Technical Assistance Report-Implementing the Public Financial Management Bill

This Technical Assistance Report discusses recommendations for implementing the Public Financial Management (PFM) Bill in Uganda. The finalization of the PFM Bill has reached an advanced stage with Parliamentary approval expected in the coming months. The IMF mission strongly recommends that the petroleum revenue management provisions in the Bill are revisited. In particular this should include clearly defining the nature of the funds and its objectives. The reporting arrangements should also be clarified and the specialist audit arrangements specified. Finally, the Bill should be enhanced to ensure that investment risks are limited and to protect against its use as a development fund.


This Technical Assistance Report discusses recommendations for implementing the Public Financial Management (PFM) Bill in Uganda. The finalization of the PFM Bill has reached an advanced stage with Parliamentary approval expected in the coming months. The IMF mission strongly recommends that the petroleum revenue management provisions in the Bill are revisited. In particular this should include clearly defining the nature of the funds and its objectives. The reporting arrangements should also be clarified and the specialist audit arrangements specified. Finally, the Bill should be enhanced to ensure that investment risks are limited and to protect against its use as a development fund.

I. Implementing The Law-Development of Financial Regulations

A. Status of the Public Financial Management Bill

1. The mission reviewed a draft of the PFM Bill dated November 2013.1 An early draft of the Bill was developed with technical assistance from FAD. At the time of the mission, the latest draft of the Bill was subject to a detailed review between officials and the Parliamentary Committee. The Bill, as it is currently drafted, suffers from a number of conceptual and drafting weaknesses which will impact on the government’s ability to implement its provisions. These issues were summarized and provided to officials before the meetings with Parliamentary Committees. In the view of the mission, many of these issues cannot be fully addressed through regulation and may require further amendment. The most significant concerns are discussed below.

2. The institutional coverage of the Bill is not clearly expressed, either in definitions or in individual clauses of the Bill. Since the East African Monetary Union (EAMU) protocol currently envisages general government and eventually public sector reporting, careful thought should be given to transitional provisions for expanding the coverage of the provisions of the Bill. It is also unclear what institutions specific clauses apply to, as they are not clearly specified. It was agreed that this could be addressed by adding schedules to the Bill, with lists of entities by name for the various classes of entities defined, with a power for the Minister to amend and add to these.

3. The fiscal principles specified in the Bill currently mix principles and process requirements, and need amendment. The Bill should not elevate consistency with the National Development Plan (NDP) to a fiscal principle, with the NDP constraining the targets set in the Charter. Instead, the NDP should comply with the fiscal principles, and be constrained by the fiscal objectives in the Charter. The need for consistency can be included as a process requirement, although the NDP can quickly become out of date. Moreover, the requirement to raise revenues to fund government programs (rather than only fund programs affordable from revenues) is not a fiscally responsible principle.

4. The macroeconomic and fiscal policy related documents must be firmly anchored in the fiscal principles set out in the Bill. The charter of fiscal responsibility, budget framework paper, and half-year and annual reports need to explain how these principles are achieved. Reports on deviations also need to link back to the fiscal responsibility principles as well as the measurable fiscal objectives in the Charter. These issues should be further elaborated in regulations.

5. Provisions for virements, supplementary budgets, and the contingency fund will require further elaboration and detail in the regulations. The Bill, as it stands, would allow a budget to be implemented which is very different to that passed by Parliament. The unit of appropriation is too broad (vote level) and the virement rules permit large changes in expenditure. A virement can be up to 10 percent of a vote and this is not specified as 10 percent in total for each year, but can be understood as10 percent for each virement, an issue which should be addressed in the regulations. This is compounded by the ability to spend 10 percent above the level of a vote upon approval approved by the Minister. The latter can create very negative incentives for Ministers to lobby for additional funds and diminishes the role of Parliament in approving a budget, and this provision should be removed during the drafting process. These provisions should be clearly specified, in detail, in the regulations.

6. While supplementary budgets are not usually funded from a contingencies fund as required in the Bill, the provisions are designed to limit the historically high level of supplementary estimates in Uganda. Given the size of the contingencies fund, it is unlikely to be sufficient to fund the potential supplementary budgets based on the average 8 percent level of supplementary estimates in the past 3 years. However, if officials consider it is operationally feasible in future years, this may not be a significant issue. The regulations should clarify the operation of the fund, since the current drafting of provisions for the contingencies fund mix a contingency appropriation with a separate contingencies fund and therefore tend to mix fund accounting and reporting with reporting on a contingency appropriation.

7. The fragmented and in places repetitive reporting provisions are a significant issue in the Bill. For example, clauses 14 and 16 are partial in places, inconsistent, and unclear. The reports should be well specified in both the Bill and the regulations and instructions, and be clear as to who submits the reports to whom, by when and the publication requirements. The definition of the accounting standards to be used is unclear and should reflect Article 16 of the EAMU protocol on financial management, accounting and reporting, which states that partner states shall harmonize their financial accounting and reporting practices. Recent discussions within the EAC suggest that this is likely to be national standards based on International Public Sector Accounting Standards. The Bill could also set out appropriate transitional provisions to allow for implementation.2

8. The provisions in the Bill covering the Petroleum Revenue Fund have considerable issues. These are set out in detail in Section IV which includes governance, and definition regarding revenues and reporting arrangements.

9. A number of provisions in the draft Bill will require transitional provisions to enable a phased implementation of the Bill. The current requirement that the Treasury Memorandum for the preceding year is presented with the budget documents for the budget year is a good example. Since production of the Treasury Memorandum, an essential piece of the accountability loop is currently significantly delayed, measures will need to be taken to speed up this process and in the interim transitional provisions provided. Examples of other areas where transitional provisions may be required are:

  • To address the loose specification of institutional coverage of the Bill, entities to which the law applies should be included in a schedule to the Bill, classified appropriately with the timing for the inclusion of each class into the provisions of the law;

  • Accounting Officers’ production of financial statements within two months of the financial year-end, will require a financial year to implement;

  • Accountant General’s production of the consolidated annual financial statements by end-September of the following financial year; and

  • Establishment of Audit committees for all votes, will require increasing the number of Audit committees from one per sector (16) to one per vote, which will require significant effort and financial resources.

B. Overview of Regulations Outline

10. A new set of regulations will need to be developed to support operationalization of the Act, when it is passed. The new act will bring together macroeconomic and fiscal policy making as well as budget preparation, approval, and management. These were previously covered by the Budget Act and its associated instructions, and the more downstream budget execution and control, and accounting and audit requirements, which were covered by the regulations to the Public Finance and Accountability Act and its associated instructions. A completely new part of the PFM framework is the management of petroleum revenues. A proposed structure of the new regulations is briefly explained below. An outline is presented in Appendix I. The resulting outline is not intended to be exhaustive, however it is critical that the regulations, and supporting instruments, such as the agreement with the Bank of Uganda address to the extent possible the weaknesses in the Bill as outlined in this report. It is likely that further issues requiring regulation will arise as the policy and institutional implications of implementing the law are discussed during the actual drafting of the regulations.

