Uganda
Technical Assistance Report-Report on Government Finance Statistics

This Technical Assistance Report discusses technical advice and recommendations given by the IMF mission to the authorities of Uganda regarding compilation and dissemination of government finance statistics in accordance with the guidelines of Government Finance Statistics Manual 2001/14. The IMF mission, in collaboration with the Uganda Bureau of Statistics, the Bank of Uganda, and the Ministry of Finance, Planning, and Economic Development, produced a new institutional table that can be further reviewed, adopted, and used as a standard for the entire government of Uganda in the production of statistics. The IMF mission recommends that the government of Uganda continue to disseminate government finance statistics to the IMF and the East African Community and start disseminating a broader range of public sector debt statistics through the World Bank.

Abstract

This Technical Assistance Report discusses technical advice and recommendations given by the IMF mission to the authorities of Uganda regarding compilation and dissemination of government finance statistics in accordance with the guidelines of Government Finance Statistics Manual 2001/14. The IMF mission, in collaboration with the Uganda Bureau of Statistics, the Bank of Uganda, and the Ministry of Finance, Planning, and Economic Development, produced a new institutional table that can be further reviewed, adopted, and used as a standard for the entire government of Uganda in the production of statistics. The IMF mission recommends that the government of Uganda continue to disseminate government finance statistics to the IMF and the East African Community and start disseminating a broader range of public sector debt statistics through the World Bank.

Executive Summary

In response to a request for IMF technical assistance (TA) by the Ugandan authorities and in consultation with the IMF’s African Department (AFR), a government finance statistics (GFS) TA mission from the IMF’s Statistics Department visited Kampala during July 8–18, 2014. The mission was led by Mr. Rocha, with participation by Mr. Robinson. Mr. Johan Mathisen, the IMF GFS Advisor at the East Africa Technical Assistance Center (AFRITAC), and an East African Community (EAC) Secretariat representative, Mr. Michael Gitau, participated in the mission. The mission was undertaken within the context of project STA_UGA_2015_04 and was aligned with the improvement of fiscal data requirements associated with the convergence criteria under the East African Monetary Union (EAMU) Protocol.

The main objectives of the mission were to (i) assist the Ministry of Finance, Planning, and Economic Development (MOFPED) and the Uganda Bureau of Statistics (UBOS) in the compilation and dissemination of GFS in accordance with the guidelines of the Government Finance Statistics Manual 2001/14 (GFSM 2001/14); and (ii) provide GFS training to counterpart staff.

Main findings and recommendations

MOFPED, UBOS, and the Bank of Uganda (BOU) jointly revised the GFSM 2001/14 implementation plan to reflect Uganda’s particular circumstances and implementation capacity. The plan is to be developed further and presented at an IMF (East AFRITAC)/EAC regional workshop in Arusha, Tanzania on August 24-29, 2014. The mission recommends implementing the final plan as envisaged to ensure that Uganda meets statistical requirements under the EAMU Protocol.

The mission, in collaboration with MOFPED, UBOS, and the BOU produced a new institutional table that can be further reviewed, adopted, and used as a standard for the whole Government of Uganda (GOU) in the production of statistics. Future reviews of the institutional table should be performed regularly and jointly by the three institutions for consistence in data reporting across macroeconomic frameworks.

The mission assisted GFS compilers in incorporating data for extrabudgetary units to build GFS for a full general government sector. The mission recommends that this work should continue to attain standard fiscal data reporting requirements under the EAMU macroeconomic convergence criteria.

The mission assisted officials in aligning the GFS public sector debt statistics template with the World Bank’s template, recommended that the GOU compile and report other debt instruments, such as “currency and deposits” as well as “insurance, pensions and standardized guarantees,” and that all these debt statistics be submitted to the World Bank’s Public Sector Debt Statistics database.

The mission recommends that the GOU continue to disseminate GFS to the IMF and EAC and start disseminating a broader range of public sector debt statistics through the World Bank.

The mission would like to thank the staff of the MOFPED, UBOS, and the BOU for their hospitality and support, which contributed greatly to the success of the mission.

I. Introduction

1. The mission’s main objectives were to review Uganda’s compilation and dissemination of government finance statistics (GFS), provide recommendations for improvement to align current practices with the Government Finance Statistics Manual 2001/14 (GFSM 2001/14) framework, and provide GFS training to counterpart staff.

