of the current situation and fiscal constraints; (iii) exploring possibilities for containing non-salary recurrent costs, but in a manner that does not adversely affect readiness and operational capabilities; and (iv) similarly reviewing construction plans and equipment costs. None of these options would be easy, but as seen above, the option of doing nothing carries grave fiscal risks and would most likely generate significant threats to Afghanistan's development. 4.26 Explore with donorsprospectsfor predictable externalfunding for the security sector over the medium-term, in line with agreed objectives for the security sector and with a clear road-map for gradually phasing out such funding over a number of years as Afghanistan progresses toward a sustainable fiscal position. 48 4.27 Ensure that the fiscal sustainability "lens" is applied to future decisions, with a view to preventing decisions or actions from inadvertently worsening o f the fiscal sustainability o f the security sector inthe short- or medium-tern. A notable example would be to avoid further wage increases or loss o f control over staffing establishment size. Of particular importance will be instituting program- and policy-based budgeting o f security sector expenditures 49 L costs for the ANA are expected to decline in coming years, its sustaining costs if it reaches the targeted size will be very substantial. Start-up investments for the ANP mean that expenditures by the Ministry o f Interior are likely to surge for the next several years, and counter-narcotics expenditures although subject to uncertainties and fluctuations will remain very large. Table 5.1: Functional Breakdown of Expenditures for the Security Sector (2002/03 to 2004/05) US$ million Note: Thedata in the above table are not strictly comparable with the budget data used in the other tables andfigures in Chapter 4 and Chapter 5, due to dgerent categories and data uncertainties. Source: MoF data, information from other sources, and staff estimates. 5.4 Turningto the economic classification o f security sector expenditures (i.e. as betweenwages and salaries, purchases o f goods and services, transfers, and capital investments), data on External Budget expenditures are not available, butrough estimates were made and are presentedalong with available data on the Core Budget inFigure 5.2. While Core Budget expenditures on the security sector not surprisingly have high salary and non-wage recurrent components (estimated at 84% in 2004/05 and budgeted at 70% in 2005/06), it is striking that around half of the External Budget for the security sector is spent on recurrent costs (an estimated 40% in2004/05 and 53% in2005/06), with a highcomponent o f wages and salaries in both Core and External Budgets. This reflects massive donor funding o f and direct payments for salaries o f the ANA, counter-narcotics forces, the National Directorate o f Security, and now expected for the police (via U S support). As discussed in Chapter 4, this poses a major threat to fiscal sustainability, as these costs could not be absorbed by the Core Budget if financing through the External Budget declines. There are also issues related to maintaining sound PFM controls in a situation where salaries are being made by different donors invarious ways, bypassing the Government payroll. B.ImbalancesandAnomalies,SequencingIssues, andContributingFactors 5.5 A detailed analysis o fthe structure o f security expenditures i s beyondthe scope o fthis volume, as it gets into technical requirements etc. which are highly sector-specific. Moreover, it is difficult to comment extensively even on the broader structure o f security sector expenditures untilthere i s inplace a national security strategy and sector strategies for the main components within the security pillar. And the information base i s weak, especially with respect to the economic classification of security sector expenditures inthe External Budget. Nevertheless, some general comments are offered below. 51 Figure 5.2: Economic Classificationof Security Sector Spending, 2004/05 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% L 0% Core External Source: MoF budget documents andfiscal reports, staff estimates. Recurrent i Capital APPARENT IMBALANCES AND ANOMALIES 5.6 The l o w spending on the justice system i s strihng (see Figure 5.1) and i s reflective o f the fact that reforms in the justice sector have made much less progress than improvements elsewhere in the security system. This seems very much disconnected with the concerns voiced by most Afghanistan about justice and crime (see, for example, CSIS, 2005). There has been a sharp increase in spending on the justice sector in the past year, which if it continues may remedy this imbalance. However, it should be recognized that l o w funding i s not the only factor behind lagging progress in the justice sector, and that more fundingwithout resolving other problems i s unlikely to achieve sustained improvements. 5.7 The national police until recently appears to have been underfunded (Table 5.1), although with the large amounts programmed for spending on the ANP in 2005/06 and plans for the coming years, it would seem that this imbalance i s inthe process o f being corrected. However, similar issues arise with respect to the sustaining costs o fthe ANP as inthe case o f the ANA (see Chapter 4). 5.8 It is difficult to assess the adequacy of counter-narcotics expenditures due to imperfect data and major fluctuations in spending. Moreover, it i s questionable whether the National Drug Control Strategy (NDCS) promulgated in 2003 is effectively guiding resource allocation in this sector. It seems much more likely that spending on counter-narcotics i s occurring in an ad-hoc manner and in response to various short-run pressures. Some donors, however, most notably the USA, have given indications o f medium-term commitments o f funds for counter-narcotics purposes. 5.9 Although information i s scanty, it also seems based on the available budgetary data that the Ministry of Border and Tribal Affairs (MoBTA) is being grossly underfunded inrelation to its heavy responsibilities related to border controls. However, there appear to be large amounts o f off-budget spending, and inparticular salaries of border militia may be paid through off-budget channels. A major initiative is underway to strengthen border control forces and a conference with regional countries is to be held on this subject inearly 2006. 5.10 Spending on mine action has increased from the $20-30 million per year range in the late 1990s to a peak o f nearly $100 million annually in the most recent year for which information i s available 52 (virtually all in the External Budget). This has been driven both by the enormous increase in funding available for Afghanistan's reconstruction and by the requirements for de-mining associated with reconstruction projects, e.g. roads and other infrastructure. Substantial funding will need to continue in order to enable Afghanistan to achieve the ambitious target put forward in its MDG report to rid the country o f anti-personnel mines by 2013 and o f other explosive contaminants by 2015 (Afghanistan Government, 2005b, pp. 109-110). 5.11 A more general issue i s the relative prioritization as between actions to establish, expand, and strengthen security forces and measures to buildcapacity and effectiveness in security sector management and oversight bodies. While developing security forces in a timely manner is essential to achieve good results in terms o f security on the ground, the role o f effective management and oversight o f security forces by the civil authorities also i s crucial (which was entirely lost duringthe long period o f conflict). Inthis regard, the slow progress in administrative refoms in key management and oversight bodies, including notably MoI, i s a matter o f concern (see Chapter 2). In addition, the administrative strengthening and restructuring o f MOD has occurred outside the Government's interim pubic administration reform arrangements. 5.12 With regard to counter-narcotics, the current institutional arrangements include the recently- formed Ministry of Counter-Narcotics, which i s mandated to play a policy oversight and coordinating role. The longer-tem fiscal and expenditure management implications are not yet clear. Pay structures are not fully clear but appear to be high and variegated at least in some parts of the counter-narcotics apparatus, with direct payments o f salariedtop-ups by donors playing a prominent role. There i s a significant fear that the costs o f counter-narcotics activities will spiral upward. The question o f how best to focus investments (alternative livelihoods, interdiction, eradication, etc.) needs to be continuously pursued. It i s difficult to capture all information regarding counter narcotics funding. With regard to the size and composition o f counter-narcotics expenditures, the following points emerge: The allocation of resources for 2004/05 appears to provide about two-thirds o f expenditure allocations for alternative livelihoods and governance and one third for law enforcement, reflectingthe priorities o f the NDCS. Resources could increase considerably from 2004/05 to 2005/06, to roughly $1.6 billion, a considerable share o f total public spending in Afghanistan. This reflects the judgment o f major participants that levels o f programming and fundingin2004/05 (even ifwell balanced and inline with strategy) were well below levelsneededto have an impact on the problem; Allocations to law enforcement have nearly trebled from 2004/05 to 2005/06, reflecting the greater emphasis on interdiction and eradication by key donors. Allocations to alternative livelihoods and governance programs are also set to rise in2005/06 (but not by so great a multiple, so that the 2005/06 program allocates a higher share o f resources to law enforcement, and a lower share to alternative livelihoods and governance, than in2004/05). The numbers for alternative livelihoods are complicated by methodological problems as to what "counts" - but it i s clear that in 2005/06 the balance has shifted from alternative livelihoods toward law enforcement. Finally, there i s very little funding allocated for demand reduction within total counter-narcotics spending. SEQUENCINGISSUES 5.13 Interms of sequencing - the prioritization of activities and expenditures over time - there also appear to be some serious issues, although it must be recognizedthat careful consideration o f sequencing aspects may have been difficult or impossible inthe initial, "emergency" phase following 2001. Thus the discussion below reflects the ex-post "de-facto" sequencing which emerges from the data on expenditures rather than explicit decisions on sequencing made at the time. 53 5.14 Based on the overall structure o f expenditures, and keeping in mind the very large spending on international security forces (Coalition and ISAF), the sequencing o f security sector expenditures over the past four years has been broadly as follows: 0 The initial focus was on following up the anti-Taliban war with a large international security presence, including the Coalition with primarily anti-terrorisdcounter-insurgency objectives and ISAF which was mainly intended to provide security for the Afghan government and in Kabul (albeit with a mandate to expand beyond Kabul that was taken up only very slowly). 