11. The key headings proposed follow the structure of the PFM Bill. The regulations should identify those areas for which the MoFPED should develop further guidelines, instructions and manuals, and the process for consultation with the broader community. A summary of the proposed structure and contents is as follows:

Purpose and scope

This part of the regulations sets out a purpose of the regulations and elaborates on the key principles of the Constitution and the PFMB. It should be clear in the regulations, the application and effective date.

Macroeconomic and fiscal policies

This part of the regulations or where more appropriate, the Charter, should make provisions to elaborate on the indicators used to assess the alignment of the fiscal responsibility Charter and associated accountability documents, such as the budget framework paper, medium term debt strategy, and economic updates to the principles in the Bill. It requires definition of the minimum indicators, reports, and publication requirements. It defines the process and contents of the BFP.

Budget preparation, approval, and management

This part should set out the main principles, roles, responsibilities, and reporting requirements of the annual budget cycle, including timely production of entity level documents, ministerial policy statements, and annual budget estimates, in accordance with the government’s medium-term planning framework. It should also describe the rules leading to the establishment of budget ceilings as well as the process for developing, considering, and approving new budget spending proposals. The section should specify the regulations governing the execution of the budget, including grant of credit, budget release, multi-year commitments and commitment control, carry-over rules, virements, supplementary estimates, and in-year reporting.

Contingencies fund

This part should expand on the procedures for accessing the contingencies fund, including the definition of “disaster”, and who is responsible for declaring a disaster and authorizing release of the disaster portion of the fund.

Cash, asset, and liability management

This part should explain the operations of the Treasury Single Account (TSA), including the scope and coverage of the Consolidated fund (UCF). It should provide detail as to the conditions for setting up special funds. It should include provisions for opening, maintaining, and closing bank accounts as well as bank reconciliation to the general ledger. It should specify the rules, procedures and responsibilities for issuing loans, guarantees, and write-offs. Requirements for public-private partnerships, public investment planning and management, and non-financial assets should also be included here.

Accounting and audit

This part should include provisions related to the design, adoption, and maintenance of the chart of accounts. It should also regulate accounting issues not covered in the Bill, such as the responsibility for setting accounting standards and associated accounting policies. The obligation to provide access to accounting information and records may be prescribed here. On external audit, any provisions deemed necessary in the context of the PFM system, should be covered.

Petroleum revenue management

This part should specify and clarify the provisions in the law relating to the objectives of the fund, including an exhaustive definition of all revenues payable into the fund, the banking and accounting, audit, and reporting requirements, to address the lack of clarity and gaps in the Bill. It should enhance the minimum standards for qualifying investments, and risk management and reporting. It should also require the Investment Advisory Committee to prepare an investment policy in line with the regulations.

Roles and responsibilities

This part should set out responsibilities for all significant aspects of the public financial management system in Uganda. In line with the coverage of the law, it should define the responsibilities of accounting officers, chief executive officers of public corporations, state enterprises, and special funds as well as local government chief accounting officers.

C. Summary of Recommendations

12. The mission recommends that the MoFPED should:

  • Recommendation 1.1: Set up a drafting committee representing each functional area to guide the development of the regulations to the act as proposed;

  • Recommendation 1.2: Present regulations to Cabinet for approval; and

  • Recommendation 1.3: Conduct training on the changes to regulations to key PFM actors, once approved.

An outline for the regulations is set out in more detail in Appendix I.

II. Charter of Fiscal Responsibility

A. Introduction

13. The charter of fiscal responsibility plays a key role in the new Ugandan framework for managing the public finances. Well designed fiscal objectives or rules, specified in the Charter and then put into operation through the Budget Framework Paper (BFP) and annual budget process, can provide a useful check against a bias towards increased expenditure and deficits. This check will be crucial for Uganda as petroleum revenues begin to flow into the public finances. As importantly, a Charter which requires high levels of fiscal transparency in reporting on performance against those objectives, should improve the regular operation of fiscal policy by increasing the executive’s accountability to Parliament and the public, for good fiscal management.

B. Role of the Charter in the Fiscal Framework

14. The role of the Charter in the fiscal framework should be to:

  • Set enduring measurable fiscal objectives that will guide the detailed fiscal strategy set annually in the BFP, and constrain budget making from year-to-year. These objectives are the main mechanism translating the fiscal principles in the PFM Bill into concrete action.

  • Specify the detailed reporting arrangements against those objectives that will promote the maximum transparency for outside observers. That will enhance the credibility of those objectives as a constraint, both within the government and to outside observers.

Currently, the PFM Bill broadly achieves this aim, and is subject to the comments made on fiscal principles in the previous section.

C. Choice of Fiscal Objectives

15. The choice of fiscal objectives in the Charter needs to support fiscal discipline in a way that is appropriate to Uganda’s current circumstances. Those notably include the likelihood of a significant influx of oil revenues in the near future. The “resource curse” associated with these revenues in other countries is widely recognized, and is a key motivation for the authorities in establishing a more effective PFM framework. These revenues, along with Uganda’s currently relatively low debt level as a result of recent debt relief, also make the East African Community monetary union convergence criteria alone an inappropriate basis for setting fiscal objectives.

Deficit limits

16. Previous FAD advice3, on the basis of international experience with fiscal frameworks, suggested that measurable fiscal objectives are most effective and durable when they:

  • Can be clearly explained to policy-makers and assessed by the public (simplicity);

  • Provide an unambiguous guide for setting the annual fiscal stance (operability);

  • Allow fiscal policy to stabilize macroeconomic fluctuations (counter-cyclicality);

  • Facilitate multi-year expenditure planning, and minimize sudden and disruptive changes in the levels of taxation or expenditure (medium term); and

  • Ensure fairness between generations (sustainability).

17. Previous advice concluded that targeting a variant of the non-resource, non-grant or “domestic” balance would likely be appropriate in Uganda. The PFM Bill includes the principles that the fiscal balance without petroleum revenues should be maintained at a sustainable level over the medium term, and the government should maintain prudent and sustainable levels of public debt. The key measurable fiscal objectives in the Charter should operationalize these principles.4

18. Placing a clear annual ceiling on the non-oil, non-grant deficit as a share of GDP is simple, and will help promote fiscal discipline. As elsewhere in the world, there is a trade-off between the simplicity and operability of the rule, and the extent to which it can allow counter-cyclicality of fiscal policy.

19. An annual ceiling does not, of itself, take account of cyclical developments in the economy. The absence of a sufficient planning buffer against the ceiling, could result in either pro-cyclical fiscal policy or a frequent resort to the deviation clause for suspending the Charter in the PFM Bill. This could weaken the credibility of the Charter as a constraint on policymakers. Typically this problem is dealt with either through:

  • The use of a cyclically-adjusted fiscal balance (e.g., in Chile and Norway). The drawback of this approach is the complexity and potential lack of transparency in a “black box” cyclical adjustment; or

  • By specifying that the target operates over a pre-defined medium term (e.g., at a three-year horizon). The drawback of this approach is that it does not place an unambiguous constraint on expenditure in the most important year, the financial year to be decided in the upcoming annual budget. It would permanently allow a deficit above the ceiling, so long as later years expected it to be corrected. This deprives the finance ministry of an important card in negotiating overall expenditure levels, and also means that overspending in-year can never breach the target.