2. The major tasks of the mission included (i) assisting authorities in finalizing the national GFSM 2001/14 implementation plans; (ii) providing advice on improving the institutional, derivation, and bridge tables, and on updating the General Data Dissemination Standards (GDDS) metadata and plans for improvement, and on preparing the Government Finance Statistics Yearbook (GFSY) and International Finance Statistics (IFS) questionnaires for submission; and (iii) providing advice on how to further automate the compilation and dissemination of GFS, widen data coverage, increase data frequency, and include balance sheet data.

3. The remainder of this report is divided into five sections. Section II discusses the legal and institutional environment, the institutional structure of general government, and the coverage of the compiled and disseminated fiscal data. Section III deals with (i) the chart of accounts, information systems, and issues concerning current fiscal statistics; (ii) data compilation; (iii) dissemination; and (iv) GDDS and Special Data Dissemination Standards (SDDS) metadata issues. Section IV presents the country’s current situation in terms of consistency of GFS and other macroeconomic data. Section V presents the country’s status concerning the migration plan to the GFSM 2001/2014. Section VI addresses resources, training and additional technical assistance (TA) issues.

4. The appendices to the report are as follows. Appendix I shows the institutional structure of the general government. Appendix II lists the general government entities and public corporations. Appendix III shows the bridge tables linking national and GFSM 2001/14 classifications. Appendix IV contains the derivation table of Uganda’s budgetary central government’s aggregates and GFS 2001/14 aggregates. Appendix V includes the classification code changes in GFSM 2014 relative to GFSM 2001, and Appendix VI is a list of officials met during the mission.

II. Institutional and Data Coverage of the General Government and Public Sectors

5. The 2003 Public Finance and Accountability Act (PFAA) forms the regulatory framework for public financial management in Uganda. It specifies the regulations of the financial management of the government; prescribes the responsibilities of persons entrusted with financial management in the government; regulates the borrowing of money by government; and establishes the responsibilities of the Ministry of Finance, Planning, and Economic Development (MOFPED) together with the powers that may be delegated to senior officials. The Act also defines the responsibilities of the Secretary of the Treasury and those of the Accountant General.

6. The PFAA also determines that the accounts of all public organizations shall be audited by the Auditor-General and that all public organizations “shall, within four months at the end of its financial year, prepare and submit to the Minister of Finance audited annual financial statements relating to its operations during its financial year.” The financial statements “shall be laid before Parliament by the Minister not later than one month after the Minister receives them.” In addition, the Act states that the financial year shall start on July 1 and end on June 30 of the following year. Moreover, Uganda adopted (cash basis) International Public Sector Accounting Standards in 2006, which help to ensure GFS-consistent financial reporting by public entities (See GFSM 2014 Appendix 6).

7. Ongoing legal and institutional public finance reforms will likely improve the authorities’ capability of compiling timely and more comprehensive GFS. In particular, the Public Finance Bill (Bill No. 5 from 2012), which was under discussion in Parliament at the time of the mission, seeks to clarify reporting responsibilities of government entities and harmonize reporting periods. The bill may also provide for improved statistics on Ugandan government contingent liabilities. Finally, the legislation requires that MOFPED consolidate accounts for the various levels of government, which will be facilitated by the October 2013 introduction of a single treasury account.

8. The general government sector in Uganda comprises three subsectors: budgetary central government (BCG) units, extrabudgetary units (EBUs), and local governments. There are no state governments in Uganda. The BCG subsector includes the offices of the executive, legislative, and judicial branches, plus various other government entities. The EBUs subsector comprises a significant number of universities, other educational institutions, research institutions, councils, commissions, and various other semi-autonomous agencies. Local governments comprise districts, municipal councils, and town councils.

9. The mission reviewed the sectoral classification of all public entities in Uganda. Working meticulously with MOFPED, Uganda’s Bureau of Statistics (UBOS), and the Bank of Uganda (BOU) staff members, several institutional units were identified for resectorization and some were classified for the first time. The new sectorization increased BCG to 124 entities from 66 entities reported previously, and EBUs were reduced to 63 institutions from 141 institutions (Appendix I). A preliminary comprehensive list of general government and public corporations is presented in Appendix II.