0 After an interval o f about a year, probably reflecting decisionmaking, start-up requirements, and breaking through political constraints, the focus was on establishing and progressively staffing up the ANA (the latter process continues, and subsequently strengthening o f MODalso has been occurring, albeit outside o fthe Government's public administrationreformmechanisms). 0 Shortly thereafter, and with increasing momentum over time, the emphasis with respect to international security forces was placed on setting up Provincial Reconstruction Teams (PRTs) in a growing number o f provinces - now reaching 22, in all regions o f the country. Although not envisioned in the Bonn Agreement, this turned out to be the mechanism for I S A F expansion outside o f Kabul (albeit slow and limited). The mandate o f the PRTs goes well beyond security narrowly construed to encompass local governance and development objectives. 0 There has been a growing emphasis on counter-narcotics spending, culminating with the strong national campaign against opium poppy planting in 2004, accompanied by much more aggressive interdiction activities based on small but high-quality forces built up over the past several years. 0 De-facto, the reform and strengthening o f the police appears to have been given relatively low priority in the initial years. Reforms in M o I also have lagged. More recently, the focus has belatedly shifted to the ANP which i s now being provided with higher levels o f knding through the External Budget. 0 With the continuing insurgency incertain parts o fthe country, apparently inlargepart funded and organized from across the border inPalustan, the counter-insurgency role o f the Coalition forces has again been receiving greater emphasis, including aggressive operations against insurgents' sanctuaries inremote mountainous areas. 0 All along, thejustice sector and border control forces have made considerably less progress than other parts o fthe security sector. 5.15 Much o f this de-facto sequencing, which emerged from the exigencies o f the situation faced in 2001 and immediately thereafter, would appear to make good sense. However, the delay untilrecently in moving ahead with M o I and police reforms, relative to the heavy focus on the ANA, may not have been appropriate given the complementarities between internal security and counter-insurgency efforts. While reform and strengthening o f the justice system undoubtedly involves complex and difficult issues and takes time, the slowness inmoving forward inthis area has been problematic. CONTRIBUTING FACTORS 5.16 The issues related to prioritization and sequencing outlined above would appear to stem in large part from shortcomings inthe PFM system inthe security sector. But first, the lack o f an agreed national security strategy, while not surprisinginthe initial period following the end o f major combat inlate 2001, clearly has been a constraint hindering effective prioritization o f expenditures. Preparation (currently well advanced), debate, approval, and promulgation o f the national security strategy will greatly facilitate effective prioritization. The same applies, to varying degrees, to prioritization of expenditures within sectors, although some o f them have a better strategic anchor for expenditure decisions than others. 5.17 Second, the overwhelming dominance o f the External Budget in total security funding has made itvery difficult for the Government to prioritize across different sectors andmajor programs. This is true 54 o f many other sectors as well, but as noted in Chapter 3 the situation appears to be worse in the security sector. Difficulties in prioritizing are accentuated by the massive spending by international security forces and their direct engagement in security operations. 5.18 Third, the above problems and fragmentation of decisionmaking across donors make it more difficult for the National Security Council (NSC) to fully carry out its mandated strategic, leadership, and coordination functions, although the situation in this regard has improved in 2004 and 2005. Nevertheless, effective prioritizationremains difficult inthis context. 5.19 Fourth, fragmentation indecisionmaking and fundingacross sectors andprogramsmay have been exacerbated by the "lead donor" approach. This may have encouraged line ministries to go their own ways (in partnership with the concerned lead donor), rather than following the lead set by the NSC and security sector coordinationmechanisms. 5.20 Fifth, under the lead donor approach progress indifferent sectors and programs has been at least to some extent dependent on the lead donor - its resources available for fundingthe sector, the degree o f pro-activity exhibited, and the effectiveness o f the advisory and other support it broughtto bear. 5.21 Sixth, the lag in reforms and capacity building in the key security sector management and oversight bodies undoubtedly has hindered effective prioritization. Line ministries, for example, for the most part lack the capacity for sound program design and budgeting. C.Usingthe Budget as a PrioritizationTool 5.22 Afghanistan's national budget, like the budget o f any country, i s the appropriate tool for deciding among competing claims and for allocating resources according to national priorities - best set out in a widely-owned national development strategy. As has been seen earlier in this volume, the budget in Afghanistan has not been able to play this role effectively inthe security sector, despite significant efforts at coordination. Box 5.1 goes into some degree o f specificity indescribing the limited role o f the national budget process indifferent sectors o fthe security pillar. 5.23 The dominance o f the External Budget i s an impediment to malung the national budget an effective tool o f prioritization that i s not likely to substantially change inthe near future. Nevertheless, it should be possible to achieve greater cohesion and prioritization around the national budget, with expenditure decisions anchored in a national security strategy and individual sector strategies. What i s needed i s to get away from the fragmented, inthe past indeed almost isolatedpatterns of decisionmaking and resource allocation. This will require further enhancing the capacity o f the National Security Council to effectively carry out its coordination role vis-A-vis the security sector ministries and various donors. And MoF will need inparticular to exercise an adequate degree of oversight with re,spect to the budget process. 5.24 The lead donor system may need to be revisited, although under effective national coordination its disadvantages may be mitigated and its advantages (clear division o f responsibilities and at least some degree o f individual donor responsibility and accountability for achieving progress, as well as helping avoid confusion among donors) may come to the fore. However, if it i s retained, the lead donor approach should not mean that each donor goes its own way, even in partnership with the concerned line agency. Lead donors needto operate inthe context o fnational oversight and Government-led coordination. 55 Box 5. 1: The Roleofthe BudgetinDeterminingPrioritiesinDifferent Sectors Military: The planned force size of the ANA was determined with reference to political considerations and the intention at that time to absorb substantial numbers of armed combatants in the new army The ANA'S pay and grading structure was determinedby the Office of Military Coordination(OMC) and the US Government, which also has roughly estimatedthe medium-term"sustaining costs" of the ANA. Although the Ministryof Defense (MOD)and National Security Council (NSC) have been increasingly brought into the decision making process, allocation decisionshave been made largely outside of the nationalbudget framework. Giventhe impact of defense spending on the national budget, the previouslack of engagement with MoF on the overall impact of defense spending decisions and fiscal sustainabilityissueshasbeen a major weakness. Moreover, the estimatedlonger-termsustainingcosts of the ANA haveappearednot to havebeenfully discussedor agreedwith the Government. Police: The recurrent costs of the nationalpolice so far have by and large been determined and disciplined by the national budget, despite the fact that the ANP force size also was set in the absence of a national threat assessment. Base salaries, food allowances, and operations and maintenancecosts, as well as a number of top-ups and incentives have been paid through the budget, reimbursed through the Law and Order Trust Fund for Afghanistan (LOTFA). However the development budget, unlike the operating budget, has been entirely determined by the main funding partners for the law andorder program(the