68826 Public Expenditure Review for the Solomon Islands Government Revised June 2011 The Ministry of Finance and Treasury has been working with the World Bank and the Pacific Financial Technical Assistance Center to assess how well the budget process has been working over the period 2006-10 and to explore options for improving that process. This note provides a non-technical summary of the initial findings and recommendations. I. MIXED RESULTS 1. The Solomon Islands Government (SIG) and its donor partners can claim credit for a number of achievements in education and health in recent years. School enrollment has increased in the last several years and net1 primary school enrollment is now well above 90 percent.2 The recent introduction of a fee-free policy through school year 10 is likely to yield further increases in enrollment. The low maternal mortality rate of 103 deaths3 per 100,000 live births is especially impressive because it is less than what is seen in the region and it falls within the medium range for human development. Table 1: Selected Public Service Delivery Indicators Gov’t Births spending Net Gov’t attended by per Average secondary Life Spending skilled student on years of school expectancy per capita Maternal health education schooling enrollment at birth on health mortality personnel Deaths per (PPP US$) (PPP US$) 100,000 live Average (Years) (Percent) (Years) Average births, (Percent) 2000-09 2010 2001-09 2010 2005-08 2003-08 2000-08 Solomon Islands 665 4.5 30.2 67.0 123 103 85 Fiji 922 11.0 79.1 69.2 121 210 99 Papua New Guinea .. 4.3 .. 61.6 .. 470 39 Samoa 395 .. 64.2 72.2 185 .. 100 Timor-Leste 488 2.8 31.4 62.1 91 380 19 Tonga 521 10.4 66.2 72.1 114 .. 99 Vanuatu 994 .. 38.1 70.8 104 .. 93 Lower middle income 659 .. .. .. 135 .. .. Low human development .. 4.1 30.9 56.0 .. 822 39 Medium human development .. 6.3 57.0 69.3 .. 242 74 East Asia and the Pacific .. 7.2 62.6 72.6 .. 126 91 Numbers in bold face indicate instances where the Solomon Islands do better than other countries or groups of countries. Sources: Expenditure data are from World Bank EdStats database, WHO, MOFT and World Bank staff calculations. Service delivery indicators are from UN Human Development Report, 2010 and UN ESCAP Statistical Yearbook for Asia and the Pacific, 2009. 2. SIG can achieve greater returns for its citizens from its spending. Data from the health and education sectors suggest that several neighboring countries are able to get equal or better results for less money per capita. Some of the difference in performance could be explained by varying degrees of development and even by demographics and geography. Even 1 Net enrollment is the number of students enrolled in a level of education who belong in the relevant age group, as a percentage of the population in that age group. Gross enrollment rates include students outside the relevant age group. 2 UN data published in 2010 show a lower rate of 67 percent for the period 2001-09. 3 From the World Bank Interim Strategy Note. UN data show 220 deaths per 100,000 live births. 1 so, the analysis conducted so far suggests that better results are blocked by problems in matching budgetary allocations to needs and problems with unproductive spending. 3. SIG spends more per capita4 than most of its neighbors for health services but does not consistently get better results. Life expectancy, for example, is below what has been achieved by several neighboring countries. The number of births attended by skilled health personnel may be low relative to several other countries in the region.5 (See Table 1.) Although health care is meant to be free6, there is some evidence that it may not be equitable: much of the health expenditures are for activities in Honiara. Some recommendations for improvements can be found in the World Bank (2010) report on Health Financing Options. 4. SIG spending on education per student over the period 2006-08 exceeded the average for lower middle income countries, yet the Solomon Islands lags behind several neighboring countries in terms of results. Table 1 shows that the average years of schooling per person (as a proxy for competencies) is lower than Fiji and Tonga and is just barely better than what PNG achieved. The net secondary school enrollment rate is only 30 percent7 compared to 38 percent in Vanuatu, 64 percent in Samoa, 66 percent in Tonga, and 79 percent in Fiji. Estimates of adult literacy vary greatly, with the 1999 census figure cited at 76.6 percent. However a detailed literacy assessment undertaken by the Asia Pacific Bureau of Adult Education8 in Honiara and Malaita in 2007 found that only 28 percent of students completing primary school were found to be literate, and under half of secondary graduates. Given high SIG spending in this sector, it is likely that more could be achieved per education dollar in terms of impact and quality. Recommendations for reform can be found in the Fast Track Initiative (2009) Education Sector Public Expenditure Review. 5. The geographic distribution of provincial grants is very uneven. Provincial grants make up only about 2 percent9 of total government expenditure with actual transfers sometimes less than what was allocated. The distribution of grants is very uneven. Rennell and Bellona were to receive per capita grants of roughly SBD1,500 in 2010, while Malaita was to receive SBD19.10 Real per capita grants to provinces are shown in Figure 1. Household survey data from 2005/06 indicate the situation should be reversed since Malaita has one of the poorest populations and Rennell and Bellona has one of the richest, as illustrated in Figure 2. The situation will need further study. It is possible that spending by sector ministries in each 4 Based on UN population estimates and World Bank purchasing power parity adjusted exchange rates. The population estimate from the 2009 census is 1.4 percent smaller than the UN estimate. Some observers believe the actual population could be as much as 11 percent more than the UN estimate. In that case, SIG would be spending less per capita than most countries other than Timore-Leste and Vanuatu. 5 The UN reports a much lower estimate of 43 percent for the number of births attended by skilled personnel. 6 “Out of pocket� contributions from households were estimated to finance only 6 percent of total health costs in 2008 according to the MHMS Finance Report for Year Ending 31 December 2008. Much of the financing comes from Australia which in 2009 provided over one-third of total public expenditure in the health sector, supporting pharmaceuticals, technical advisors, and provincial grant funds. 7 The introduction in 2009 of a fee-free policy through year 10 is likely to yield further increases if it can be sustained: the policy is expensive. NZAID has led a significant program of education sector planning strengthening and budget support to major initiatives such as school grants as part of the fee-free basic education. 