10. The National Social Security Fund (NSSF) is a compulsory saving scheme, consistent with the definition of a provident fund, and is classified as a public financial institution in line with GFSM 2014 sectoring guidelines. It is a simple compulsory saving scheme that keeps contributions of each participant and of their employer on behalf of each participant in a separate account, managed and invested to obtain a return. The NSSF does not provide social insurance benefits, but the deposits can be withdrawn under specified circumstances, such as retirement.

Recommendation:

  • Review and finalize the institutional table prepared during the mission and adopt the final table as a standard for the production of consistent macroeconomic statistics in Uganda. In addition, procedures should be developed and implemented for maintaining and reviewing regularly the public sector institutional table.

11. The MOFPED currently compiles consolidated monthly and annual data for 66 BCG entities on a cash basis in a GFSM 2001 format. Tax revenues are recorded when collections are made, and categories of expense are recorded when cash payments are made. MOFPED publishes data for BCG in the annual Statistical Abstract, both in hardcopies and disseminated on the Ministry’s web site. The Statistical Abstract also presents aggregated data on local governments, selected information on EBUs, and some information on public corporations. It also contains selected data on public sector debt. MOFPED also publishes GFS for BCG on a monthly basis on its website.

12. MOFPED reports annual (GFSY) budgetary central and local GFS and monthly (IFS) BCG finance statistics to the Statistics Department at the IMF. The monthly series cover July 1997 to March 2014, and the annual series run from 1997/98 to 2012/13, with work ongoing to report 2013/14. GFS activities in MOFPED are coordinated by the Macroeconomic Policy Department, using inputs from various departments including the Accountant General’s Office, Directorate of Budget, Tax Policy, BOU, as well as local governments. The immediate challenge going forward will be to increase coverage of BCG data to cover all entities on the revised institutional table.

Recommendation:

  • Gradually increase coverage of BCG data to cover all entities on the revised institutional table, and resubmit historical IFS and GFSY data.

III. Accounting and Statistical Systems

13. A well-defined and detailed chart of accounts (COA) is a good starting point for compiling sound GFS. Uganda issued a substantially revised COA in March 2011, which has since been implemented in all BCG units, all local governments, and some EBUs. Remaining EBUs and most public corporations use a private accounting system aligned with International Financial Reporting Standards.

14. The mission confirmed that Uganda’s new COA is broadly aligned with GFSM 2001 classification codes. This was accomplished by classifying detailed BCG, EBU, and local governments’ data in the IMF’s Classification Assistant tool. The resulting bridge tables (Appendix III) show alignment with GFSM 2001. The related derivation table (Appendix IV) shows that the main national and GFSM 2014 aggregates are consistent.1 Given the wide implementation of the new COA and that other public entities generally produce financial statements aligned with International Financial Reporting Standards, it appears feasible to compile general government (and annual public sector) finance statistics. Importantly, the mission provided a list of the new accounting codes that were produced as part of the GFSM 2001 to 2014 update (Appendix IV) for consideration in future updates of the COA.

15. Data for BCG are compiled using a variety of source data. Tax revenue data are from the Tax Policy Department within MOFPED. Nontax revenue data are compiled by the Treasury Department within MOFPED. Expense and transactions in nonfinancial assets and financial assets and liabilities data are obtained from Uganda’s Integrated Financial Management System (IFMS), which reflects detailed classification of expense that are highly consistent with the GFSM 2001 coding system. Expense incurred by BCG donor projects is not recorded in IFMS, but is estimated and added by MOFPED. Data on transfers are from the Budget Department within MOFPED, while external grants data are from the Aid Liaison Department. COA procedures are in place to modify the classifications when new accounts are added to IFMS. The accounts are generally prepared on a cash basis.

16. The mission revised and updated GFSY and IFS questionnaires based on disaggregated source data provided by the MOFPED. The high frequency IFS questionnaire was aligned with GFSM 2001 and is updated through May 2014, with substantially more detail compared to previous submissions. A preliminary 2012/13 GFSY questionnaire was also produced jointly with the authorities.