EC, Germany,UK,UNDP, andthe US) incollaborationwith MoI. Certain priorities, such as the ID card system, have not been funded despite the fact that they have remained the highest priority for the MoI. Recently, anew pay and gradingstructure for the ANP has beenproposed, based on discussions inajoint Govemment`donor working group, althoughMoFwas notdirectly involved. Consequentlythe new structure, which brings ANP salaries to parity with those of the ANA, has not been assessed from the point of view of fiscal sustainabilityor budget composition. Judiciary: A substantialproportion of the day-to-dayoperating costs (salaries and non salary recurrent expenditures) for the Ministry of Justice, the Supreme Court, the Attorney General's office and the prison service are covered by the Core Budget. However, development spending, equivalent to almost 200% of the operating budget, is almost exclusively channeled through the External Budget. Moreover, significant recurrent costs (e.g. furniture, office renovation, staff training, etc.) are embedded within project finance in the External Budget. There are concerns regarding staffing numbers, for the courts in particular, with wide differences reported between the Office of Administrative Affairs and the Ministry of Justice, particularly with regard to sub-nationaljudicial staffing. The payment of top-ups from the ExternalBudget remains a major area of concern and will result in substantially higher salary costs oncethe majority of salariesmigrateto the PRR scale. Disarmament, Demobilization, and Reintegration of a-combatants (DDR): DDR has been funded through the External Budget, with expenditure levels and allotments determined by the National Budget Committee, based on proposals by ANBP. Adjustments have been made at mid-year through the Mid Year Review exercise. As this program carries with it no long-term fiscal obligations for the Government, and given the political sensitivities and uncertainties involved, the fact that this was not funded throughthe Core Budget i s probably not an issue of concern. The annual budget process and mid-year review exercisehave allowedthe Government to exercise some control over bothlevelsandallotments, to the extentrationally required. Counter Narcotics: The Core Budget, which provides salaries and operatingcosts of permanent institutions, captures very little counter narcotics-relatedexpenditure. In 2004105 the basic salary and operatingcosts of the CNPA were paid from LOTFA, which meant they were in the Core Budget. At present, the UK is funding much of the CNPA budget directly, with the objective of transferringthis to the nationalbudget as the PRR processproceeds. With these exceptions, all other counter narcotics-related institutionsandactivitiesare paidfor throughExternalBudgetprograms. The programs in the 2004105 budget are largely congruent with the NDCS and its priorities. However, the memorandumlisting activitiesfor counter-narcotics as a cross-cuttingtheme is a mixed bag that provides little insight into how resource allocation relates to strategic priorities. Some mismatches between investment and operating budgetshave occurred, e.g. an investmentbudget was providedfor CNPA without the associatedoperatingbudget. In 2005106, the US will be providing substantial resources for CN, largely on a task force basis; so most of the programmedUS resources (totaling close to $800 million) are likely to be outside the budget. Other donor resources are expectedto bewithin the budget, and the UK is activelypursuingastrategy of "budgetization" for all of its aid. 5.25 As more recurrent expenditures inthe security sector are shifted from the External Budget to the Core Budget, prioritization will be facilitated. However, as noted in Chapter 4, fiscal constraints dictate that this shift will need to be a carefully managed, fairly gradual process. Moreover, failure to prioritize effectively now i s likely to lead to increasingly serious problems inthe future, related to both sequencing and fiscal sustainability. 5.26 Inadditionthere are some more general lessons andpointerswithrespect to the technical aspects of prioritization, which are summarized inBox 5.2. These are far from a panacea especially inthe case o f 56 inter-sectoral prioritization, but ifthe more fundamental elements described above are inplace, such tools can be useful ifapplied as part o f a soundbudget process. Box 5.2: How to Prioritize? Prioritizing expenditures - determining what requires the most funds, what should be done first, what should be done later, and what not at all - is not an easy task in any sector, and security is no exception. Sound PFM principles (Box 3.1) can provide some guidance. A few additional pointers are listed below, separately for intersectoral and intrasectoral prioritization. Prioritization across sectors: Sophisticated technical approaches may require a great deal o fresources and even so are highly imperfect and can result inmisleadingconclusions. Simpler rules of thumb may work better, andjudgment i s called for. Intersectoral allocation decisions need to be anchored in an overall strategy (in this case the national security strategy). Identifying and mitigating gross anomalies in intersectoral allocations can be a powerful tool (e.g. the serious underfundingo f thejustice sector, discussed earlier). International patterns o f spending and outcomes across sectors may provide some guidance on where finds should be allocated. Identifjmg bottlenecks and directing expenditures to alleviate them may be a useful short-run tool approach. Some linkage between sectoral allocations and performance may make sense, but this needs to be watched carefully and corrective actions taken to ensure that large imbalances between sectors (in relation to the overall strategy) do not emerge over time. Finally, as a very important matter o f sound public finance management, it i s essential that intersectoral prioritization be comprehensive -no sector should be considered sacrosanct or immune from scrutiny. Prioritization within sectors: Technically, prioritization within sectors is more manageable than intersectoral prioritization because the alternative tend to be more comparable. Intrasectoral allocation decisions needto be anchored ina sound sector strategy. Technical analysis (e.g. cost-effectiveness analysis) can be used to evaluate individual project proposals and also to make more generic prioritization decisions. Ruthless culling o f projects with lower benefits and/or which do not have good prospects for rapid completion will improve the overall timeliness, efficiency, and effectiveness o f sector programs. Complementarities need to be factored into prioritization decisions (e.g. there i s no point in having a police force that i s not adequately equipped - see Box 5.4). Correcting geographical and other disparities may help guide prioritization. Source: VolumeI,Chapter 5 5.27 With respect to inter-sectoral allocations o f budget resources, simple "rules of thumb" may be helpful whereas sophisticated technical approaches are difficult to apply, require significant resources, and could give problematic results. Having a strategic anchor (i.e. national security strategy and sector strategies) which sets clear objectives will facilitate use o f simple prioritization tools. Correcting gross imbalances should be the first priority. Relevant international comparisons may provide some guidance, for instance the size and cost o f police forces (in relation to population, and talung into account geographical conditions) innearby countries and other poor countries broadly comparable to Afghanistan. However, such comparisons need to be used carefully and avoid use o f less relevant comparators, for example those with inappropriately high cost structures as in countries with higher average per-capita income. Addressing bottlenecks may be a useful focus inthe short run. 5.28 With respect to the External Budget, having comprehensive budget information will itself be useful in guiding the prioritization process and making all stakeholders, including donors, aware o f imbalances and anomalies that need to be corrected. Donors also can make use o f some o f the tools outlined inBox 5.2, and can respond to Government leadership when the Government applies such tools. 5.29 With respect to intra-sectoralprioritization, some o fthe technical tools available are more reliable and can be applied, even inthe security sector. A notable example is economic (or socio-economic) cost- benefit analysis. As discussed in Box 5.3, a cost-benefit analysis o f landmine clearance in Afghanistan yields some useful guidance on prioritization, both in terms o f which types o f land contaminated by landmines carry the highest net socio-economic benefits if cleared (and therefore should be cleared first 57 with limitedresources) and with respect to the choice o fmine clearance techniques (manual, mechanical, use o f mine detection dogs). In the present context o f reconstruction, de-mining is in many cases an essential component to enable projects to be constructed and benefits to be achieved, and should be an integral part o f such projects." A good example is rehabilitation or construction o f roads. De-mining i s also important for irrigationprojects, and for many agriculturalprojects. Box 5.3: Socio-Economic Cost-Benefit Analysis of LandmineClearance Afghanistan faces one o f the worst landmine problems in the world, a legacy o f more than two decades o f conflict starting with the occupation by the Soviet Union in the 1980s. A mine action program has been underway in Afghanistan for well over a decade and has achieved considerable progress in de-mining, although very large areas o f mine-contaminated land remain. During 2000-2001 a study was carried out, which aimed to assess the socio-economic costs o f landmines and the corresponding (net) benefits o f mine clearance, with a view to assessingthe returns to the mine action program and providing guidance with respect to choice o f de-mining techniques andprioritization o f different types o f land for mine clearance. Along with their tragic consequences for individual victims, their families, and communities, the socio-economic costs o f landmines can be divided into three broad categories: (i) loss o f life, health, human production potential, and human welfare resulting from mine accidents; (ii)denial o f access to mine-infested land and loss o f associated production or consumption benefits from such land; and (iii) distortion o f behavior due to the existence o f mines with consequent costs, including longer travel distances, journeys not undertaken due to threat o f landmines, etc. In the case o f Afghanistan loss o f livestock that accidentally detonate mines is another significant socio-economic cost. Correspondingly, the socio-economic benefits o f mine clearance can be summarized as follows: 0 Gains in economic productivitydue to reduced humanlosses from mine accidents. 