20. An alternative is to specify an annual limit, but with a pre-defined two- or three-year correction period where unexpected revenue shortfalls, as a result of a cyclical downturn, take the deficit above the limit. In that event, there would be no suspension of the Charter under the PFM Act, as long as the pre-defined correction was planned. If, however, the excess deficit was a result of increased spending, for example as a result of a national emergency or major disaster, then the deviation clause in the PFM Act would need to be brought into effect. This formulation provides unambiguous guidance on the next year’s annual fiscal stance, so long as the rule is currently being met, but avoids the rule being excessively pro-cyclical.

21. An annual limit does not provide flexibility for the government to exceed the ceiling in a given year or years, if it believes that there are investment or development projects that are of sufficiently high value to justify a higher deficit. The EAC are considering whether a category of such exceptional infrastructure projects could be defined and an exemption allowed. We would not recommend including such an exemption in the deficit objective, because of the potential for its exploitation as a “loophole” in the framework. Instead, the level of the deficit ceiling should be set with the balance of development spending and fiscal sustainability in mind, and potential projects prioritized and sequenced over the medium term to live within that ceiling.

22. The level of the non-petroleum deficit limit is a crucial medium-term fiscal decision. Given constraints on the level of public debt, it effectively defines the pace at which Uganda’s natural resource assets will be depleted through transfers from the Petroleum Fund. The fiscal principles in the PFM Bill require the government to manage petroleum resources and other finite natural resources for the benefit of current and future generations. Previous FAD advice5 recommended that the level of the target be set by anchoring it to the principle that the estimated benefits of resource revenue and external grants be kept constant over a fixed 40 or 60 year “catch up” period—i.e., constant between the next 2 or 3 generations. This remains a good principle for determining the target in the Charter.

23. A ceiling on the non-petroleum deficit set in this way should be the primary fiscal objective guiding policy setting. However, for the period prior to significant oil revenue coming on-stream, it is likely to be inappropriately loose. It will effectively allow the government to borrow more now in anticipation of revenues that have not yet materialized, and whose size and timing is inherently volatile and uncertain. Consequently, the government should consider, at least in the first Charter adopted, supplementing it with a ceiling consistent with the EAC monetary union deficit limit, applied in the same manner. That would support prudent management of oil revenues by preventing excessive borrowing in anticipation of their arrival, and help support convergence with the EAC limits.

Debt limits

24. Measurable fiscal objectives on the level of public debt should operationalize the fiscal principle in the PFM Bill that debt must be maintained at a prudent and sustainable level. Adopting in the Charter a hard annual ceiling on gross public debt, potentially making reference to the ceiling in the EAC convergence criteria, is a natural way of doing so. However, given the tendency for public debt levels to be very volatile in the face of a shock, and the headroom Uganda currently has against the 50 percent EAC ceiling, we recommend targeting fiscal policy to maintain a buffer beneath that hard debt ceiling. This can be done by setting a lower target ceiling for public debt, at the end of the five-year forecast horizon, along with an additional requirement that public debt be projected to fall as a share of GDP by that point.

25. The level of the target ceiling needs to be chosen carefully, and in light of decisions on the deficit ceiling. However, it should provide a sufficiently large buffer to allow space for the public finances to absorb a large shock without breaching the 50 percent ceiling. It should also be set at a level that prevents fiscal planning being based on sharply rising debt ratios over the short to medium term. The government should consider whether the lower target should be set on nominal debt as a share of GDP, rather than in NPV terms. This would be a simpler and more transparent measure.

Supplementary targets

26. Supplementary targets, in addition to these primary deficit and debt targets, could further encourage sound management of the public finances. Particular areas where the government has been considering targets in the Charter are:

  • Efforts to improve revenue collection; and

  • Explicit limitations on withdrawals from the Petroleum Fund, including linking the funds to investment or infrastructure expenditures.

27. The government’s aim to increase the revenue to GDP ratio over time, from its current very low levels, could be included as an objective in the Charter. This would help hold the government to account. However, the authorities should be cautious in including this target in the Charter on two counts:

  • They should avoid a situation where fiscal and expenditure plans are set on the basis of assumed improvements in revenue collection in line with a fixed target. Failures to improve collection—an item not fully in control of the revenue authorities—in line with the target will result in expectations of expenditure that are too high to be financed from actually available revenue, putting the deficit ceiling at risk.

  • Including a suitably stretching numerical target in the Charter on revenue to GDP increases will mean a significant risk that the authorities fall short of it in any given year. That could mean frequent invocation of the deviation clause in the law suspending the Charter, again, damaging its credibility.

28. If the Charter contains a target on improved revenue collection, it should explicitly require the revenue forecasts used in the BFP and budget process to be prudent and realistic. They should be based not on the target but only on (i) concrete and properly costed tax policy measures to increase revenue and (ii) realistic assumptions about success in improved revenue collection. Revenue forecasts should not be based simply on the assumption that the target will be met. The government should instead consider adopting a fixed cautious forecasting assumption, that improvement in revenue collection and reductions in the tax gap should only be included in forward projections when they have been successfully realized in outturn figures (see below, paragraph 34 for further discussion of revenue forecasting in the Charter). Any additional revenues realized above forecast should be subject to the supplementary budget provisions as provided for in the draft Bill and supporting regulations.

29. The primary constraint on excessively front-loaded spending of oil resources should be the deficit ceiling calculated excluding petroleum revenues. However, an additional constraint in the Charter on withdrawals from the Petroleum Fund and/or the kinds of expenditure items that can be financed by them could potentially be a helpful addition. There are a range of forms such a target could take, including:

  • A cash ceiling on the amount of resources that can be appropriated from the Petroleum Fund in each financial year. This would determine the split between Petroleum resources and government debt that finances the non-oil deficit. It could be a fixed amount, or linked to the level of in-year revenues so as to ensure a steady build up of the fund for future generations;

  • A specific floor on government investment spending, helping to ensure that the resources funded by the higher non-oil deficit do not leak into recurrent expenditure; and

  • A less tightly drawn link requiring that the government only appropriate resources from the Petroleum Fund, where they are to be used to finance capital investment. Given that cash is fungible, this target would be largely presentational. However, it could still potentially be helpful in establishing in the public mind the need for oil resources to be managed sensibly for the benefit of future generations.

Definition of fiscal aggregates

30. The coverage and definition of fiscal aggregates used to measure fiscal objectives in the Charter must be clearly specified, and as far as possible determined independently of the government. This reduces the potential pressure to engage in “creative accounting” of expenditure to move it outside the deficit ceilings. This is important to ensure the transparency and external credibility of the constraint they are designed to impose.

31. Targets should be based on fiscal aggregates, compiled using an international standard such as SNA95 or GFSM 2001. Where possible, they should be based on consolidated general government, with prudent forecasting of those elements of the consolidated general government not directly controlled through the annual budget. Concepts such as “petroleum revenues” and “grants” should be clearly defined and their data source specified. In the case of petroleum revenues, this can be by reference to the definition and reporting in the PFM Bill.

D. Specifying the Operation of Fiscal Policy in the Charter

32. The Charter should set out and explain the basic operation of fiscal policy to support delivery of the objectives. It should explain the purpose of the timetable and processes for making fiscal policy set out in the PFM Bill.