Recommendation:

  • MOFPED regularly update and submit the detailed GFSM 2001-compliant IFS questionnaire produced by the mission. Annual data submissions on BCG assets and liabilities should be expanded to produce a partial balance sheet.

17. Annual local GFS are compiled based on annual reports filed in hardcopy with the Accountant General’s Office within MOFPED. While districts report on a cash basis, municipalities and town councils report on an accrual accounting basis. Approved estimates of revenue and expenditure (recurrent and development) for local governments are issued by the MOFPED. The local GFS are made available on MOFPED’s Internet website, as well as transmitted to the IMF for publication in the GFSY.

18. The frequency, timeliness, and level of detail of local GFS may improve with the rollout of electronic transactional information systems in Uganda. The rollout of IFMS so far covers 40 local governments, and there are other improvements in software tools to collect local government budget data. A complete IFMS rollout to local governments is envisaged to begin within the next two years. In the meantime, this data source can be used for compilation of partial local GFS, automated by integrating the bridge tables in the budget transactional information system.

Recommendation:

  • Gradually collect more detailed and timely local government data as it becomes available, and automate processing of such data by integrating the GFS bridge table with the information system.

19. An important goal of the Ugandan authorities is to build full GFS coverage of the general government sector by adding data for EBUs. The main impediments have thus far been incomplete reporting by EBUs, fiscal year inconsistencies, methodological discrepancies, and timeliness issues.2 Notwithstanding these difficulties, MOFPED has begun assembling an annual dataset for EBUs based on audited reports, which are published in the Statistical Abstract. The most recent such publication was in June 2013 with data for 2004/05–2010/11.

Recommendation:

  • The new Public Finance Bill and its regulations should harmonize EBUs’ fiscal years, clarify reporting responsibilities, and increase the frequency of reporting by EBUs.

20. The mission used data for EBUs to generate partial GFS for this subsector, as well as trial GFS for a consolidated general government sector. GFS for the EBU subsector were compiled using a multiperiod IMF Classification Assistant for 2004/05 through 2010/11 for 37 of the 66 EBUs for which data were available.3 GFS for the consolidated general government sector were compiled using a multisector IMF Classification Assistant; revenue and expenditure data for BCG, EBUs, and local governments for 2010/11 comprised the accounts. The Classification Assistant produced key GFSY questionnaire tables.

21. The fact that the mission developed different bridge tables for BCG, EBUs, and local governments demonstrates the potential for compiling GFS for the general government sector and for expanding data coverage for the GFSY in the not-too-distant future.

Recommendation:

  • Continue efforts to extend the EBU dataset and use bridge tables prepared by the mission to include EBU finance statistics in data submissions to the IMF.

22. GFS consolidation for the general government sector is quite simple in the case of Uganda as only one of the three important types of accounts for consolidation is relevant. The three types of accounts are assets and liabilities, interest, and grants. Uganda does not currently produce complete and detailed GFS data on assets and liabilities, and there are no interest flows across subsectors of general government because BCG borrows for the entire sector. Therefore, Uganda can focus its consolidation efforts on grants. Uganda’s COA includes a full complement of GFS codes on grants, such as to/from foreign governments, to/from international organizations, to/from other levels of government, and current and capital breakdowns for each of the three types of grants. However, the national tabular presentations of GFS emphasize a “budget support” and “project support” classification of grants. To facilitate consolidation, Uganda would benefit from compiling data according to the classification of grants required by GFSM 2014.

23. MOFPED currently compiles annual data for public financial corporations and for some of the nonfinancial corporations. The data are published in the Statistical Abstract. The Public Finance Bill requires that all public corporations file annual reports with the Accountant General. MOFPED plans to use these data when they become available to produce public sector finance statistics.

24. The mission carefully reviewed Uganda’s GDDS metadata on government debt and on government operations, as well as SDDS metadata on central government operations. Although the last update of the metadata was in April 2013, some updates were found to be required. Further revisions will be necessary as the coverage of GFS expands to the general government sector.

25. General government debt data are also published by the MOFPED in its annual Statistical Abstract at a much aggregated level. General government external debt data are compiled by the MOFPED and reconciled with the BOU quarterly. General government domestic debt data are compiled at the BOU and are shared with MOFPED. The PFAA states that only the MOFPED can contract loans and issue guarantees on behalf of local governments and public corporations. There are no data available on other accounts payable for local governments and EBUs, and they cannot incur any other types of debt. Hence, there are no debt statistics for local governments. Public corporations debt data are not compiled by the BOU, but they do collect currency and deposits, loans, and some recent data on other accounts payable for financial public corporations.