8 Educational Experience Survey. Education, Language and Literacy Experience in the Solomon Islands. October, 2007. Asian South Pacific Bureau of Adult Education and Coalition on Education Solomon Islands. 9 No judgment is implied regarding the percentage. Much depends on what functions are devolved and how. 10 While population estimates are imprecise, the relative population sizes between provinces is less in doubt, with Mailaita the largest and Rennel and Bellona the smallest. 2 province help smooth out the distribution (although health data do not support this hypothesis). In addition, there are some scale effects to the extent that there are fixed administrative costs that must be financed regardless of the levels of provincial populations or poverty. Figure 1: Provincial Grants per Capita in 2010 Figure 2: Households in Lowest 3 Household Expenditure Deciles 1,800 1,600 40 % of Total Households 1,400 35 1,200 30 1,000 25 SBD 800 20 600 15 400 10 200 5 0 0 Central Isabel Malaita Rennel and Rennel and Isabel Central Malaita Western Guadalcanal Choiseul Temotu Makira-Ulawa Western Guadalcanal Choiseul Honiara, Town Temotu Makira-Ulawa Bellona Bellona Source: Ministry of Finance and Treasury expenditure data, Bulletin Source: Solomon Islands: Analysis of Poverty from 2005/06 HIES, 14/2010 and World Bank staff calculations. Table 17. 6. It’s also striking that the various constituency development funds help to ensure that MPs have a much larger role in many local citizens’ lives than their provincial governments do. For example, in the Province of Malaita, MPs could disburse up to SBD$24.6 million in 2010 whereas the Provincial Government spending would amount to about SBD$6 million that year. Unfortunately, there is no effective requirement that the actual use of these monies and their impact be monitored and reported.11 II. DIAGNOSIS 7. Some of the difficulties in getting good results can be traced to local conditions beyond government control. Administration costs and travel costs, for example, are increased by the geographic dispersion of the islands. Road construction is made difficult by the mountainous terrain on most islands. SIG is heavily dependent on expensive imports due to the fact that few products are manufactured in the Solomon Islands because the population is too small to justify major investments in manufacturing. Imports are expensive because the islands are quite remote from many trading partners. Yet, these obstacles are also common to most other neighboring countries12 – and many of those countries are generally performing better with regard to public service provision. 8. Some of the problems are within government control, the budget process in particular. A good budget process should do the following four things: 1) allow the government to reach agreement, in consultation with citizens, on what shall be achieved with its money; 2) provide the right amount of money to line ministries and Provincial Governments so they can achieve government goals; 3) ensure that line ministries and Provincial Governments 11 Guidelines for how the RCDF should be used were issued in 2010 but these did not include adequate reporting requirements. 12 There are some differences too. For example, the population of Samoa is not widely dispersed across many islands. The population of PNG is segregated by many deep mountain valleys. 3 make good use of the resources entrusted to them; and 4) make it possible for officials, citizens, and donors to check if government goals were achieved so that people within the government administration can be held accountable for results. More succinctly, one should look for a well formed budget, faithfully executed with full transparency and accountability throughout. None of these conditions are presently well met. 9. The budget formation process needs strengthening. The PER team found three main issues related to budget formation: a) policy priorities, plans, and budgets are not well linked or integrated; b) budgetary allocations are made with little consultation with the line ministries or citizens and without a feed-back process to facilitate corrections; and c) there is a general lack of consistency over time in allocations to public services.  It is not clear how budgets support SIG policy and plans. The Ministry of Planning requests that each ministry develop a four year corporate plan. Ministries are also asked to prepare an annual or operational plan for each year of the corporate plan. These plans lay out in more detail, what will be achieved from the implementation of each activity. These plans typically include a large number of goals, often far more than could feasibly be achieved under present human and budget constraints. National plans set targets for outputs and/or outcomes but these are not generally accompanied by cost estimates. By contrast, the national budgets are organized in terms of inputs and do not include clear expectations for service delivery targets. National plans and budgets are not linked: the budget does not show clearly and consistently what activities are to be delivered and how spending contributes to government policy priorities.  Consultations with line ministries have been poor. The only portions of a ministry budget that are systematically deliberated upon are a small fraction of the recurrent budget set aside for new activities,13 and the development budget.14 Decisions regarding most of each ministry budget are largely incremental, meaning that the bulk of the allocation decisions are percentage increases over the previous year’s baseline allocations – decisions are made with insufficient regard for what ministries should achieve or how much their goals would cost.  There are few mechanisms for consultations with citizens. Consultations with citizens and provincial governments are not used well in forming and prioritizing national plans and line ministry corporate plans. The budget process includes no formal processes for hearing citizen input and encouraging debate regarding allocations of resources. Unlike in many other poor countries, there is very limited discussion within media or civil society regarding resource allocation decisions included in the budget. Examples of citizen participation include community oversight committees (Sierra Leone) and citizen compacts (Kenya).  There may be a need to better align budgetary allocations with ministry requirements. There is a general lack of consistency over time in allocations to large, priority public services such as policing, health, or education. For example, in Table 2 below, the allocation to the Ministry of Education for personnel increases as a share of the personnel budget in 13 The process also makes it is difficult to know with certainty whether bids for alleged new activities represent genuine new activities or instead represent inflated costs for ongoing activities. 14 The PER team received mixed reports as to how much scrutiny is given to development budget expenditures. 4 2009 and decreases in 2010 – even while the allocations to the same ministry for other recurrent needs move in the opposite direction, falling in 2009 and rising in 2010. 