Top-down budgeting

33. The Charter should clearly enshrine the top-down budgeting process which is implied by the provisions in the PFM Bill, and which is designed to further fiscal control and proper prioritization of expenditure.6 In particular, it should explain and set out the purpose of:

  • The Charter first setting measurable fiscal objectives to operationalize the principles set out in the PFM Bill;

  • The BFP, next, establishing a set of medium-term expenditure ceilings, for each of the next five financial years based on the objectives in the Charter and a prudent macro-economic and fiscal forecast;

  • The annual budget, then, allocating funds to particular sectors for the coming financial year within the expenditure ceiling fixed in the BFP;

  • Sector ministries, then, prioritizing the use of their resources within their fixed budgets as approved by Parliament; and

  • Proper reporting and accountability on the execution of those budgets.

Prudent revenue forecasting

34. For a framework based on measurable fiscal objectives and a medium-term expenditure framework to be effective in improving fiscal control, it is crucial that revenue forecasting is prudent and realistic. Persistent over-optimism on revenue forecasts significantly damages the credibility of such frameworks. As a prospective oil-exporter, Uganda will face particular challenges. Added to the inherent volatility of oil prices and uncertainty of production volumes, is the temptation toward optimism on these prospects as a means of increasing the planning envelope for future budgets.

35. The Charter should establish mechanisms to help ensure fiscal policy is based on prudent revenue estimates. At a minimum, this should include the requirement that forecasts be based only on announced revenue policies, transparently published revenue forecast determinants linked to the macroeconomic forecast, and a cautious approach to anticipating potential improvements in revenue collection. Increased transparency over the assumptions behind the forecast, and a clear presentation of the forecasting record can help correct and control forecast biases. This should be a central requirement of the Charter (see Section E below).

36. Further, the Charter should require a comprehensive and quantified fiscal risk statement. This should show both the sensitivity of the fiscal balances to alternative macroeconomic assumptions, including in particular alternative assumptions over oil prices and revenues, but also the impact of the realization of discrete fiscal risks. These should include guaranteed loans, state-owned enterprises, local governments, PPPs, and tax expenditures. Box 1 suggests contents for a fiscal risk statement in line with those discussed in the context of the East African Monetary Union.

Uganda: Suggested Contents of a Fiscal Risk Statement

Macro-Fiscal Sensitivity Analysis

Discussion of the macroeconomic forecasting record in recent years, comparing the assumptions used in budget forecasts against actual outcomes.

Sensitivity of aggregate revenues and expenditures to variations in each of the key economic assumptions on which the budget is based (e.g., impact of exchange rates and interest rates on revenues and expenditures), with explanation of underlying mechanisms. Possible methods and presentational devices include alternative scenarios or fan charts. In conducting these exercises, it is desirable to take into account the correlations among different shocks.

Public Debt Sustainability Analysis

Sensitivity of public debt levels and debt servicing costs to variations in assumptions regarding e.g., exchange rates and interest rates. Impact of debt management strategy on the government’s risk exposure.

Policy and institutional framework for government borrowing and on-lending: projected statement of inflows, outflows, and balances; and disposition of loan repayments and nonperforming loans.

Specific Fiscal Risks

Discussion of discrete sources of risk to the government’s fiscal forecast, including:

Contingent Liabilities: Expected value and government’s gross exposure to contingent liabilities—especially central government guarantees (e.g., to public enterprises); reporting to include broad groups of guarantees, but also any major individual guarantees. Rationale and criteria for the provision of guarantees.

Banking sector: Deposit insurance schemes and—to the extent that the authorities feel this does not generate moral hazard—risks from the banking sector. Information on costs of past bailouts/recapitalizations/preemptive financial support.

Legal action against the central government: Past claims (including amounts) and the face value of current claims, including a disclaimer that reporting the risk does not indicate government acknowledgement of liability.

Natural Disasters: Fiscal impact of disasters in recent years. Level and operation of possible contingency reserve for natural disasters (if applicable).

Sub-National Governments

Legal framework for intergovernmental fiscal relations, and summary of recent aggregate sub national government financial performance and financial position.

State-Owned Enterprises

Policy framework for SOEs (pricing policy and dividend policy). Financial performance and position of the SOE sector and the largest SOEs.

37. Finally, the government should consider further policy options that can support the realism and credibility of forecasts. This could include:

  • Pre-committing to the use of specific forecasting assumptions designed to be prudent. Examples include using historical averages or external consensus forecasts for oil prices, and basing revenue projections on the assumption of flat rather than improving tax gaps;

  • Developing a deliberately cautious forecast for planning purposes, set below the government’s central forecast;

  • Requiring external validation/auditing of key forecast assumptions, for example by the Parliamentary Budget Office or the Economic Policy Research Council; and

  • Explicitly comparing forecasts with those of other institutions.

E. Reporting and Accountability

38. The Charter should specify the timetable and minimum content of reporting on performance against fiscal targets, to ensure it remains transparent and the government is accountable.

Timetable for reporting

39. The timetable for reporting should be clear, based on the provisions in the PFM Act, and avoid excessive updating of fiscal forecasts. The government should produce full macroeconomic and fiscal forecasts and report on performance against the objectives in the Charter in the following documents:

  • The annual Budget Framework Paper, published by the end of March.

  • A Half-Year Fiscal and Economic Update, published no later than the end of October.

  • A Full-Year Final Report on Fiscal and Economic Outturns. This should be published when final outturn data on the fiscal aggregates is available. Ideally it should be timed to coincide with (or form part of) the annual Budget Framework Paper.

40. The current reporting timetable could be streamlined. At present the PFM Bill (section 14 and 16) requires economic and fiscal updates by the end of February and the end of October. To reduce the burden on staff producing economic and fiscal forecasts, the first of these public updates should be aligned with the BFP projections (approved by Parliament by end-February, section 9).

Content of reports

41. The Charter should specify minimum content for reports. The PFM Act specifies basic requirements, but the Charter can and should enlarge the list as:

  • The aggregates that have to be explicitly forecasted and published will depend on the targets set in Charter; and

  • Forecasting approaches can evolve, with particular issues rising and falling in importance over time (as for example forecasting of oil revenues will soon become crucial to Uganda’s fiscal forecasts). Updating the specific requirements in the Charter allows more flexibility than including them in the Act, but still provides the necessary assurance of continued transparency.

42. The mission discussed the contents of the Charter with a working group of officials. The draft Charter in Appendix II (attached as a separate document) sets out a specification for minimum requirements for the forecast. It is particularly important that the specified reporting on oil revenues and future production, and forecasts for public debt and its future evolution, are included.

43. The Charter should also specify some key analytical tables, to help Parliament and the public hold the government to account for performance against the objectives. These should include:

  • Comparisons to outside, independent forecasts;

  • Comparisons to previous official forecasts, with explanations for changes separating them into policy changes, changes to economic determinants, and other changes;

  • Transparent tables setting out the key fiscal determinants, and their changes since the previous forecast, to assist in understanding reasons for revenue deviations;

  • Data on budget execution; and

  • Sensitivity analysis and alternative scenarios for the fiscal aggregates under different assumptions from the baseline, as part of the fiscal risk statement discussed above (paragraph 37).