0 The corresponding savings inmedical costs. The improvement inhumanwelfare attributable to reduced losses from mine accidents. 0 The net economic benefits attributable to returning landcleared o f landmines to productive use. 0 Reduced losses o f livestock from mine accidents. The study estimated these benefits (per unit o f land area) based on data compiled by the UN Mine Action Centre for Afghanistan (MACA) and compared them with the unit costs o f mine clearance using different techniques (manual, mechanical, use o f mine detection dogs, and community-based de-mining). On this basis, the study derived socio-economic benefit-cost ratios for de-mining various types o f land (roads, irrigation systems and canals, agricultural land, residential land, and grazing land) using different techniques. The socio-economic benefits associated with any particular piece o f land vary, and the cost o f de-mining is not predictable in advance but varies with the number o f mines planted and other conditions (e.g. amount o f rubble covering up mines). Nevertheless, the broad averages on both benefit and cost sides provide useful guidance for prioritization: On the benefit side, clearance o f irrigation systems and canals carries the highest socio-economic returns (as measured by benefit-cost ratios), followed by roads, agricultural land, and then, with much lower returns, grazing land. In essence, the economic value o f the land not surprisingly i s the main determinanto fthe socio-economic returns to mine clearance. Onthe cost side, use of mine-detection dogs was found to be the most cost-efficient method o f mine clearance for all types of land on which use o f this technique is feasible (although there is continuing debate about the conditions under which use oi dogs i s appropriate); mechanical mine clearance i s the most expensive technique and may be called for only in situations where use o f other techniques i s too dangerous; and manual clearance falls inbetween. These findings provide some basic guidance ("rules o f thumb") on choice o f mine clearance techniques and prioritization oj different types o f land to de-mine. The study also estimated the internal rate o f return o f Afghanistan's mine action program as a whole, which at 28% (19% based on a sensitivity analysis with more highly conservative assumptions) was considerec sufficiently high tojustify the public resources devoted to mine clearance at the time. Source: Byrd and Gildestad (2002). 5.30 Such formal technical analysis i s not as easy to apply in other parts o f the security sector, as socio-economic benefits may be difficult to specify precisely, and the relation between expenditures on security sector programs and outcomes i s less clear and direct. However, the alternative livelihoods component o f the counter-narcotics effort, where projects have clear development-related objectives, should be amenable to cost-benefit analysis. Moreover, where benefits are hard to specify or measure, some form o f cost-effectiveness analysis may be a useful tool for selecting among alternatives for achievingbasically similar objectives. And finally, even without conducting formal analysis, availability of information on results and outcomes, and regular monitoring o f such information while also making ~~ See Box 6.1 for a discussion o f implementation issues with respect to de-mining. 58 use o f data on costs (which should be available from budget and other documents), would provide useful guidance for prioritization. 5.3 1 Complementarities are very important and can provide useful guidance on prioritization within sectors. For example, expanding the size o f the police force requires a range o f complementary inputsfor the police officers, as outlined in Box 5.4. Failure to take into account such requirements drastically reduces the benefits o f any staffing increases (and associated training). Factoring in complementary expenditure requirements can be very helpful in several respects. First, it allows prioritization decisions to be made on the basis o f the full costs o f the different programs, so that on the cost side the choices are comparable. Second, complementarities may provide guidance on short-run expenditure prioritization, i.e. to prioritize items that are essential for the effective functioning o f the security service concerned and therefore complement other spending (e.g. on salaries). And finally, including complementary expenditures i s the only way to assess whether force sizes and programs are affordable; there would be no point, for example, in training a sizable police force or army that cannot be properly equipped and maintained over the medium term. I Box 5.4: Equippingthe PoliceForce The near-term staffing target for the ANP is 50,000 by the end o f 2005, with an eventual size targeted at around 62,000. Along with recruiting and training the police officers, they need to be properly equipped in order to carry out their functions effectively. With respect to small arms for the police, as of early 2005 the Ministry o f Interior had approximately 22,000 serviceable weapons, mainly AK-47s. The Government considers that there is a need for at least 30,000 sidearms and another 20,000 AK- 47s. Fundingis also required for ammunition. A second issue is vehicles. M o I recently procured about 1,500 Russianjeeps to augment its small fleet o f existing vehicles. UNOPS has been operating approximately 1,100 of the same types o fjeeps for election work, to be turned over to M o I after the elections. In addition, different types of vehicles, including pickup trucks and buses, have been donated directly by Germany and Japan. This brings the total M o I fleet o f serviceable vehicles to about 3,000. However, the ratio o f police officers to vehicles is still too high, with an identified shortfall o f around 5,500 vehicles. A third issue is fuel, which if inadequate means that vehicles cannot be used effectively for operational purposes. Thus adequate allocations for fuel are essential. A fourth issue is utilities. Police stations cannot function well without reliable power supplies for example. This requires money to pay electricity bills, and/or special arrangements where grid power supply is unreliable. A fifth issue is communications, which remains a problem for MoI. Last year, Motorola entered into negotiations with the Ministry of Interior to build a comprehensive nationwide communications system (not including border police communications, currently being funded by the US Department o f Defense). However, shortfalls in donor funding have placed this project on hold. With respect to uniforms, a distinctive new uniform for the Afghan police was introduced last year. Approximately 7,000 police officers had been outfitted as of early 2005. The Government requires more funds to equip the rest o f the police force with new uniforms complete with personal gear, rain gear, and cold weather gear. Source: Adapted and expandedfrom Afghanistan Government (2005a, Theme 8 - Security, Justice and Law and Order, Bo2 5.32 Inconclusion, technical tools may be usehl inproviding guidance for decision malung on inter- sectoral and especially intra-sectoral expenditure allocations. However, such tools need to be used selectively and flexibly. Moreover, they cannot substitute for improvements inmore fundamental aspects likethe need for a national security strategy and sector strategies, capacity development insecurity sector management and oversight bodies, bringing in the Ministry o f Finance on broader fiscal aspects o f security sector expenditures, and enhancing donor alignment on prioritization o f the External Budget under Government leadership. 59 CHAPTER6. BUDGETEXECUTIONFOREFFECTIVEAND EFFICIENTUSEOFFUNDS 6.1 The third key outcome o f a PFM system, no less inthe security sector than inother sectors, i s the effective and efficient utilization funds to achieve stipulated purposes. This chapter first reviews key PFM principles related to budget execution, arguing that different or exceptional arrangements for the security sector are not justified and detract from public finance management as well as security sector objectives, and in particular hinder effective and efficient utilization o f resources. The chapter then discusses areas for strengthening budget execution in Afghanistan's security sector. The final section looks at the implications o ftaking a service delivery approach to the security sector, with some discussion o fperformance indicators and monitoring. A. Key PFMPrinciplesfor BudgetExecution 6.2 Once a budget has been approved and monies appropriated, the goal i s to ensure that resources are used efficiently and effectively in pursuit o f strategic sectoral priorities. This requires careful monitoring and evaluation o f operational performance, both within the security services themselves and by civil servants inthe relevant management and oversight bodies. As a generalrule, funds appropriated should be spent for the purposes and inthe amounts intended. This i s necessary for sound fiscal planning and management as well as for the operational effectiveness o fthe security forces. Stability inpolicy and funding, particularly during the budget year, is important for operational performance. It i s difficult to assess stability in policy if there i s no strategy for the different components o f the security sector and funding is not predictable. At the same time, without transparent and comprehensive security budgets, it i s difficult to achieve predictability o f funding. 