15 This same pattern is repeated when the data are adjusted for inflation and population growth. A similar puzzle can be seen for the Ministry of Provincial Government which saw its inflation adjusted per capita allocation to personnel fall sharply by almost 40 percent in 2009 while its allocation for other goods and services increased by 8 percent!16 These changes appear hard to justify when the needs of the population are generally constant over time and when government policy objectives have not fundamentally changed over the last several years. This suggests that the current budget process may not promote good matching of allocations to needs.17 Table 2: Selected Budgetary Allocations, 2008-10 Personnel Other Recurrent 2008 2009 2010 2008 2009 2010 Share of Budget (%) Min. Education 36.4 42.7 41.1 20.6 19.9 20.3 Min. Police 12.4 10.9 11.9 7.1 7.8 7.5 Min. Provincial Government 1.6 0.8 0.8 6.7 7.2 6.8 Min. Rural Development 0.7 0.5 0.6 0.7 1.2 1.2 Inflation Adjusted Changes (year on year, %) Min. Education 8.3 35.1 -22.3 9.6 -2.0 13.7 Min. Police -17.4 1.8 -11.7 7.5 10.6 6.9 Min. Provincial Government 171.2 -39.6 -26.3 -13.7 8.0 5.9 Min. Rural Development .. -15.3 -14.3 663.2 81.9 6.0 Sources: Ministry of Finance and Treasury data and World Bank staff calculations. 10. Actual spending does not resemble approved allocations. While execution across some Ministries has been relatively close to what was approved by Parliament, there are more ministries where actual expenditure has very substantially exceeded or fallen short of budgets. This is shown in Table 3 below. In 2009 for example, 6 of 30 budget heads were spent within plus or minus 10 percent of their original approved recurrent allocations.18 That year was a difficult year in which SIG controlled resources fell short of budgeted targets by almost 14 percent.19 Yet it is interesting to contrast this outcome with the outcome from 2008 when 16 of 30 budget heads (53 percent) were spent within ±10 percent of budgeted amounts despite a 12 percent shortfall in total SIG controlled resources. Moreover, the 2008 outcome represented a slight improvement over the 2007 outcome when only 14 of 29 budget heads (48 percent) were spent at levels close to allocated amounts. While more work is needed to fully explain what happened, this comparison suggests that resource constraints are not the only cause of spending deviations. The fact that some budget heads were over-spent during periods of resource shortfalls may offer a partial clue – such spending only increases the need to reduce spending elsewhere. 15 This sort of variance is too large to be explained simply by some ministries being more successful in others in winning additional funding through the budget bid process. 16 The inflation adjusted changes in actual spending are as pronounced as those for budgeted allocations. 17 While much of the problem could be put down to the budget process, there are also issues related to the quality of record keeping and database management. In 2010, for example, all personnel allocations were too low because of erroneous information about actual personnel spending in 2009 which was to have been inflated by a cost of living adjustment for 2010. 18 Changes to approved allocations via supplemental budgets are not included here. Thus, the deviations reported in Table 2 represent the sum of sanctioned and unsanctioned within year adjustments. 19 SIG controlled resources include all revenues and all direct budget support but excludes project support and sector support. 5 11. Execution of the development budget is very weak. While SIG contributes only a fraction of the funding for the development budget, Table 3 shows that fraction has been under- spent in most ministries. In 2006-09, for example, no more than two ministries spent within plus or minus 10 percent of their approved allocation from the consolidated portion of the development budget. The majority of ministry allocations were under-spent. This is partially the consequence of SIG placing a higher priority on recurrent obligations (notably the wage bill) but it may also a reflection of generally weak capacity for project implementation throughout SIG. More exploration is needed. By contrast, donor-funded expenditure has been split with roughly half the expected contribution being over-spent and half being under-spent. Table 3: Budget Execution, 2006-09 2006 2007 2008 2009 Deviation in SIG Controlled Resources (% of Budgeted Amount) 0.3 4.8 -11.9 -13.6 Recurrent Spending a/ Good Execution (Number of budget heads within ±10%) 10 14 16 6 Poor Execution (Number of budget heads) 5 15 14 24 Over +10% of approved allocation 2 8 2 4 Under -10% of approved allocation 3 7 12 20 Spent without allocation 0 0 0 0 Total Budget Heads with Allocations or Expenditures 15 29 30 30 Development Spending, Consolidated Good Execution (Number of budget heads within ±10%) 0 2 1 0 Poor Execution (Number of budget heads) 9 13 25 26 Over +10% of approved allocation 1 1 0 1 Under -10% of approved allocation 8 12 25 25 Spent without allocation 0 2 0 0 Total Budget Heads with Allocations or Expenditures 9 17 26 26 Development Spending, Non-appropriated b/ Good Execution (Number of budget heads within ±10%) 3 10 4 5 Poor Execution (Number of budget heads) 14 13 23 20 Over +10% of approved allocation 6 6 10 3 Under -10% of approved allocation 8 7 13 17 Spent without allocation 7 2 1 2 Total Budget Heads with Allocations or Expenditures 24 25 28 27 a. Excludes sector budget support for health and education. b. The outcome in 2009 reflects changes in the accounting of support for police and justice programs as well as some under- counting in other ministry programs rather than actual under-spending. Sources: MOFT and World Bank staff calculations using most recent database from January 2011. 12. There are many reasons why the budget that is delivered does not match the budget that was approved. Here is a partial list:  Cash reserves are insufficient. As in all countries, macroeconomic forecasts to provide estimates of the amount of revenue available versus the amounts needed for spending are not always accurate. When revenue falls short of forecasts, or when needs exceed forecasts, then the only option available is to disrupt spending after cash reserves are exhausted.  Ministers have changed their minds about what money should be spent on. They have reallocated resources to different areas throughout each year by means of Cabinet Letters, Contingency Warrants and Supplementary Budgets. The need to make such changes implies either that SIG is not getting the approved budget right the first time or some people do not respect the approved budget as a legally binding instrument for expenditure control. 6  Current laws, regulations, and procedures do not provide adequate expenditure commitment controls. MoFT revised its financial instructions in 2010. If these are effectively implemented, they will significantly improve accounting and financial management practices.  