F. Summary of Recommendations

44. The mission makes the following recommendations:

  • Recommendation 2.1: We recommend placing an annual limit on the non-resource, non-grant fiscal deficit, with a pre-defined two-year correction period where the limit is breached as a result of unexpected revenue shortfalls.

  • Recommendation 2.2: In setting a target for the non-resource, non-grant balance the government should consider adopting the principle that it should keep the estimated benefits of oil revenues and grants constant between generations.

  • Recommendation 2.3: The Charter should adopt a hard ceiling on gross public debt, potentially making reference to the EAMU convergence criteria, applying in every financial year. However, it should also include a lower target level for public debt over the medium term.

  • Recommendation 2.4: If the Charter contains a target on improved revenue collection, it should explicitly require the revenue forecasts used in the BFP and budget process to be based not on the target, but only on (i) concrete and properly costed tax policy measures to increase revenue and (ii) realistic assumptions about success in improved revenue collection.

  • Recommendation 2.5: The government should consider potential forms for an objective on withdrawals from the Petroleum Fund, but be careful to ensure such an objective does not unintentionally undermine the main deficit constraint on excessive front-loading of their usage.

  • Recommendation 2.6: Targets should be based on fiscal aggregates compiled using an international standard such as SNA95 or GFSM 2001.

  • Recommendation 2.7: The Charter should establish institutional mechanisms to help ensure fiscal policy is based on prudent revenue estimates.

  • Recommendation 2.8: We recommend the government produce full macroeconomic and fiscal forecasts and report on performance against the fiscal objectives twice a year, in the BFP and a half-year update. The annual budget should be based on the BFP forecasts.

  • Recommendation 2.9: The Charter should specify the minimum content for fiscal reports, in line with the draft Charter included in Appendix II.

III. Petroleum Revenue Management

A. Introduction

45. Provisions for the Petroleum Fund have been made in the PFM Bill. There are weaknesses in the governance and accountability provisions when compared to the Santiago Principles for Wealth Funds.7 These are commented on in Appendix III. The main issues are set out below.

46. Proceeding with the Bill as it is currently drafted will present considerable risks. The issues with this part of the Bill are extensive and include problems that may undermine the potential effectiveness of public finances. The Bill is still being discussed and changed in the parliamentary committee process currently underway. It is important that the version that emerges from that process is tested against the Santiago principles and the detailed list of issues in Appendix III. The final version should deal with the issues raised, given that many of the issues cannot be dealt with by regulation.

Objectives for the Petroleum Fund

47. There are no objectives for the Petroleum Fund. In the absence of these, there is little guidance for the government in making decisions about the transfers to the annual budget, apart from the fiscal principles which are currently not well specified nor properly linked to the decision making provisions in the Bill and any implicit restrictions placed by the Charter (as noted in Appendix II). Even when improved, these principles apply to the management of all the government monies, while the Petroleum Fund should have its own objectives.

48. An example of a suitable objective that could be included in the PFM Bill is as follows. The Petroleum Fund is to ensure that petroleum revenues are used for the benefit of both current and future generations. This is done by providing stable and sustainable disbursements to the central government budget through the Consolidated Fund, and investing the remaining revenue for utilization in future budgets.8

The nature of the Petroleum Fund

49. The exact nature of the Petroleum Fund is unclear due to a lack of precision in the definitions part of the Bill and various clauses.9 The Bill states that petroleum revenues may only be used for the financing of infrastructure and development projects, and not recurrent expenditures, but this does not define the purpose of the Fund. This lack of clarity was confirmed in discussions with officials who had different understandings of the Petroleum Fund as being (i) a fund that holds the Petroleum Revenues transferred from the Uganda Revenue Authority (URA) and an Investment Portfolio managed by the BoU; or (ii) only the investment portfolio managed by the BoU. While regulations can deal with banking and transfer arrangements, they cannot resolve the confusion in the Bill over the nature of the Fund.

50. The arrangements in the PFM Bill lack the protections of having an independent and well-constituted governance board tasked with achieving the objectives. Instead the governance function sits with the Minister of Finance who is required to: make the agreement with the BoU to manage the funds, appoint the Investment Advisory Committee, approve the Investment Policy, monitor the performance of the BoU with regard to the Fund (including thought the reports of the BoU and the Investment Advisory Committee), monitor the Accountant General/PS/ST/BoU with regard to transactions relating to the Fund, approve the Annual Plan for the Fund, and approve the Annual Report of the Fund.

51. The Minister’s roles are extensive and time consuming, and may present tensions between the short term budget demands and those of the Fund. If some of these roles are delegated, this may place too much responsibility on officials who may have official roles that conflict with the, currently unspecified, objectives of the Fund. For example, there is an obvious tension between meeting short-term budget demands and the longer-term objective that should be applied to the Fund of preserving wealth for future generations.

Revenues from investments

52. The PFM Bill does not require all revenues, income, dividends, proceeds, gains, earnings, or other moneys or assets received from the investment of Petroleum Fund to be retained within and constituted as a part of the Petroleum Fund. These revenues should include: (a) interest on deposits and securities; (b) dividends; (c) proceeds from the sale of stocks and shares; and (d) earning from derivative transactions.

Investment Advisory Committee and investments

53. The Investment Advisory Committee is not constituted in a way to ensure that it fulfills its functions. Instead of consisting of experts on investment policy, markets, and performance, two of the five positions are allocated to the Secretary to the Treasury and the Permanent Secretary of the Ministry responsible for Petroleum activities, or their representatives. It is crucial that positions are filled by representatives with the “substantial experience, training, and expertise in financial investments, portfolio management, or investment law” that is required for each member by clause 65 (4). This will inevitably imply that individuals with appropriate skills will need to be recruited, rather than sourced from civil servants within the respective ministries. Established civil servants may also have responsibilities that conflict with the objectives of the Fund, such as pressures to invest in development projects or local investments that may expose the resources of the Petroleum Fund to unacceptable levels of risk.10

54. The prohibitions on risky investments in clause 61 of the PFM Bill are too limited and should be extended, so that the Investment Advisory Committee cannot include them in the investment policy to be approved by the Minister. The Bill should specify the acceptable issuers of an internationally convertible currency deposit or a debt instrument denominated in any internationally convertible currency. There should also be a requirement to have an investment grade rating for the central banks of other countries when investing in their interest bearing instruments. A limitation could be specified in the percentage of the portfolio that can be in securities and the acceptable stock exchanges for those securities. Further limitations could be placed on the use of derivatives.

55. A prohibition can be added that Petroleum Fund resources shall not be invested in any instrument issued by a bank, corporation, or individual resident in Uganda or owned or controlled by a Ugandan national or company registered in Uganda. This would help protect the Petroleum Fund from being used as a development fund, as the development aspects can occur through the transfers from the Petroleum Fund appropriated each year as part of the annual budget, while the Fund itself is protected by suitable investment provision to preserve it for future generations.


56. Protections from unauthorized withdrawals are missing. This could be achieved by requiring that the funds that are withdrawn in breach of clause 57 shall be treated as losses of public money. The PFM Bill is not adequate in its treatment of losses and unless other laws cover this to permit timely recovery of losses from civil servants, public office holders, and private sector individuals and entities, then the Bill should be strengthened by adding suitable provisions.