6.3 Well-functioning financial management information systems are critical if decision makers and public sector managers are to obtain the financial data they require for controlling aggregate expenditure, prioritizing expenditure among and within sectors, and generally operating in a cost-effective manner. It i s also extremely important that irregularities identified in the course o f monitoring performance be addressed. Failure to do so may create or reinforce a climate inwhich non-compliance can flourish. 6.4 Transparent and competitive procurement practices are a critical element in achieving value for money and cost-effectiveness. Procurement should be open to public scrutiny with expenditure fully accounted for. Purchases should also reflect agreed threat analysis, and equipment should be relevant to agreed tasking. 6.5 The security sector should meet the same standards o f accounting as non-security bodies. Security ministries should maintain an internal audit unit and consider the establishment o f an inspector- general function. The auditor-general should audit security ministry/ agency accounts regularly, along with those o f the security services. The results o f these latter audits should be reported to the legislature and irregularities addressed in a timely manner. Additionally, cash flow and expenditures should be monitored closely. Methods o f verifying the number o f personnel in the security services and employed by the security ministries and agencies and o f linkmg salary and wage payments to actual employees facilitate these tasks. Tracking studies, carried out by or for the security ministries / agencies, can be an important tool for ascertaining whether resources are reaching their intended targets. Finally, as in any sector, feedback from monitoring and evaluation into strategic planning i s critical. 6.6 Collecting and using information on security sector performance, as in the case o f other sectors, should be an objective. One indicator i s security service preparedness, which is a h n to an output. Clear objectives, specification o f preparedness, and measures o f performance supported by monitoring and 60 evaluation should all be encouraged. Value for money audits by the external auditor also help focus attention on efficiency and effectiveness issues, and the defense and other security sectors should not be excluded from the mandates o f the external auditor and other oversight bodies. Where a more performance-oriented budgeting system i s being implemented, the security sector shouldbe included. 6.7 These principles are only partly followed in Afghanistan's security sector. Lack o f information regarding public expenditures on security, sector strategies (in process o f being completed), and the weakness or absence o f key institutional monitoring and oversight mechanisms (such as the Afghanistan Financial Management Information System, AFMIS, and hitherto the National Assembly) make it difficult to assess the effectiveness and efficiency o fresource utilization inAfghanistan's security sector. B. StrengtheningBudgetExecution 6.8 The budget can be a useful policy tool only ifit i s properly executed. The Government has made significant progress in developing institutions and systems to raise fiduciary standards, including implementation o f the AFMIS in the Treasury, establishment o f the Central Procurement Facility, and promulgation o f procurement, public finance and expenditure management, and audit laws. However, because most o f the expenditures in the security sector are in the External Budget, these systems are bypassed. Ongoing refoms to strengthen accountability mechanisms will develop institutions and systems that give civil servants the incentives and tools to use public resources effectively, but this will take some time. Many of these reforms will also reduce corruption, which may be considerable incertain parts o f the security sector. 6.9 Currently, execution o f the ordinary budget proceeds largely inaccordance with the usual process o f meeting day-to-day recurrent costs for both the security services and associated civilian staff. Based on the approved annual budget, quarterly advances are provided to security sector entities, against which recurrent costs (O&M) and food allowances are paid. Innearly all cases, execution o f the development budget i s via donor funded projects inthe External Budget. The only notable exception i s DDR. 6.10 The ANA payment system has been a parallel budget execution system, although OMC-A i s strengthening the budget department o f MOD,and future salaries are expected to be paid through the new payroll system. The U S Army/OMC-A maintain full expenditure authority over U S Government funding. Salaries for AMF were provided through the national budget, but the AMF has now been phased out through DDR. Food allowances for the ANA had been provided in-kinddirectly by the U S Army but are now paidthrough the Core Budget. Investmentexpenditures are procured directly by the U S Army. 6.11 Core Budget funding is accounted for through the Government / Treasury accounting system. In the case o f reimbursement through ARTF, expenditures are verified through AFMIS by the ARTF Monitoring Agent. The Auditor General's Office currently conducts an audit o f Government accounts, although the audit methodology (sampling frame and extrapolation technique) needs to be upgraded. In reporting all operating expenditures, the standard economic object classification i s used, allowing comparative analysis between sectors. 6.12 Execution of the External Development Budget differs substantially from the recurrent or operating budget inthat, with the exception o f some Core Budget funding for law enforcement (US$16.8 million in 2004/05), mine action (US$3.9 million), and Justice (US$3.1 million), all other development budget spending i s drawn from the External Budget. The External Budget i s fully projectized, under funding from bilateral donors. As a consequence, donors procure works, supplies, and services directly through their own procurement services. For the External Budget, accounting and auditing proceduresare determined by the policies o f each bilateral donor. As financial information i s internal to bilateral 61 agencies, actual recurrent spending in the External Budget remains unknown. However, given the requirement to move toward an integrated security budget, greater disclosure will be required to enable the longer-term fiscal implications o f External Budget spending to be calculated. 6.13 Procurement. An external Procurement Advisor (Agent) has been retained by the Government, reporting to the Afghanistan Reconstruction and Development Services (ARDS) and supervised by the Ministers of Planning and Reconstruction. However, in spite o f being the procurement agent for over 17 national Ministries, it has not so far handled procurement for any part o fthe security sector. 6.14 A new ProcurementLaw has recently been passed. Hitherto, procurement of supplies, services, and works in the security ministries has not been unified around a single procurement system due to the different sources o f finance and the multiplicity o f actors. Instead, all core operatingbudget expenditures appear to follow the various procurement laws established by the budget entities directly. What is more, none o f the 110-150 procurement staff from line ministries who have received training in basic and specific procurement rules have come from security-related ministries. With the approval o f the new Procurement Law, this situation will need to change. Indeed, OMC-A has embarked on adapting procurement processes for MODto the new law. Currently, Government security sector fundingis limited to payroll and nonpayrollrecurrent costs, so no large procurements have taken place to date. 6.15 LOTFA has its own set o f procurement procedures and policies, covering both payroll and purchase o f supplies such as vehicles and radio sets as well as works contracts for construction o f police offices. Procurement i s based on open international competitive bidding overseen by the management support unit, which i s formally responsible for "procurement o f goods, services and works on a transparent and competitive basis." 6.16 Procurement o f defense items by bilateral agencies i s determined by the bilateral procurement policies o f the different contracting authorities (US DoD, UK Government, and so on). Clearly, if procurement policies are to be unified within the framework o f the national budget, bilateral commitments in support o f national security will need to be progressively drawn within the national budget framework. In essence, this would require that defense and non-defense procurement be harmonizedto enhance (i) transparency, (ii)cost-efficiency, (iii) cost-effectiveness. However, given and bilateral donors' procedures and interests, this i s unlikely to be possible any time soon. In the case o f counter-narcotics, large contracts are being tendered, with few market operators able to deliver in such a complicated political and security environment, and contract costs probably reflecting this situation. 6.17 In the case of mine clearance, discussed in Chapter 5, there are some issues related to procurement (i.e. contracting) for this activity, summarized inBox 6.1. While very specific to de-mining, these issues and the options for addressingthem do provide an illustration o f how constraints elsewhere in the security sector might be addressed. The general message i s that while the specific features may vary, itis extremely important to ensure that soundprocurement principles are followed. 