Government officials have diverted (vired) resources from productive uses. Virement is an official process by which ministries are able to reallocate funding between line items. Virements can serve an important role in correcting the inevitable mis-allocations built into the budget every year. Unfortunately, most ministries have used virement for purposes that do not contribute substantially to the achievement of policy objectives. These include travel, allowances, and entertainment expenses. (See Box 1.) These diversions appear to be increasing over time.  Donor funding is volatile. Donor support often exceeding forecasts for some projects while falling short for others. Box 1: The Record on Virement Ministries typically submit a very large number of virements during the course of a year. The Ministry of Foreign Affairs had just over 300 virements approved in 2009, the Ministry of Health had nearly 400 virements that year while the Ministry of Police submitted roughly 450. The amount of funding that is vired is often a substantial portion of overall allocations. For example, they accounted for 14 percent of the approved 2009 recurrent allocation to the Office of the Prime Minister and Cabinet and 18 percent of the recurrent allocations to the ministries responsible for commerce and forestry. It is not uncommon for line ministries to submit requests for virement of funds only days after the appropriated budget becomes effective. Although the Treasury Division has often tried to deter the practice by turning them down when they are scrutinizing the applications, their usage has remained high. Some of the highly placed officers within line ministries lobby for the approving authority (the PS/the minister) when Treasury Division rejects their requests. There are no clear guidelines in the revised Financial Instructions to regulate the use of virements in the execution of the budget. Moreover, there no set minimum or maximum amounts that can be transferred through virement within a budget head provided the accounting officer responsible for the line ministry concerned has approved the request. 13. There has been little accountability for the use of resources and results achieved. The public has not been able to see what activities and programs the government was using money for, nor the results achieved, because of the way the budget is presented and because of delays in reporting actual outcomes. Line ministries have not been held to account by Ministers, in part because there is little information about what activities they will deliver or results they will achieve. Some of the reasons for this situation are listed below:  The legal framework is weak. It does not require due diligence for equity and efficiency in public service provision. It does not provide enough restraint for fiscal imprudence and taking undue fiscal risks. It does not provide a framework for improving government performance or accountability. Roles of Parliament, the Caucus, and the Office of the Prime Minister and Cabinet are not clearly spent out. In addition, it is likely that not all existing laws are well enforced. More exploration is needed.  Success has not been well defined for most activities. National plans and corporate plans typically include a large number of policy goals only some of which include performance 7 measures – and only some of those can actually be monitored by the line ministries. Many of the performance indicators are not measurable either because they are poorly formulated or are beyond ministry capacity. And, of course, it is not possible to design monitoring systems for those goals for which success is not defined.  There isn’t any central office systematically keeping track of policy outcomes. The Office of the Cabinet and Prime Minister, for example, has been largely focused in recent years on tracking political rather than technical outcomes. In addition to a central office, it will also be important for each line ministry to develop the capacity to ensure that policy goals are on track.  The Public Accounts Committee (PAC) has limited capacity. The PAC in many countries is empowered to sanction ministers and public officials for fiduciary breaches in the course of a fiscal year, based on a review of the public accounts. This role could be emphasized more. Unusually, since the 1980s, PAC in SIG also invites officials to defend their budgets. At the culmination of these hearings, the PAC issues a formal report with recommendations. The team was informed that, so far, the PAC reviews have not resulted in any substantive requests for changes beyond correcting errors spotted by PAC – calculations, spelling mistakes, incorrect page numbers et cetera. During these hearings the PAC has also sought to review progress on ministry audit action plans which they are required to submit as part of the review of the OAG’s report. This review is not done in a systematic way and effectiveness is restricted because of the limited number of parliamentary sessions.  Information on outcomes and results is scarce and often of mixed quality. This is true for national indicators such as population, GDP, and inflation but also true for various aspects of sector performance such as school enrollment, literacy rates or maternal mortality. The team noted in particular that there did not appear to be any general record keeping standards being maintained throughout the government for key documents and decisions. In a positive move, however, in 2010 SIG became the second government in the Pacific region to adopt International Public Sector Accounting Standards for its financial records.20 SIG is now also releasing fiscal outcomes to the press on a quarterly basis and has created a new website for MOFT which will also facilitate information sharing with the public.  Budgetary data is not collated or presented optimally to either the public or to decision- makers. Information regarding budgetary allocations and outcomes is scattered between three different documents (the Recurrent Estimates, the Development Estimates, and the Establishment Register), even though comprehensive understanding of the resources available to various ministries requires information from all three. The documents are not user-friendly, and not easily available to the general public. The distinction between the Development and Recurrent budgets is not clear, with both budgets containing operating and capital costs. Information regarding donor-funded projects presented in the Development Budget is provided by donors, and is often inadequate to provide a clear understanding of the purpose, nature, and cost of projects. The data that is provided by donors is sometimes inaccurate.  Some important information for management and accountability is not readily available. Until recently the Solomon Islands Government had little capacity to bring together existing information on levels and trends in payroll, recurrent, and development 20 IPSAS cash reporting provides a consistent approach to the presentation of financial information thus leading to greater transparency and accountability. Support comes from the Pacific Financial Technical Assistance Centre. 