26. The mission held meetings with staff members who prepare public debt statistics within the MOFPED and the BOU in order to obtain a time series of general government debt. Currently, Uganda only provides data on external debt for the World Bank. During the mission, the World Bank’s public sector debt template was populated with quarterly data from 2009 to 2013 including information on general government domestic debt, “currency and deposits” and debt data from the BOU (loans and other accounts payable). Additionally, source data were provided and verified. The mission could not obtain debt data for “insurance, pensions and standardized guarantee schemes,” but was able to have a spreadsheet compiled with public debt data for the period above for general government and for the BOU.

Recommendations:

  • Source data for “currency and deposits” and “insurance, pensions and standardized guarantee schemes” should be organized into a dataset to improve compilation, validation, and dissemination of general government debt data by the BOU.

  • Develop procedures and mechanisms for the collection of public debt data from public corporations.

27. Concerning general government debt data, the MOFPED requested that the mission evaluate and suggest improvements to the draft Uganda Debt Statistical Bulletin. The mission, inter alia, recommended that the set of tables used in the dissemination of GFS debt data be incorporated into the bulletin, and that a clear revision policy be defined and made public through the same publication. The publication is intended to be produced on an annual basis and, ultimately, evolve into semiannual and quarterly publication.

IV. Consistency of GFS Datasets and Other Macroeconomic Data

28. Uganda established the Public Finance Statistics (PFS) Committee in 2010 to address public finance fiscal data reconciliation and quality improvement issues. The PFS Committee is comprised of members from UBOS, MOFPED, the BOU, and the Local Government Finance Commission. The committee meets regularly to ensure consistency between GFS and other macroeconomic data.

29. At the request of the authorities, the mission focused on internal GFS data consistency. To perform a consistency check on BCG data, the TA mission obtained detailed data for fiscal years 2010/11 and 2012/13 on tax revenue, nontax revenue, expense, arrears, transactions in nonfinancial assets, and domestic and external grants from MOFPED, and classified these data using the IMF Classification Assistant.4 Ugandan authorities indicated that the detailed data were consistent with published estimates.5 As part of this process, the mission identified several misclassifications. For example, employment-related social benefits were classified as “compensation”; bad debt was classified as an “expense”; transfers to nongovernmental units were classified as “grants”; and various types of arrears categories were classified as “liabilities” and not coded to GFS expense classifications. Nevertheless, as noted above, the mission found that total revenue and expense values in the Classification Assistant and in the source data were comparable.

Recommendation:

  • Correct the above-highlighted misclassifications, including removing the expense line items for “arrears,” distributing the amounts shown to subitems of the relevant GFS types of expense, and introducing a memorandum item for the total stock of arrears.

30. To check for local government data consistency, the mission obtained data from MOFPED that were based on detailed accounts presented in local governments’ audited financial statements. These data were used to perform GFS compilation using the Classification Assistant. The compilation produced accurate results, as values for categories of revenue and expense produced by the Classification Assistant were largely consistent with the original source data. Classification Assistant balancing items were also consistent with source data.

31. To ensure data quality of EBUs’ audited annual reports, the mission confirmed that the Classification Assistant organized accurately categories of revenue, expense, net acquisition of nonfinancial assets, and acquisition of financial assets and incurrence of liabilities. The mission also confirmed that balancing items were calculated consistently by the Classification Assistant.

32. To meet its statistical needs, UBOS also has a dataset for BCG fiscal data, as well as for local government data and it is developing a database for public corporations. BCG data are obtained through distinct institutions, and data sources for other sectors are obtained through hardcopies of financial reports. Consistency between UBOS and the MOFPED data should be reviewed during future GFS TA missions.

33. For the moment, only one of the data sources being used to compile Uganda’s GFS data is published and disseminated independently in any official document. Consolidated financial data for BCG is published in an Audit Report on the Accountant General’s website. However, the new Finance Bill, if it passes Parliament, includes provisions for the Accountant General to publish consolidated accounts for central government (including BCG units and EBUs), local governments, and public corporation. Such action will impose more discipline on the production of public sector fiscal data in the country.