6.18 Another area o f procurement related to the security sector - very important in the current situation in Afghanistan - i s procurement by international security forces for various goods and services that they require, ranging from certain security services to water, fuel, other goods, construction, construction materials, etc. Although difficult to ascertain with any precision, these activities are sizable and comprise a significant source o f demand in the Afghan economy. As noted inVolume I, Chapter 7, such contracts in many cases comprise an opportunity for development o f the Afghan private sector which should be exploited. As argued with respect to other security-related procurement, these also should follow good-practice principles for procurement, particularly with respect to transparency and competition. However, it appears that such principles often are not followed in contracting for services by international military forces. 62 Box 6.1: Contractingfor Landmine Clearance De-miningwork has some specific characteristics that affectthe contractual arrangements for carryingit out. First, it is demandingwork technicallyand safety-wise(around0.5% of de-minerswere killed or injured on thejob from mine accidentsin 1999, and training demandsare high includingfor mine-detectiondogs). Second, it is difficult to predict costs with great precisionex ante (even through survey), as minefields vary and ex-post mine clearancecosts will dependto aconsiderableextent on the specific characteristicsof aparticularminefield. Third, the quality requirementsfor de-miningare very high - a minefield that is certified as having been cleared should not haveany live minesremaining. Fourth, monitoring i s quite difficult, with respect to both outcomes (for the most part low quality of work unfortunately becomes evident only if a mine explodes on previouslyde-minedland) and costs (since the actual degree of difficulty of de- mining any particular minefield becomes evident only ex post, and is known far better to the de-mining team than to any outsideentity). And finally, it shouldbe notedthat the "de-mining industry" in manymine-affectedcountries is not very competitiveinterms of the numbersandqualificationsoforganizationspreparedto carry out mineclearance activities. For all o f these reasons, non-competitive contractual arrangements are common, and the de-mining industry is widely consideredto be vulnerableto fraudandcorruption. Despite the obstacles, good procurement practices are applicable to mine clearance contracting. Draft guidelines for mine action contracts (WMAS, 2003) recommend competitive tendering of such contracts. Particular attention needs to be focused, however,on the prequalificationof suitable candidatecontractors (based on specifiedcriteria); systematic assessment o f tenders (with a clear scoringsystemnot based on price alone); contract pricing (there i s an ongoingdebate about whether fixed-price contracts or cost-plus contracts are more appropriate); contract enforcement (contractorssometimes deliberately bid low in the expectationthat contract enhancements will come later); and monitoring and evaluation (in particular post- contractmonitoring including inspectionfor quality andthat the benefitsfrom landclearedof mines are beingreaped). Afghanistan built up significant de-mining capacity in the 199Os, consisting of a number of non-government entities undertaking de-mining activities under contract with capacity building support from the UN Mine Action Centre for Afghanistan (UN MACA). Currently there are nine "implementing partners" working on mine clearance and disposal of explosiveordnance as part of the Mine Action Program for Afghanistan (MAPA). By all indicationsthe de-mining sector in Afghanistani s reasonablyproficient and very cost-efficientby internationalstandards. Thus the prerequisitesfor competitive tendering of mine clearance contractswould appear to be inplace, as long as the important good-practiceprovisionsoutlined above and inmore detail inUNMAS (2003) are followed. 6.19 Accounting and reporting. As indicatedearlier, accounting standards andreportingrequirements should be uniform as between the security sector and non-security sectors. Within the Core Budget this condition i s basically met, although there i s room for further improvement inthe case o f security as well as other sectors. Major improvements in terms o f timeliness, quality, and harmonized information standards for expenditures in the External Budget are needed. It i s also very important that information on staffing and payrolls be improvedinboth Core and ExternalBudgets. 6.20 External scrutiny and audit. It i s essential that security expenditures be properly audited by a suitable external auditor, and that findings o f audit reviews are made available to the Government and public. Core Budget expenditures (including those on security) are audited through regular Government procedures, whose quality needs to be improved. But it is very important that External Budget expenditures have proper external scrutiny and audit, and that findings are shared with the Government. LOTFA's accounts also need to be regularly audited by an external auditor on a timely basis. With the formation o f Parliament, review o f audit reports by the legislature shouldbecome a regular practice. 6.21 In sum, security expenditures within the Core Budget are essentially subject to similar control mechanisms, accounting and reporting requirements, and procurement arrangements as in the rest of the Core Budget. The main difference is the absence of a strong internal audit function - a role that in the case o f the civilian budget i s to a large extent played by the Monitoring Agent o f the ARTF (see Box 6.2, Volume I). absence o f such a mechanism - albeit extraordinary (and temporary) - in the security The sector means a weaker control framework despite the beneficial application o f other elements of the budget executionprocess for the Core Budget. 63 6.22 Security spending in the External Budget occurs outside these Government processes. Although this spending is subject to financial controls exercisedby donors, associated disadvantages are potentially very important with respect to the quality and consistency o f information flows, and procurement. Another major problem i s the wide variety o f salary levels, payment channels, and top-up arrangements in the security sector, which risk fragmenting the accountability and loyalties of security sector staff. C.Toward Performance Measurement inthe Security Sector 6.23 It is difficult if not impossible to measure outcomes in much o f the security sector, although as noted inChapter 1the importance o f security as an enabling factor indevelopment becomes very clear in its absence. Moreover, it i s also difficult to identify performance indicators that do not inadvertently generate inappropriate incentives for security agencies. Nevertheless, and notwithstanding data limitations, as in the case of other public services it i s necessary to monitor performance as best possible and in the process generate useful information for feedback and decision malung. Appropriate performance indicators inevitably will vary across sectors, and as important as the information itself i s ensuring that it feeds into subsequent decisions including through the budget process. While detailed consideration o f performance indicators and monitoring in the security sector would get into technical aspects beyond the scope of this volume, some initial ideas for different sectors are put forwardbelow, for illustrative purposes and further consideration. The Government, in defining its MDGs, has established targets, indicators, and baselines for the security sector (mainly for Defense), and moving toward a performance-based budgeting system for the security sector will keep efforts focused on outcomes. 6.24 In the case of defense, PFM specialists generally consider that indicators o f preparedness (although they are more akm to output indicators than to outcome indicators) are the most appropriate performance indicator inpeacetime. Another set o f indicators which may be useful are the number, types, and results of training exercises conducted. It shouldbe noted inthis context that the Government has put forward the level o f defense expenditures as a ratio to GDP as a (negative) performance indicator for its MDGtarget o fenhancing security. While other things equal, a lower level o fdefense expenditure allows a greater proportion of Government budgetary resources to be spent on other development priorities, the relationshp between the level of defense expenditure and actual security i s complex and unlikely to be linear ineither the positive or the negative direction. 6.25 For thepolice, and for the internal security system more generally, the ultimate outcome o f local security and rule o f law i s very difficult to measure but can be roughly proxied by some indicators. Official crime statistics are a partial and not very accurate (negative) indicator o f security and rule of law, since they are affected by various factors such as the extent to which crimes are reported and recorded. Hence these are hard to use in monitoring performance over time or especially across countries. Household surveys generating information on the number and types o f crime encountered by households can provide a useful check.on the validity o f official statistics (see UNODC, 2005), and they can also shed light on people's perceptions of the local security situation and their degree o f trust inpolice.12 The effectiveness o f the police and prosecutorial services can be measured by the percentage o f different categories o f cases resulting in convictions, based on which type o f evidence (confessions, physical evidence etc.). The availability o f female police officers and separate detention facilities for women are also indicative o fmitigating the risks women face during detention. 6.26 In addition, the performance of thejustice system can be measured by various indicators like numbers o f cases handled, timeliness of processing court cases, and households' perceptions of the justice ''SeeBruce andNeild (2005) for adetaileddiscussion on this topic. It contains a chapter onpolice oversight andthe use of indicatorsandanother chapter on strengtheningthe use of indicators,includingindicatorsfor delivery of safety, security, and access tojustice. 