8 expenditure to build a comprehensive picture of how resources were being used, or examine changes over time. During the PER process, the PER team was required to build a new database to assess levels and trends in expenditure as much of this information could not be generated by SIG systems. Crucially, only input information is currently available to decision-makers and the public. 14. It is important to deal with the formation, execution, and accountability problems together because they are mutually reinforcing. The incentive to budget well increases when three conditions are met: i) line ministry officials have confidence that they will be consulted; ii) ministry officials have confidence that allocations provided will match what was agreed during consultations; and iii) the approved budget is the final word -- ministers cannot press for changes except in clear cases of emergency. Similarly, the incentive to execute each budget faithfully improves when line ministry officials believe it meets their needs and when they are held accountable for their decisions. III. A TEN YEAR STRATEGY FOR REFORM 15. The proposed strategy is based on just a few principles. First, government officials will deliver better results when they are empowered to do so with adequate funding and good information and when they know they will be held accountable for what happens. Second, the reform program should focus on getting the basic functions of government working well before attempting more sophisticated reforms. Third, be mindful of how long each reform step takes and which reforms must precede others. Fourth, to avoid over-burdening the system, proceed with a few ministries at first and then gradually expand the reforms to the remaining ministries with the benefit of lessons learned. Fifth, build on existing programs of donor support as much as possible. 16. A ten-year strategy based on these principles is laid out below, accompanied by brief explanations of how and why each strategic element is needed.  Decide what is to be achieved and how you will know when you have succeeded. Hold annual consultations with citizens and provincial leaders on their needs.21 Build the capacity of SIG central ministries to agree with line ministries on feasible and measurable goals. Change the format of budget documents so that links between activities and government goals are clear to decision makers and line ministry officials. If these changes are made, and if the budget process is used to get good results at the community level, then people will see that the government is more responsive to their needs. The public will have better information regarding what government money is being spent on and it will be easier to demonstrate success. At the same time, failures will also be easier to detect and correct. 21 Consultations cost money. Annual consultations limited to a few days are completely affordable. Larger consultations, such as might be conducted for a Poverty Reduction Strategy Paper, which are national in scope and conducted at multiple levels every 4-5 years, require substantial financing and planning – but yield commensurately more valuable results. This is an important option for the medium-term. 9  Make sure the line ministries know how to do each task and how much they cost. Decision makers and SIG officials need better information. As of 2010, all they have is information about previously approved allocations to ministry inputs (staff, paper, electricity, vehicles et cetera), an estimate of an affordable increment to add to those allocations, and line ministry wish-lists for new activities. More is needed. Information about actual spending patterns from the previous year could help in detecting gaps between what is needed and what was provided but this has not been available early enough to be useful in the budget cycle. This information can also help detect the need to improve capacity or internal controls with various ministries. Looking forward, it would be helpful to begin recording the actual cost of activities and programs in addition to inputs – this is a necessary first step to gaining knowledge about the unit costs of public service delivery. The new Chart of Accounts (CoA), devised as part of the FEMSP, is a good tool for this purpose because it will allow line ministries to gradually begin recording expenditures against activities in addition to inputs.22 Information about the future cost implications of various programs and projects would also be helpful. This could help officials reject projects that cannot be affordably maintained in the future and ensure that proposed spending does not exceed forecasted resource availability. In order to better present such information, SIG may also consider following other countries by placing the annual budget estimates within a rolling multi-year framework, once the annual budget process has been improved. SIG may wish to seek donor technical assistance for these reform elements – much of this agenda is not well addressed under existing programs of donor support.  Ensure ministries and provincial governments have the resources they need to achieve the goals set out for them. As more expenditure information becomes available, especially for fixed costs and unit costs, it should become possible to change the budget process so that baseline allocations are informed by consultations on what line ministries need to achieve agreed goals, as well as what they actually spent in each previous year. Baseline allocations could also be informed by the findings from expenditure reviews conducted for a few key ministries each year. If these reforms are completed, then the coordinating ministries (MOFT, MDPAC, MPS) will be better able to help line ministries achieved their agreed goals. Budget allocations will be better aligned with needs. This will help give ministry officials confidence that their efforts in budget formation are being rewarded.  Ensure that ministries and provincial governments use their resources well. Conduct expenditure reviews in key ministries to ensure that resources are being well used. Use supplemental budgets to adjust spending plans for new circumstances and needs but also require that they are fully paid for -- from new revenues, grants, or from cuts in low priority programs. In addition, new spending should not be authorized outside of approved budgetary allocations: the use of Cabinet Letters should be discontinued. Ministries should be discouraged from viring money to unproductive uses such as travel and entertainment even while allowing virement for essential adjustments in favor of public service delivery. Access to contingency warrants should be limited to unforeseen and consequential requirements for public service delivery. If these changes are made, then more resources will reach their 22 The design is well advanced. If there is any short-coming, it would be the lack of any direct coding of expenditures within provinces. Instead, assumptions are made based on various configurations of division, activity, and account codes. These may be misleading when actual spending does not conform to the assumptions made. 10 intended destinations as expressed by approved allocations. This will make it much easier for ministries to implement their programs and achieve their goals.  