V. GFSM 2001/14 Implementation Plan

34. The GFS Technical Working Group in Uganda is the Social Economic Statistics Technical Committee, which is comprised of representatives from MOFPED, the UBOS, and the BOU. The Committee’s main objectives are to improve GFS and to define a strategy for migrating to GFSM 2001/14. MOFPED formed a high-level GFS Committee comprised of commissioners from the various internal departments to fast-track GFS implementation. MOFPED also has a Sector Statistics Committee (SSC) that is comprised of representatives from all departments, which implements directives and recommendations of the GFS Committee.

35. The mission worked with MOFPED, BOU, and UBOS officials to revise the GFSM 2001/14 implementation plan, which aims to meet EAMU Protocol fiscal statistics requirements. Following the mission, the plan was finalized and presented at a joint IMF East AFRITAC/EAC regional workshop in Arusha, Tanzania during August 24-29, 2014. Implementation of the plan will result in several important improvements in the Uganda’s GFS, such as coverage expansion to the general government sector by 2015. The target date for adopting a modified cash basis of reporting is 2017. In addition, according to the plan, Uganda will disseminate estimates of the stock of financial assets and liabilities by 2017. The implementation plan should be aligned with MOFPED’s new Sector Strategic Plan for Statistics, which will cover 2013/14–2017/18.

article image
article image
article image

VI. Resources, Training, and Technical Assistance

36. Work on GFS in the MOFPED is coordinated by the Macroeconomic Policy Department of the Directorate of Economic Affairs, which compiles data using inputs from various departments within the ministry, from IFMS and from other institutions (see paragraphs 9 and 20 for further information). GFS reporting to both the IMF and EAC is performed by MOFPED. A March 2010 Memorandum of Understanding (MOU) between UBOS and MOFPED clarifies the statistical responsibilities of both organizations. The MOU determines that MOFPED shall be responsible for remitting fiscal data for publication in the IMF’s GFSY and IFS after reconciliation and validation by the PFS Committee and that “UBOS shall be responsible for disseminating official Public Finance Statistics at the national, regional, and international levels.” According to the MOU, “upon verification and validation by PFS Committee, UBOS shall recognize fiscal data and publications produced by MOFPED as official.”

37. The mission had as one of its primary objectives providing hands-on training for data compilers. During the mission, the following training sessions were conducted: (i) revisions to the GFS framework as a result of the GFSM 2001-to-2014 update; (ii) identifying institutional units and methods for sectorization; (iii) understanding GFS classifications and framework; (iv) using the classification assistant; and (v) operationalizing consolidation. As part of its new joint effort to assist partner states in meeting EAMU Protocol convergence criteria, the IMF (East AFRITAC) and the EAC plan to conduct a regional workshop in Victoria Falls, Uganda on November 17–21, 2014. The following week, November 24–28, IMF/EAC jointly sponsored GFS training will be conducted for selected financial officers in Uganda.

38. The MOFPED requested that the mission provide a review of Uganda’s COFOG mapping so that it is aligned with GFSM 2001/14. The mission conducted a hands-on exercise to build bridge tables between Uganda’s classification of functions of government and the COFOG.

Recommendation:

  • Programs and projects in recent or current use should be selected on a specific file and a list compiled with the main objectives of each of the program/projects in order to adequately classify programs/projects into their relevant COFOG categories. In case a project is multifunctional, the mission recommends that indicators be used to split multifunctional projects into multiple COFOG classifications.

39. UBOS requested that the mission make recommendations on the next steps in terms of database development for the public sector and on the extraction of GFS data from public corporations’ financial statements. The mission evaluated the existing database, suggested possible improvements and initiated a process of providing assistance on how to use financial statements to generate GFS data.

Recommendation:

  • To the extent possible, the various agencies should share the same databases and data sources to guarantee consistency among distinct macroeconomic statistical systems, optimize the use of resources, and avoid duplication of efforts.

40. A follow-up mission is tentatively scheduled to take place before May 2015.

Uganda: Technical Assistance Report-Report on Government Finance Statistics
Author: International Monetary Fund. Statistics Dept.