Planning, reporting, and auditing requirements

57. The planning and reporting provisions should be complementary and well specified. This is not the position in the PFM Bill. There is an annual plan by the BoU for the petroleum revenue investments, which is approved by the Minister and Parliament (clause 69). There are three reporting authorities: the BoU which prepares semiannual and annual reports including financial statements for the investment funds (clause 67 and 70); the Accountant General who prepares semiannual and annual accounts for the Fund (clause 59); and the Minister who reports to Parliament on transfers, sources, volumes, and value of petroleum revenues and who tables the financial statements (clause 60).

58. If the Fund was constituted as an institution with a governing board, there could be an Annual Plan from the Board and an Annual Report that covered the performance of the Fund against its objectives and included the results of the investment portfolio. The BoU would report to the Board and the public on the performance of the investment portfolio and flows in and out of the investment portfolio. The Accountant General would report on the transactions relating to the Consolidated Fund, for use in funding the annual budget.

59. Since there is to be no governing board, as envisaged in the current draft of the Bill, the current planning and reporting provisions will need to be expanded in the regulations. The annual report on the Petroleum Fund prepared by the Accountant General should include the requirements that are currently applied to the Minister’s report, and should include other requirements such as an explanation of how the Fund is performing against its objective. The Minister could table this report in Parliament.

60. The PFM Bill does not require the BoU to report against suitable benchmarks. The publicly available monthly reports should be against benchmarks, with more comprehensive benchmarking information in the BoU’s annual report. The specific benchmarks can be set in the investment policy and/or in regulation, but the requirement to have them should be in the Bill.

61. The audit provisions in the PFM Bill need to be strengthened to reflect the specialist nature of audit of investment funds. There should be a requirement for the petroleum revenue investment funds managed by the BoU to be audited by an audit committee of independent and highly qualified members, with reports to be periodic and annual, and publicly available. This is specialized auditing to give assurance about the operations of the investment portfolio, and would be in addition to the statutory external audit by the Auditor General or his agents.

Other provisions

62. The mechanisms for transfer of the seven percent of petroleum revenues specified in the Bill to local governments are inflexible, should forecasts be inaccurate. The transfers are appropriated in the annual budget based on forecasts of the petroleum revenues. There is no mechanism in the Bill for dealing with situations where the actual revenues vary from the forecast revenues. The regulations should include the process for dealing with significant variances, if these cannot be absorbed by the central government.

63. The PFM Bill provides for a Petroleum Pricing and Valuation Committee, but does not include essential requirements regarding this committee. At a minimum, the PFM Bill should set out the membership, appointment provisions, terms and conditions for members, functions, powers and reporting requirements. If it is simply an internal government advisory committee made up of public officials, then it may not need to be specified in the Bill and could be dealt with in regulations.

B. Implementation Issues

64. The implementation of the Petroleum Fund will require the development of committees, a contract with the BoU, policies, and changes to systems and processes and other capabilities. Table 2 notes the developments required by the MoFPED and comments on the resources and timeframes. The BoU will also have considerable development work to do, to set itself up to fulfill the function of the manager of the investment funds. The Revenue Authority and the ministry responsible for petroleum will also have development work, which is not covered in this report.

Table 2.

Uganda: Activities to Set Up the Petroleum Fund

article image

C. Summary of Recommendations

65. The PFM Bill could be amended to:

  • Recommendation 3.1: Specify the objectives of the Fund;

  • Recommendation 3.2: Fully define the revenues of the Fund to include the earnings on investments;

  • Recommendation 3.3: Remove the civil servant positions designated by office on the Investment Advisory Committee and add the requirement that appointees are not to have conflicts of interest with their functions;

  • Recommendation 3.4: Extend the prohibited investments to ensure the fund is adequately protected (as suggested above);

  • Recommendation 3.5: Specify that withdrawals that are not in accordance with the law are to be treated as public losses and ensure that there are adequate provisions in the law to recover such losses;

  • Recommendation 3.6: Improve and further define in regulations, the planning and reporting requirements so there are suitable and consistent requirements including rationalizing the Accountant General’s and Minister’s reports and making the contents of the BoU report more suited to investment portfolio reporting, including benchmark reporting;

  • Recommendation 3.7: Provide for suitable auditing provisions for the investment portfolio;

  • Recommendation 3.8: Provide for dealing with differences in the appropriations to local government and the actual revenues of the fund or deal with this in regulations;

  • Recommendation 3.9: Set out the basic provisions for the Petroleum Pricing and Valuation Committee with the detail to be covered by regulations if necessary;

  • Recommendation 3.10: Ensure that the drafting issues and errors in the Bill are addressed before the Bill is passed into law (see Appendix I);

  • Recommendation 3.11: Plan the implementation of the Fund and related matters; and

  • Recommendation 3.12: Develop supporting regulations once the Bill is finalized so the regulations can be tightly linked to the provisions of the forthcoming Act.

IV. Cash Management and Treasury Single Account

A. Introduction

66. Good progress has been achieved since June 2013, providing a sound platform for extending the TSA and for introducing cash management reforms.11 The bank account structure has been revised to include a UCF Account and decentralized zero-balanced Vote sub-accounts to facilitate payments. All UCF revenues are credited to the UCF Account and spending limits distributed from it on a quarterly basis to a UCF Holding Account. Cash is transferred to the Vote sub-accounts based on payment listings generated through the IFMS and prior to the payments being finalized. The IFMS-generated payments are made by electronic fund transfers on the same day as the cash transfers; any close of business balance on the Vote sub-account is swept back to the UCF Holding Account.

67. Restructuring the government’s bank accounts has benefitted from complementary initiatives. These have comprised several technical and process changes, including: enhanced IFMS functionality to support the revised banking arrangements and security functions; reconfiguring the core banking application to support the required bank account structure and the GoU requirements for transaction processing; payroll processing decentralized to the accounting officers; changes in the budget fund release process; and improved interfaces enabling straight through processing (STP) for domestic payments. Implementing the restructure also benefitted from extensive change management activities including the wide dissemination of a TSA concept paper and TSA guidelines, the delivery of associated training, the closure of dormant and redundant bank accounts, and the appointment of a US Treasury resident advisor to provide ongoing technical assistance.

68. The MoFPED and the BoU cite a number of tangible improvements resulting from the combination of changes. In particular, these include enhancements in the security and efficiency of transactional processing for payroll and supplier payments, timelier reconciliation of a smaller number of bank accounts, and improved planning for service delivery from quarterly fund releases. However, realizing benefits from aggregating government cash balances and moving from the current “cash rationing” basis for fund release will take longer.

69. The MoFPED is keen to build on the current momentum to extend the TSA arrangement. Amongst its immediate ambitions are plans to: expand the TSA coverage to include project and other funds; further reduce and rationalize the number and location of operational bank accounts; complete the decentralization of the payroll; further enhance the IFMS/core banking functionality and interfacing, to enable STP of foreign payment transactions; formalize the relationships between the GoU and BoU in a series of service level agreements (SLAs); and commence building cash management capacity. The remainder of this section concentrates on providing guidance for expanding the TSA.