64 system and their indications o f willingness (or otherwise) to use formal justice channels when seeking redress. The existence o f Family Courts and Juvenile Courts and their functioning i s also indicative o f how sensitively and even-handedly the justice system deals with women and children. Furthermore, Afghanistan has signed a number o f international conventions related to gender, including the Convention on the Political Rights o f Women and on Elimination o f all forms o f Discrimination Against Women, which like the other international conventions ratified will require legislative and administrative measures to gwe legal baclung to the nation's commitment to fair treatment and advancement o f its women. 6.27 In the case of mine action, outcomes of mine clearance activities (land and areas free of landmines - with clear linkage to ultimate development benefits) can be specified and measured reasonably well, and compared with the costs o f mine clearance (see Box 5.3). On the other hand, the performance and effectiveness o f mine awareness-raising activities are difficult to measure. 6.28 For counter-narcotics, the outcome indicator at the national level can be specified relatively clearly: progressive and sustained reduction over time and eventual elimination o f the opium economy on a sustainable basis (i.e. pressures to resume cultivation are manageable), accompanied by a corresponding reduction in trade in opiates (the latter may occur with a lag due to remaining inventories). However, measuring the performance o f various actors and activities involved in counter-narcotics activities i s not easy, and linlung such performance to progress toward achieving this strategic objective i s even more difficult. Given the variety o f activities encompassed under the counter-narcotics rubric - ranging from eradication and interdiction to judiciary and penal development, demand reduction, and rural development programs - performance indicators for the different components will need to be appropriate to their varying characteristics. 6.29 More generally, in all parts of the security sector basic indicators of inputs can be useful in measuring performance. Examples include number and ratio o f staff on the job (confirmed by spot- inspections), percentage of staff in security services properly equipped, proportion o f staff who have received designated training, percentage o f budget resources spent (covering both Core and External Budget to the extent possible), etc. A related option i s expenditure tracking suweys, which are rarely used in the security sector as compared with social sectors, but which are feasible and provide information on actual flows o f funds to, for example, police stations. 6.30 Unit costs, although not an outcome indicator, nevertheless can be useful performance indicators o f the efficiency with which inputsfor service delivery are being provided. 6.31 All this leads to a rather formidable agenda for generating and effectively utilizing monitoring information. While there will be a need to focus attention and resources on this aspect, developing performance indicators and monitoring progress can start with small steps and make progress incrementally over time. 6.32 One promising option i s to generate and regularly monitor basic financial information on security sector expenditures inthe Core Budget, and encourage donor partners to provide similar information on as consistent a basis as possible for External Budget expenditures inthe security sector. This will provide useful information on expenditure of funds (a critical input) that can be monitored. 6.33 To illustrate how straightforward monitoring o f basic financial data can provide meaningful insights and feedback for resource allocation and other budgetary decision-making, Table 6.1 compares estimated actual expenditures with budget targets for various categories o f security sector expenditures in 2003104 and 2004105. K e y points that emerge from these simple comparisons include the following: 65 0 There are sizable deviations between budgeted and estimated actual spending for most categories of security expenditures and also major fluctuations over time, suggesting that there i s still a way to go before the budgettargets become reasonably goodpredictors o f actual expenditure^.'^ 0 Within the security sector, External Budget expenditures are substantially higher in relation to budget targets than are Core Budget expenditures. 0 Defense expenditures tended to perform better inrelation to budget targets than other parts o f the security sector, especially in2004/05. 0 Finally, comparing the security sector with the budget as a whole (see Volume 11, Part 1, Table l.l),isclearthatthesecuritysectorachievesconsiderablyhigherexpendituresinrelationto it budget targets than does the overall budget. 0 These observations suggest that the security sector as a whole (in relation to the rest of the budget), and withn the security sector the External Budget and Defense, are better "protected" in terms of expenditures inrelation to budget targets. Table 6.1: Deviation Analysis of Security Sector Budget 2004105 (us $ Budget target Total 1,105 1,510 Core 355 Operating 255 Development 99 External 619 870 140 1,155 Defense 563 797 141 790 Interior 143 209 146 498 Justice 13 16 125 61 Other 93 91 161 I I 84 I Source: MoF budget documentsandjiscal reports, staff estimates. 6.34 Another important area for basic input monitoring is employment levels. Staffing data, which have suffered from inconsistencies, need to be improved and regularly monitored throughout the security sector, including staff in security services paid through the External Budget. Such data can be augmented with indicators o fnumbers or proportionso f staffon dutyand available to carry out their duties. 6.35 Buildingonmonitoringof basic data on inputs,simple indicators o f output/performance couldbe compiled and used in monitoring. It is more important to collect and effectively utilize simple information than to devise highly sophisticated data and monitoring systems which are difficult to implement and become a burden on scarce Government capacity. 6.36 Finally, the institutions concerned need to take on the role o f monitoring and evaluation and treat it as a serious part o f their work. Most important they need to make meaningful use of performance indicatorsindecision making. l3However, the difficult circumstances and "start-up" aspects ofbudgeting particularly in2003/04 need to be kept inmind. 66 CHAPTER 7. CHALLENGES,PRIORITIES,AND RECOMMENDATIONS 7.1 This volume has analyzed from the PFMperspective security as an essential public service with major ramifications for Afghanistan's reconstruction and development. Inthis'concluding chapter, first the key challenges facing the security sector from a PFMperspective are outlined, based on the analysis inearlier chapters. Then the mainrecommendations for further improvingpublic finance management in the security sector are summarized, and possible contributions by donors to this effort are put forward. A. Summaryof Key Challenges 7.2 This analysis o f security sector expenditures conducted as part o f the PFM Review has identified the following key issues from a PFM and development perspective: 7.3 The needfor an integrated security sector strategy andpolicyframework, without which sound programmatic - e.g. on force sizes, equipment - and public expenditure decisions cannot be made. Preparation o f the National Security Strategy i s at an advanced stage. 7.4 Very high donor-executed spending through the External Budget, which renders achieving coherence o f spending across and within subsectors much more difficult, and also raises questions about the fiscal sustainability o f the security sector inthe medium term when donor funding may decline. 7.5 Dijjficulties in coordinating andprioritizing security sector expenditures and actions, reflecting the above factors and fragmentation o f decision-making across donors. These factors make it more difficult for the NSC to fully carry out its mandated strategic, leadership, and coordination functions, althoughthe situation inthis regardhas improved in2004 and 2005. 7.6 Concerns about the growing security sector wage bill and pressuresfor fiscally unaffordable salary increases, as well as thefiscal sustainability of security sector staffing levels. The ANA salary structure, determined apparently without reference to fiscal constraints or pay elsewhere in the civil service, has set a precedent which the police and other sectors aspire to and which will be fiscally costly. There are also doubts about whether recruitment practices are being regulated by staffing establishment size limits. 7.7 Slow implementation of priority public administration reforms and capacity building in security sector management and oversight institutions, which have lagged behind the development o f security forces like the ANA. This imbalance, and also the lack o f progress in some security subsectors (e.g. justice) carry a variety o frisks for the future. 7.8 The need to develop good governance and sound and sustainable financial management practices in the security sector, hlly in line with and integrated with national budget processes and fiduciary provisions. B.MainRecommendations 7.9 The most important message emerging from this analysis o f security sector expenditures i s that the security sector needs to be appropriately integrated into all aspects of Afghanistan's PFM system: medium-term fiscal programming (MTFF), annual budget formulation, staffing and payroll management, financial management and controls, and external scrutiny, among others. There i s no justification for 67 treating the security sector as separate or sacrosanct, and not subjecting it budgetary and fiduciary processes. Specific recommendations include the following. 