Hold decision makers and government officials accountable for good results. Thus far, there has been very little in the way of formal or informal requirements for accountability. The proposed strategy would start a long-run process that would increase accountability in several ways. Decision makers in the coordinating ministries could allow themselves to be held accountable to citizens through budgetary consultations with their provincial leaders or with organized citizens groups and journalists. The UNDP Provincial Government Strengthening Program could conceivably become a vehicle for the former while technical assistance could be solicited for the latter. It would also be helpful to strengthen the legal basis for the roles and accountabilities to be played by various officials. This is not as clear as it should be at present.  Accountability will require building the capacity to check if government goals are achieved. As improvements in expenditure information are made, it should become possible to change the format of the budget estimates to clearly show what activities and programs line ministries and their divisions are undertaking, and how much funding is used for each. Performance targets for ministries should be carefully identified and published. Systems should be established to allow citizens to join with the government in monitoring performance outcomes against targets. If these reforms are completed, then it will become possible for the government to hold accountable its ministers and Permanent Secretaries for their contributions to government successes and failures. In turn, Permanent Secretaries will be able to hold Division heads accountable for their performance. Moreover, the citizens will start to understand what Ministries are using money for, and whether their community needs are being met. 17. In the long-run, coordinating ministries could rely more on line ministry staff. It is not realistic to expect the staff in MOFT, MDPAC, or MPS to have a good understanding of whether a particular mix of inputs is well suited to the activities and programs of each and every line ministry, nor is it desirable. One could, however, expect the line ministries to have the best knowledge of what is needed to achieve the goals set out for them – particularly after they have begun collecting several years of data on actual spending arranged by activity and program. Thus, line ministries with a good track record of data collection, budgeting skill, and expenditure controls should eventually be granted more autonomy over how they manage their expenditures and their personnel. In particular, they should be allowed to 1) budget by activity and program rather than inputs (although inputs would still be tracked and accounted for); and 2) hire, fire, train, and promote staff as needed in order to best achieve their goals. 23 The 2010-15 Human Resource Management Strategy is largely consistent with these objectives. Very importantly promotion should be performance based. Moving in this direction will take many years but it would be helpful to begin considering what is desirable and feasible. SIG may wish to seek donor technical assistance for these reform elements – much of this agenda is not well addressed under existing programs of donor support. 23 Care and caution are required. Recent lessons from the case of Vanuatu are cited in O’Donnell and Turner (2005). 11 18. If one is to make use of these changes, then one must necessarily change the nature of the budget process and the presentation of the budget estimates. If allocations to line ministries are to be based on policy goals and the estimated costs of meeting those goals, then the old practice of setting incremental baselines would need to be phased out. If ministries are to focus on meeting policy goals, then the budget estimates should be revised to present information on activities and programs rather than inputs and, eventually, include indicators of results and impact. This could be done quickly by supplementing the existing format with new tables and charts where appropriate. 19. The pace of change will be governed by the pace of change within the line ministries because they hold the information needed by the coordinating ministries. The basic steps that the line ministries would need to take include: begin recording the actual cost of ministry activities and programs using the new Chart of Accounts; accumulate at least 2 years of data using new CoA codes; use the new data on actual costs to inform the unit costs of service provision; build capacity for monitoring and evaluation; and build capacity for better human resource management. Each step will take several years per ministry, so that the first set of pilot ministries would be ready for some simple form of policy based budgeting in ten year’s time, by 2021. Caveat: the proposed strategy assumes that all relevant ministry staff have the capacity to absorb additional training. This may not be true for all individuals in all ministries either due to educational constraints or time constraints arising from ongoing work obligations. Further investigation would be helpful in assessing whether the proposed strategy should be adjusted in light of such constraints. 20. Achieving government ownership and consensus among all key stakeholders will be very important to sustaining the proposed reforms because of the long time frame envisaged. Many of the necessary reforms would need to take place in MOFT although there is also much work to be done in the line ministries as well. The ongoing FEMSP is well placed to address MOFT needs although a broader program of Public Financial Management reforms may be needed for the line ministries as well, particularly for the accounting cadre that take on extra duties as budget and planning officers as well as procurement officers. 