B. Implementation Issues

70. Extending the TSA to encompass project funds will need to be accompanied by changes to the TSA structure and additional change management measures. Modifications will be required to address constraints in the TSA’s current design especially to: recognize the role of the UCF and its relationship to the aggregate TSA balance; include alternative TSA tools for managing the cash pertaining to projects, donor, and extra-budgetary funds; and review the efficacy of the daily advances to the operational Vote sub-accounts. Change activities will need to identify and cater for the requirements of development partners in accounting for the receipt and application of project funds. Box 2 summarizes international experience from Mozambique on incorporating projects into the TSA; the mission has drawn upon those lessons learned to suggest the TSA design and the processes necessary for expanding the TSA coverage to donor projects.

Mozambique’s Experience of Extending the Treasury Single Account to Projects

Mozambique is heavily dependent on external assistance with approximately 35 percent of its budget funded through external sources, much of which is project-related. After establishing a core TSA, Mozambique sought to expand its coverage to include the balances and transactions of externally funded project accounts.

Currently, foreign currency project funds are aggregated in separate TSAs, denominated in three major currencies: US dollar, Euros, and South African Rand. Foreign currency project expenditures are transacted through the IFMIS (referred to in Mozambique as the e-SISTAFE) and the foreign currency TSAs. Foreign funds converted for local spending are credited to the main TSA denominated in Mozambique currency (MZM).

Significant success has been achieved in expanding Mozambique’s TSA in this way with the TSA coverage increasing from less than 20 percent of externally funded budget execution in 2008, to 47 percent in 2013. A number of features were key in persuading development partners to channel project funding through the TSA. These include:

  • A robust accounting system supporting the TSA and satisfying the development partners’ accountability and reporting requirements;

  • Effective change management to convince development partners of the combined abilities of the IFMIS and TSA to ensure their project funds can only be applied to the intended purposes, will be available when required and subject to a full accountability framework;

  • Piloting to ensure the combined TSA and IFMIS functionality meets donor requirements, and allow for subsequent system modifications to include additional analysis and reporting;

  • A well-structured chart of account including a Source of Fund classification segment which, when combined with other segments, enables donor funds to be ring-fenced for particular purposes;

  • A flexible project-specific classification segment allowing donors and project managers discretion to introduce unique values to support donor-specific reporting requirements;

  • Functionality for executing foreign currency transactions through the IFMIS and the appropriate foreign currency TSA, improving the efficiency and direct accounting for transactions, and reducing the risk of exchange rate variation; and

  • Revisions in budget management practices supporting the year-end rollover of unutilized donor-funded budgets, recognizing the impact of significant exchange rate variations, and allowing the project/program manager greater discretion in executing the budget (e.g., less stringent expenditure controls against line items, whilst constraining aggregate expenditure for the project).

Despite these efforts, some donor-funded projects remain outside TSA. The main reasons being:

  • The development partner has yet to complete a fiduciary risk assessment of the PFM systems;

  • Existing financing agreements require compliance with donor specific practices; and

  • Donor focal points lack understanding of the functioning of the government PFM system and perceive the use of in-country systems as weakening their control over project funds.

The government’s reform program aims to expand the proportion of external funding passing through the TSA and IFMIS. Planned activities involve a range of change management and communication initiatives to provide impartial assessments of the benefits and risks to all parties, and to be responsive to donor concerns and requirements.

The Treasury Single Account design

71. The current TSA configuration incorporates some sound TSA features. These include the use of a hierarchical set of linked accounts (or sub-accounts) enabling a view of the UCF aggregated balance. Whilst this structure may be suited to the first step of establishing a TSA—i.e., rationalizing the UCF domestically funded banking arrangements— additional features will be required to enable the TSA’s expansion as the means to aggregate and manage all public funds. In particular, new design features will be required to: (a) ring-fence specific funds; and (b) aggregate all public funds whilst recognizing the unique role of the UCF and the requirements of other public funds. The design should also reconsider the mechanism for pre-funding the Vote sub-accounts, since these accounts add an additional step in the process with no added value, as the transaction information is available in IFMIS.

Ring-fencing cash balances

72. The TSA design needs to allow for the protection of certain cash balances. TSA principles seek to maximize the fungible nature of cash, by relying on the financial management systems for exercising expenditure controls. However, in practice there may be instances where ring-fenced balances in both the IFMS and banking systems are justified. In the longer term, ring-fencing may be required to avoid the co-mingling of monies from the various public funds or of the various autonomous government bodies, especially where the IFMS is not being used as the accounting/budget execution tool. In the shorter term, ring-fencing of cash balances within the TSA will likely be required for donor-funded projects and until such time, donors become convinced that earmarked funds cannot be diverted to other purposes.

73. The TSA design should allow for sub-accounts held in the BoU, which retain cash balances earmarked for specific purposes.12 Unlike the Vote sub-accounts, the balances on these sub-accounts would not be swept at the close of each business day. However, the inclusion of ring-fenced balances within the TSA structure will still provide significant benefit in allowing the GoU to aggregate those balances for determining and managing its overall cash position. Further benefits will accrue over time as the usage of and confidence in the IFMS controls grow and the demand for ring-fencing withers.

Recognizing the Uganda Consolidated Fund’s role

74. It is inappropriate to include the UCF as the TSA head account. The current design as described in the TSA Phase One Progress Report13 places the UCF bank account at the apex of the TSA structure. Whilst this may have facilitated the acceptance of and transition to the Phase One TSA structure, it will cause future difficulties. In particular, as the TSA expands beyond the domestically funded UCF operations, attempting to co-mingle all public funds within the UCF, would distract from the UCF purpose and attract resistance. The mission considers that positioning the UCF as the TSA Head Account, is inappropriate for three important reasons:

  • Aggregating all public monies to the UCF is contrary to legal requirements: The UCF is required, and its purpose defined in the PFM legal framework. However, as either an accounting or banking concept, the legal framework distinguishes the UCF from other public funds. The aggregate of all public cash cannot be shown as belonging to the UCF.

  • Placing the UCF as the TSA Head Account conflicts with its fiscal stability role: The UCF performs an important role as being the repository of all UCF receipts, and from which cash is distributed to settle obligations. It plays an important role in ensuring fiscal stability, whereby the inflows to the UCF are used as the main determinant of fund releases to Votes— a role that will likely continue until more reliable revenue and expenditure forecasts are evident.

  • The TSA Head should not be a transactional account: The UCF receives transfers from the URA, borrowing proceeds, and budget support grants, and issues cash to Votes. In a hierarchical TSA structure, the upper levels should represent aggregated views of the related lower levels (in much the same way that a Vote in the general ledger will comprise all transactions from its detailed line items). The TSA Head Account should, therefore, represent a view of all transactions and balances of public monies at the lower levels.

75. The TSA Phase Two design should introduce a TSA Head Account and relegate the UCF to a lower level within the TSA hierarchy. The TSA Head Account should be capable of providing a view of all government-controlled cash balances administered through the BoU. Relegating the UCF to a lower level, preserves its critical purpose in managing execution of the national budget.