7.10 Review the Fiscal Implications of Ongoing Security Sector Reforms: Quickly enhance the information base on security sector expenditures, and conduct an integrated security sector-wide fiscal review to forecast fiscal implications o f security sector development and reforms over the medium- and longer-term. This would serve as a basis for revisiting security sector force levels, notably for the ANA, and for discussions with donors on longer-term financing. It would also facilitate policy- and program- basedbudgeting o f security sector expenditures with a medium-term perspective. 7.11 Further Develop Policy Making Capacity in the NSC and Finalize the National Security Strategy: The National Security Advisor and the NSC are tasked with coordinating policy development and overseeing integratedstrategic planning across all security institutions. Adequate capacity i s required for these purposes. An integrated security sector strategy and policy i s needed to determine acceptable longer-term costs of security institutions and appropriately prioritize and sequence spending across and within subsectors (e.g. justice and police versus ANA). As in the case o f the ANDs as a whole (discussed below), the National Security Strategy is being developed by Afghan stakeholders, with full Government ownership (across the ministries and agencies involved), and will need to gain wide buy-in within and outside Government. Thiswill require a meaningful and substantive consultationprocess. 7.12 Treat Security as an Integral Part of the National Development Strategy Currently under Preparation: The ANDs can be used as an opportunity to ensure that the security sector both i s accorded its due importance as a development issue for Afghanistan and that the broader developmental and fiscal perspective is injected into the security sector strategy. 7.13 Implement Priority Administrative Reforms and Develop Security Sector Management and Oversight Capacity: Support reforms enhancing human resource capabilities in the Ministries o f Defense, Interior, Justice, Counter Narcotics, and in the National Security Directorate. A particular priority i s developing solid financial management capacity and practices in these line ministries. Strengthening the control framework, notably with the development o f an internal audit function, i s a priority. One option would be to focus the newly developing Internal Audit Department o f the Ministry o f Finance initially on the security sector. The vocation o f this department, as per the new Public Finance and Expenditure Management Law, i s to cover the whole o f Government. Given that a mechanism (the ARTF Monitoring Agent) is temporarily inplace for most ofthe civilian recurrentbudget, this department could initially prioritize its work programtoward the security sector. 7.14 Rigorously Enforce Staffing Size, and Set Salary Increases Based on Fiscal Absorption Capacity: It i s vital that the agreed formal staffing establishment sizes be enforced. Moreover salaries, like non salary costs, must be subject to the normal rigors o f public expenditure management and fiscal discipline -the security sector wage bill mustremain affordable. 7.15 Implementing these recommendations and other initiatives to improve security in Afghanistan, which i s a necessary condition for the country's development, undoubtedly will take time. However, the need to get a handle on the fiscal sustainability issues inthe security sector i s urgent, as there i s a risk that decisions on force levels, pay, non-salary spending, equipment, construction, etc. that are being made or implemented today will be unaffordable for Afghanistan in the future. Dialogue with donors, including on possible medium-term financial support to Afghanistan's security sector, also i s urgently needed. 68 C.Role of Donors 7.16 A s emphasized throughout this volume, donors play an even greater role in Afghanistan's security sector than they do in other sectors. Consequently they also play a very important role inpublic finance management in the security sector. Whereas donor practices inadvertently can hinder the Government's efforts to make the national budget the central instrument o f policy, reform, and resource allocation for the security sector, they can also make a major contribution to improvements inthis regard. 7.17 More specifically, donors can contribute to improving public finance management inthe security sector inthe following- ways: - Support the development o f an agreed national security strategy and corresponding strategies and policies for the individual sectors (defense, police, justice, etc.). Inthis context, reassess their views about force size and cost requirements inthe light o f near- term and medium-term fiscal constraints. Help enhance the capacity o f oversight actors (such as the Auditor General, legislature) and economic managers (Ministry o f Finance Budget Office) to address issues relating to financial management inthe security sector. Channelmore o f their assistance to the security sector through the Core Budget. For resources not channeled through the Core Budget, ensure that timely and accurate information i s made available to the budget authorities, and maximize use o f internationally acceptable procedures e.g. for procurement. Help key security ministries strengthen their budget formulation and budget execution processes. Discuss medium- and longer-term availability o f external resources for the security sector. 69 REFERENCES Afghanistan Government (2004). SecuringAfghanistan s Future. Afghanistan Government (2005a). Accelerating Economic Development: Afghanistan Development Forum 2005. http://h.ww.adjgov.af/. Afghanistan Government (2005b). Afghanistan 's Millennium Development Goals Report 2005: Vision 2020. http://www.adf.gov.af. ARTF Monitoring Agent (2005). "Quarterly Monitoring Report: Afghanistan Reconstruction Trust Fund" (various quarters). Ball, Nicole and Malcolm Holmes (2002). "Integrating Defense into Public ExpenditureWork". Paper commissioned by UK Department for InternationalDevelopment (January). Bruce, David and Rachel Neild (2005). The Police That We Want: A Handbook for Oversight of the Police in South Africa. Johannesburg: Centre for the Study of Violence and Reconciliation. Byrd, William and Bjorn Gildestad (2002). "The Socio-Economic Impact of Mine Action in Afghanistan: A Cost-Benefit Analysis". World Bank South Asia Region Internal Discussion Papers, No. 184 (January). Byrd, William and Chstopher Ward (2004). "Drugs and Development in Afghanistan". World Bank Social Development Papers: Conflict Prevention and Reconstruction, No. 18 (December). Center for Strategy and International Studies (2005). In the Balance: Measuring Progress in Afghanistan. Washington, DC: CSIS. Chalmers, Malcolm (2005). "Supporting Security in Fragile States". Paper presented at the LICUS Learning Seminar, World Bank, Washington, DC (September). Colletta, Nat J. and Michelle L. Cullen (2000). Violent Conflict and the Transformation of Social Capital: Lessonsfrom Cambodia, Rwanda, Guatemala, and Somalia. Washington, DC: WorldBank. Frank, Cheryl (2005). "Costing Crime: What Value i s There for Policy Making?" SA Crime Quarterly, no. 11. Hendrickson, Dylan, Michael Bhatia, Mark Knight, and Annabel Taylor (2005). "A Review o f DFID Involvement in Provincial Reconstruction Teams (PRTS) in Afghanistan". London: Conflict, Security and Development Group, International Policy Institute, King's College (July). Kaufmann, Daniel, Art Kraay, and Massimo Mastruzzi (2005). "Governance Matters lV: Governance Indicators for 1996-2004", draft. Koehler, Jan and Christoph Zurcher. "Conflict Processing and the Opium Poppy Economy in Afghanistan". GTZPAL Internal Document No. 5 (August). Nojumi, Mazurana, and Stites (2004). Afghanistan 's System of Justice: Formal, Traditional, and Customary. Boston: TuftsUniversity. OECD (2001). TheDAC Guidelines: Helping Prevent Violent Conflict. Paris: OECD. OECD (2005a). "Conflict Prevention and Peace Building: What Counts as ODA?" Paris: OECD (March) . OECD (2005b. Security Sector Reform and Governance: Policy and Good Practice. Paris! OECD DAC Guidelines andReference Series. 70 Perito, Robert M. (2005). "The U.S. Experience with Provincial ReconshuctionTeams in Afghanistan: LessonsIdentified". United States InstituteofPeace Special Report 152(October). Police Accountability. Police Accountability Newsletter, http://www.policeaccountability.co.z~ewsletter/backIssues.asp . Sedra, Mark and Peter Middlebrook(2004). "Afghanistan's ProblematicPathto Peace: Lessons in State Building inthe Post-September 11Era". Stewart, Frances (2004). "Development and Security". Centre for Research on Inequality, Human SecurityandEthnicity (CRISE) WorkingPaper 3 (January). UKDepartmentfor InternationalDevelopment(DFID, 1999). "Poverty andthe SecuritySector". UKDFID (2005). "Fighting Povertyto Builda Safer World: A Strategy for Security andDevelopment". DFID (March). UnitedNations Mine Action Service (UNMAS, 2003). Guidefor the Drafting of Mine Action Contracts (draftJirst edition). New York UNMAS (March). UnitedNations Organizationfor Drugs and Crime (UNODC, 2005). Crime and Development in Africa. Vienna: UNODC (June). Ward, Christopher and William Byrd (2004). "Afghanistan's Opium Drug Economy". World Bank SouthAsia RegionPREMWorking Paper Series, No. 5 (December). Wily, Liz Alden (2003). Land Rights in Crisis: Restoring Tenure Security in Afghanistan. Kabul: AREU. WorldBank(1998). Public ExpenditureManagement Handbook. Washington, DC: WorldBank. World Bank(2003a). Breaking the Conflict Trap: Civil War and Development Policy (A WorldBank Policy Research Report). Washington, DC: World Bank andOxford University Press. World Bank(2003b). WorldDevelopment Report 2004: Making Services Workfor Poor People. Washington, DC: World Bank. World Bank(2005). Afghanistan: StateBuilding, Sustaining Growth, and ReducingPoverty (A World Bank Countly Study). Washington, DC: World Bank. World Bank(2006). TheInvestment Climate in Afghanistan: IncreasingForeign and DomesticPrivate Investment. Washington, DC: World Bank, forthcoming. 71