21. A specific sequence of steps is proposed for each of the strategy elements outlined above. Each sequence would progressively strengthen the budgetary capacity of the government. The strengthening would take place over a period of several years and start with just a few key ministries each year, expanding gradually to cover the entire government. More than one sequence and more than one objective could be pursued in any given year, depending upon the capacity and interest of those involved. SIG is well positioned to undertake these reforms given the extensive support already provided by RAMSI and other donors to MOFT. This now needs to be broadened to reach the line ministries as well. Annex B to this paper provides a step-by-step plan along with estimates on timing and references to available technical assistance. It is hoped that this design can be further improved through careful exploration and discussion with all stakeholders throughout SIG. 12 IV. PRIORITIES FOR 2011/12 22. The six most important steps required to put SIG on the reform path in 2011/12 are summarized below. These include alphanumeric references, such as A1, to specific steps found in Annex B. Two of these are already agreed and good progress has been made on a third. A1. Consider increasing central government responsiveness to citizen needs for non-devolved services by using the annual Premiere’s conference as an opportunity for the coordinating and line ministries to discuss how various provincial priorities could be addressed in the annual budget. B1. Begin recording the actual cost of selected ministry activities and programs using the new CoA. Similar work should proceed with the UNDP Development Assistance Database to capture donor financed spending in the recurrent and development budgets. (Underway.) C2. Establish a policy of building and maintaining a cushion of cash reserves to protect against unexpected shocks. (This is part of the program agreed with the IMF.) C4. Use periodic Public Expenditure Reviews to help better align selected budget allocations to line ministry needs. D2. Strengthen the quality of budget execution by tightening up expenditure controls. Change the legal framework to ensure no expenditures shall be made outside of approved budgets (ie, Cabinet Letters) and all Supplementary Budgets are fully financed from new revenues and grants or specific cost savings. This could also include training for the new Financial Instructions as well as work on commitment controls, virement, contingency warrants, supplemental budgets and procurement. 13 Annex A: Related Materials European Commission, Nov. 2008. Solomon Islands Public Financial Management Performance Report. IMF PFTAC, Aug. 2010. A Note on the New Financial Instructions and the Basis of Accounting- the Next Steps. IMF PFTAC, Aug. 2010. Solomon Islands PEFA Assessment 2008 and PFM Improvements Since. Regional Assistance Mission to Solomon Islands, May 2010. Financial and Economic Management Strengthening Program. World Bank, Sept. 2009. Solomon Islands Education Sector Public Expenditure Review. World Bank, June 2010. Health Financing Options. World Bank, June 2010. EXCEL Allocations and Expenditures Database. World Bank, July 2010. Two Alternative Formats for Budget Estimates. World Bank, Sept. 2010. Budget Planning and Formulation in Solomon Islands: A Review. World Bank, Oct. 2010. Downstream aspects of the Budget Management Framework Findings, Issues, Assessments, and Suggested Way Forward. World Bank, Sept. 2010. Assessing the Readiness in the Solomon Islands Government for Results-Based Monitoring and Evaluation to Support Results Oriented Budgeting. World Bank, Nov. 2010. A Good Budget Process Should … World Bank, May 2011. Briefing Note: Levels and Allocations Solomon Islands Public Expenditure Review. 14 Annex B: An Illustrative 10 Year Program of Budgetary Reforms for Consideration and Review PLANNING BUDGETING EXECUTION AND ACCOUNTABILITY A. Decide what is to be achieved and how you will C. Ensure ministries have the resources they D. Ensure they use those resources well know when you have succeeded need E. Hold people accountable for good results B. Make sure the line ministries know how to do each task and how much they cost A1(a). Consider increasing central government C1(a). In order to minimize the need for D1. Provide training for the SIG Accounting responsiveness to citizen needs by soliciting input spending reservations, strengthen MOFT Service on the new Financial Instructions as well from civil society organizations on how to improve capacity for aggregate fiscal forecasting including as on topics such as commitment controls, the quality of spending. From 2011 onwards. real GDP, inflation, and debt burdens: ongoing. virement, contingency warrants, supplemental Note: under-estimating expenditure can be as budgets and procurement. 1-2 years per A1(b). Consider increasing central government troublesome as over-estimating. Under- ministry, 2011-12 for pilot ministries. responsiveness to citizen needs for non-devolved estimates can motivate hasty and wasteful services by using the annual Premiere’s conference spending when actual revenue exceeds D2(a). Commence a twinning arrangement to as an opportunity for the coordinating and line expectations. build the capacity of a SIG official to undertake a ministries to discuss how various provincial review of the Public Finance and Audit Act 1978 priorities could be addressed in the annual budget. C1(b). In order to help keep public spending and future reviews. 2011 This should be coordinated with the UNDP within sustainable levels, complete a new debt Program of Provincial Government Strengthening. management strategy and reach agreement on D2(b). Strengthen the quality of budget From 2012 onwards. any amendments to the Honiara Club execution by tightening expenditure controls in Agreement. 2012 the Public Finance and Audit Act and the A2(a). Strengthen the capacity of government to supporting Financial Instructions. (See Annex coordinate policy initiatives through procedural C2. Build up a cushion of cash reserves to protect for details.) Drafting and submission to improvements in the coordinating ministries as against unexpected shocks: 2011 onward Parliament in 2012. well as OPMC. 5 years, 2012-16. Note: It would be useful to review with IMF whether a minimum target of 2 months of 2 D3. Strengthen the quality of budget execution A2(b). Strengthen the capacity of government to months of forward recurrent spending is too little by introducing through training the practice of coordinate policy initiatives by bringing in new in light of volatility of fuel import prices and gold establishing agreed budget implementation talent recruited through "graduate programs" prices. strategies with the line ministries. aimed at the OPMC offices responsible for the Note: this activity could be further strengthened Prime Minster, the Cabinet, and Policy Design and C3. Further reduce the need for spending by introducing procurement planning to line 15 PLANNING BUDGETING EXECUTION AND ACCOUNTABILITY A. Decide what is to be achieved and how you will C. Ensure ministries have the resources they D. Ensure they use those resources well know when you have succeeded need E. Hold people accountable for good results B. Make sure the line ministries know how to do each task and how much they cost Monitoring. reservations by better controlling the use of ministries. contingency warrants: 2012 onward A3. Strengthen the capacity of MDPAC & LMs to D4. Strengthen the quality of budget execution select projects into the budget that serve SIG C4. Use periodic Public Expenditure Reviews to by establishing tighter controls on spending objectives, including consultation with MOFT on help better align selected budget allocations to through upgrades to FMIS and Aurion that the sustainability of new O&M obligations: 4 line ministry needs: 2-3 ministries per year, 2012 prevent spending in excess of warrant years, 2014-17 onward. This can be done with or without authorities. 