76. New financial regulations will need to clarify the relationships between the UCF, public monies, and the TSA. The relationships are not well-defined in the PFM Bill. The financial regulations should be used to ensure a common understanding that all public monies (including the UCF) are to be managed through a TSA arrangement—see also Section I and Appendix II of this report.

Vote sub-accounts

77. The MoFPED should reconsider the pre-funding of the Vote sub-accounts. Under the TSA Phase One arrangements, quarterly budget fund releases are allocated to Votes through the IFMS, to allow commitments and expenditures. Simultaneously, cash is transferred from the UCF to a UCF Holding Account—payments are thus constrained by the fund releases, which are linked to the cash available in the TSA Holding Account. Payments are generated directly through the IFMS where domestic payments use electronic fund transfers (EFT), whilst external payments use bank instruction and letter of credit mechanisms. Payment lists are produced for each Vote, and as part of the work flow process, used to transfer funding from the UCF Holding Account to the respective Vote sub-accounts. The EFTs are effected the same day, generally causing the credit balances on the Vote sub-accounts to be cleared. Residual credit balances may result where EFTs are cancelled or fail—in these instances, the transactions are canceled and the balances automatically swept back to the UCF Holding Account. The bank statements for each Vote sub-account are uploaded to the IFMS and matched to the cash book; each Vote is responsible for completing its bank reconciliation. There is scope for improving the efficacy of operating Vote sub-accounts, notably by:

  • Funding the transfers to the UCF Holding Account in tranches through the quarter.

  • Ceasing the pre-funding of Vote sub-accounts—the requirement to pre-fund and then sweep back any residual balances cause additional work and accounting transactions, but does not appear to strengthen controls. The process should be reviewed (over say a three month period) to assess the value added.

  • As a longer term objective, reducing the number of Vote sub-accounts by progressively (e.g., absorbing first the Vote sub-accounts with relatively fewer payment transactions) pooling transactions through a centrally-maintained TSA sub-account and payment process.

Extending the Treasury Single Account to projects

78. Ush-denominated project accounts represent the biggest group of government-controlled bank accounts and have a substantial combined balance. In February 2013, there were an estimated 600 project accounts held in the BoU with a total balance of Ush1.7 Trillion14 (approx. US$650M)—this balance being unremunerated and outside the BoU computation of the GoU cash position. Recent high-profile breaches in PFM procedures, leading to losses of monies from donor-funded project accounts, have increased demands for greater controls and accountability, but may have also made development partners wary of in-country systems.

79. Thus, the incentive for including project balances within the TSA is substantial but incorporating these balances will require complementary activities. In particular, accountability should be strengthened by projects using the IFMS for recording receipts and commitments, processing payments, completing bank reconciliations, and generating financial reports. Encouraging development partners to embrace the IFMS and TSA will require that MoFPED gains a good understanding of donor requirements and concerns, and a willingness to accommodate those requirements and concerted change management measures. Discussions with the World Bank during the mission indicated a willingness by the Bank to act as a pilot for including its projects within the IFMS and TSA arrangements. With the support of the US Treasury Adviser, high level sensitization of the donor community has commenced, with the overall aim of enhancing understanding of the government processes and development of a roadmap for bringing projects onto the TSA.

80. In preparation for discussions with development partners, the MoFPED should anticipate donor requirements and consider the means to accommodate them. Drawing, in part, on the Mozambique experiences illustrated in Box 2 the following potential donor requirements might be anticipated:

  • Protecting/ring-fencing of donor funds for particular purposes through the TSA and IFMS structures, procedures, and workflows;

  • Authority to spend should be restricted to project managers or other staff designated by the development partners;

  • Enforcing good practices and accountability, including requirements to transact through the IFMS using EFT, segregation of duties, regular back-up, and full audit trail of transactions;

  • The level of flexibility for managing project budgets, e.g., relaxing line item controls by allowing project manager discretion within overall budget ceilings, and subject to having received sufficient receipts of donor funds;

  • The ability to refund unspent funding to the development partners;

  • The year-end rollover of unspent monies and committed budgets through timely and simple procedures;

  • Mechanisms to account for exchange rate variations; and

  • Flexible project reporting, to be based on the standard chart of account structure but perhaps providing an additional code segment for discretionary project use and analysis.

81. The work plans for implementing future TSA phases must allow for significant change management activities. During the course of the mission, a high-level meeting with senior donor representatives facilitated by the US Treasury resident adviser and supported by AFRITAC East took place to explain some of the technical aspects of the TSA and encourage donor participation in the process. Subsequent meetings with the World Bank indicated a willingness to participate in the testing of the TSA through inclusion of World Bank funded projects in the TSA arrangements, which the mission strongly supports. Further change management activities will include, but not be limited to:

  • Requirements elaboration, i.e., convening sessions with individual and groups of donors to understand their common and specific requirements, and to describe solutions;

  • Involve donor representatives in confirming system requirements and testing the configured solutions;

  • Cooperate with donor fiduciary risk assessments;

  • Phasing the implementation, i.e., sequencing might involve: (i) including the Ush-denominated project monies, (ii) including the foreign-denominated accounts, and (iii) reducing the level of direct disbursements;

  • Development of a Donor/MoFPED memorandum of understanding, i.e., setting out the roles and responsibilities of each party etc; and

  • Piloting, i.e., involving one or more donors and projects to use the IFMS and TSA providing feedback to improve their operations, and to serve as a champion for encouraging participation by other donors.

C. Legal and Regulatory Framework

82. The Treasury Single Account arrangements are not defined in the new Bill. There is no definition of the TSA in clause 4; neither are banking arrangements defined nor specified in clause 33, which deals with banking arrangements. The responsibilities of the Accountant General do require that office to, in the management of votes, operate a Treasury Single Account, as defined in regulations. This clause is problematic since it may limit the coverage of the TSA to votes, thereby potentially missing the opportunity to bring extra-budgetary funds, including petroleum funds within the TSA. This issue is an example of the coverage of the Bill discussed in Section I of this report, and should be addressed at the drafting stage.

83. The regulations to the Act should define the TSA, its coverage, and distinguish between the UCF and the TSA. The regulations should not only define the banking arrangements of the TSA, but also anticipate future stages of cash management development during which the benefits of the arrangements can be more fully realized. This will require regulations regarding the investment of surplus cash balances and borrowing to meet short term cash shortages.

D. Summary of Recommendations

84. The mission makes the following recommendations:

  • Recommendation 4.1: Review the TSA structure and introduce a TSA Head Account, and relegate UCF to a lower level within the TSA hierarchy;

  • Recommendation 4.2: Review the efficiency of operating Vote sub-accounts;

  • Recommendation 4.3: During drafting of the legislation, include a definition of the Treasury Single Account and its coverage, with appropriate transitional provisions;

  • Recommendation 4.4: Develop new financial regulations that define the TSA, and clarify the relationship between the UCF, public monies, and TSA;

  • Recommendation 4.5: Work with donors to establish their common and specific requirements for management and accounting of project funds through the TSA; and

  • Recommendation 4.6: Continue to expand the coverage of the TSA and roll-out of the IFMIS, in particular, bringing project accounts into the TSA.

Uganda: Technical Assistance Report-Implementing the Public Financial Management Bill
Author: International Monetary Fund. Fiscal Affairs Dept.