2011 and onward. donors. B1(a). Update the Chart of Accounts to enable SIG D5. Review the laws and Financial Instructions to begin tracking expenditures by programs, C5(a). To better align allocations for operations with the intention of replacing the portions activities, and districts. 2011 and maintenance with government assets referring to procurement with a comprehensive (buildings, vehicles etc) and projects, begin procurement law governing SIG, its provincial B1(b). Design and implement training courses for training ministries on how to produce forward governments, and its SOEs and establishing basic the implementation of the new CoA including estimates. protections against improper procurement, guidance on how to define programs and broad Note: This training will be more efficient with the sanctions, and systems of redress for complaints groups of activities. Provide extensive TA to benefit of program or activity data that would from ministries or suppliers. By 2013. ministries to assist them in accurately become with completion of step B3. implementing the CoA, and assigning line-items to E1. To ensure good data on actual spending is programs and/or activity-codes as well as district C5(b).To better align allocations for operations available to inform budget decisions and budget codes in policy-relevant ways. At least 1 week per and maintenance with government assets, use reviews, continue strengthening the FMIS ministry within 1 year per ministry, 2011/12 for improved information on attribution of capacity to produce reports comparing actual pilot ministries expenditure through the Development and planned spending for any year or quarter by Assistance Database and the CoA ministry, program, activity or district. Efforts B1(c). Begin recording the actual cost of ministry implementation process where adequate data is should include providing all line ministries with activities and programs using the new Chart of available. From 2016 onward. Prerequisites access to this data. Ongoing. Value of the data Accounts (CoA). 2 years per ministry, 2013-14 for include B2 plus prior work to identify domestic is increased with progress on B2. pilot ministries capital spending by object and program for both domestic and donor finance spending. E2. As actual data on spending against programs 16 PLANNING BUDGETING EXECUTION AND ACCOUNTABILITY A. Decide what is to be achieved and how you will C. Ensure ministries have the resources they D. Ensure they use those resources well know when you have succeeded need E. Hold people accountable for good results B. Make sure the line ministries know how to do each task and how much they cost B1(d). Similar work should proceed with the UNDP and/or activities becomes available in various Development Assistance Database to capture C6(a). To better align all budget allocations with ministries, begin introducing new tables in the donor financed spending inside and outside the needs, and to improve the degree of Budget Estimates with approved allocations and recurrent and development budgets. 2011 onward. consultation with line ministries, develop an actual spending listed by programs and activities analytical capacity within MOFT (PEAS). 2011 – in addition to the traditional tables with input B2. Accumulate at least 2 years of data using new onward. data. This will allow more accountability for CoA codes and DAD, 2 years per ministry, 2014-15 spending against recognizable goals and for pilot ministries. Prerequisite is B1. C6(b). To better align all budget allocations with activities. From 2016 onwards. Prerequisite is needs, and to improve the degree of B1. B3. In selected ministries, and where considered consultation with line ministries, develop a necessary to achieve a particular policy purpose, planning and budgeting capacity within selected E3(a). To strengthen the ability of PAC to hold use the new data on actual costs to inform the unit line ministries. ministries accountable for results, MOFT can costs of service provision: 3 years per ministry progressively provide PAC with more with at least 1 week of training per ministry per C7(c). To better align all budget allocations with information as and when needed. Ongoing. year, 2016-18 for pilot ministries, prerequisites are needs, and to improve the degree of B1 and B2 consultation with line ministries, revise the E3(b). To strengthen the ability of PAC to hold Financial Instructions in 2013 to introduce a fully ministries accountable for results, work with the consultative budget process, with a reasonable UNDP Parliamentary Strengthening Program to production cycle, beginning in 2014. The goal is progressively improve the depth and quality of to provide an opportunity for each ministry to annual PAC reviews of budgetary outcomes: 7 defend its budget submission, and for MOFT to years, 2012-18, prerequisite is E1. This effort examine that submission, as being reasonable should become especially rewarding with and consistent with all of its needs (including progress towards A2 and E2. personnel, other recurrent, and capital investments) rather than defending only the bids E4(a). Strengthen the legal basis for for new activities. Design and implement transparency and accountability in government training to support the change. operations. Consider the introduction of incentives for good results and sanctions for 17 PLANNING BUDGETING EXECUTION AND ACCOUNTABILITY A. Decide what is to be achieved and how you will C. Ensure ministries have the resources they D. Ensure they use those resources well know when you have succeeded need E. Hold people accountable for good results B. Make sure the line ministries know how to do each task and how much they cost C7(d). To better align all budget allocations with serious wrongdoing. 2 years, 2013-14. needs, and to improve the degree of consultation with line ministries, implement a E4(b). Study the merits and challenges fully contestable budget process. From 2014 associated with the introduction of model onward. Prerequisites are B1, C3, and E1. contracts for senior officials. This activity should begin no sooner than 2016 when pilot ministries have well defined and costed programs. E5. Start an educational campaign for media on what kinds of information are needed to assess budgetary spending on public service delivery and where to find it. 2014 onward. Prerequisites are B2 and E1. E6(a). To increase accountability to the citizens, begin informing district populations, by radio and community bulletin boards, of the central government allocations due for key public services. 2014 onward. E6(b). To increase accountability to the citizens, introduce various ways for communities to play a role in monitoring the government budget and its impact. Examples include community oversight committees (as in Sierra Leone) and citizen compacts (in Kenya). 5 years, 2014-18. 18