Eswatini: Technical Assistance Report on Government Finance Statistics Mission (July 6-12, 2022)
Author:
International Monetary Fund. African Dept.
Search for other papers by International Monetary Fund. African Dept. in
Current site
Google Scholar
PubMedClose
Publication Date:
06 Dec 2022
eISBN:
9798400226960
ISBN:
9798400226960
Language:
English
Keywords:
BCG debt statistics; BCG GFS data; BCG GFS source data; fair value; nonfinancial asset
Download PDF (475.9 KB)

This technical assistance (TA) mission on Government Finance Statistics (GFS) was conducted during July 6-12, 2022. The main purpose of the mission was to review the progress made by the authorities in implementing previous TA recommendations and provide further support to strengthen the compilation and dissemination of GFS in line with international standards set out in the Government Finance Statistics Manual 2014 (GFSM 2014).

Abstract

This technical assistance (TA) mission on Government Finance Statistics (GFS) was conducted during July 6-12, 2022. The main purpose of the mission was to review the progress made by the authorities in implementing previous TA recommendations and provide further support to strengthen the compilation and dissemination of GFS in line with international standards set out in the Government Finance Statistics Manual 2014 (GFSM 2014).

Summary of Missions Outcomes and Priority Recommendations

1. A Government Finance Statistics (GFS) and Public Sector Debt Statistics (PSDS) technical assistance (TA) mission was conducted in Mbabane, Kingdom of Eswatini during July 6–12, 2022. This mission was conducted under the Data for Decisions (D4D) Fund1, a multi-partner initiative aimed at strengthening the quality of national statistical outputs to better support economic policy making in low- and lower-middle income countries.

2. The primary objective of the mission was to support the authorities in their efforts to enhance the quality and availability of fiscal data and evaluate progress made in relation to the previous mission’s recommendations. The mission covered the following six topics: (i) reviewed the classification of public sector entities and updated of the Public Sector Institutional table; (ii) identified source data needed for compiling fiscal statistics for the entire public sector; (iii) assessed consistency and accuracy of annual GFS for the Budgetary Central Government (BCG), including expenditure according with Classification of Functions of the Government (COFOG); (iv) facilitated the compilation of a draft financial balance sheet; (v) brainstormed on ways to compile high frequency BCG GFS; and (vi) explored ways to improve the process of compiling and disseminating PSDS, considering expansion of instrument and institutional coverage.

3. This mission held discussions mainly with members of the GFS Technical Working Group (TWG), composed with officials from the Ministry of Finance of The Kingdom of Eswatini (MOF), the Eswatini Central Bank (ECB), and Eswatini Central Statistical Office (CSO)2. The mission agenda3 summarizes the order of the issues discussed with authorities, and the implementation status of recommendations from the previous TA mission is reported in Appendix III.

4. The coordination of GFS and PSDS compilation process was discussed. Covid 19 pandemic restrictions have impacted implementation of the Terms of Reference developed to guide the work of the GFS TWG. Members of the TWG agreed to use recommendations from the mission to develop a medium-term action plan. The mission encouraged the development of business process for compiling GFS and PSDS to guarantee continuity and consistency of the GFS and PSDS compilation process.

5. Understanding the perimeter of the general government (GG) and wider public sector remains important. The mission reviewed and updated the Public Sector Institutional Table (PSIT). Authorities are expected to finalize the review and validate the current PSIT before publishing this on the authorities’ official websites.

6. There is significant fiscal activity that takes place outside of the BCG sub-sector, through the expenditures of Extrabudgetary Units (EBU) and local governments (LG). Grants from the budget to other general government units were around SZL 3.9 billion in 2020 (~6% of Gross Domestic Product (GDP). Comprehensive GFS would turn these transfers into additional wages and salaries, or use of goods and services, providing a different picture of government expenditure compared to just looking at the BCG data alone. In addition, EBUs and LG units have some own revenues, including LG property taxes and various fees and charges, and again fully capturing this revenue and the additional expenditures it finances would provide a more comprehensive fiscal picture for Eswatini. The Authorities have made limited progress in collecting financial data of other government units that are outside the budget. To assist the authorities with the data collection process, the mission created a summary table to detail the type of data and possible providers.

7. The Country reports annual BCG data to the IMF GFS database. The mission reviewed the consistency and accuracy of GFS for fiscal years (FY) 2018 – 2020, that were submitted to the IMF GFS database against the Medium-Term Fiscal Framework (MTFF) and concluded that transactions such as revenue, expense and transactions in assets and liabilities are preliminary. Compilation of GFS for FY 2021/22 was initiated during the mission.

8. Monthly fiscal data exist but the authorities do not compile and disseminate high frequency GFS. The source data is the same as the annual, authorities expressed concern over the use of high frequency data due to preliminary nature of the data that results from the lack of closing periods. The mission recommended that the authorities should compile and disseminate high frequency data, even if this data comes with a lag, clearly stating that it is preliminary data and updating and revising data as the final outturn data becomes available.

9. The authorities have made some progress in compiling Functional Expenditures (COFOG) for the BCG, but more work remains to be done. COFOG based expenditure data is still limited to BCG and only captures the spending of part of BCG. The authorities should work to ensure all expenditure can be classified by function ensuring that total expenditure in the functional breakdown is as close as possible to total expenditure by economic categories.

10. Drawing on data from the Statement of Assets and Liabilities, the mission compiled a draft financial balance sheet. Different source data should be considered to put together a financial balance sheet, and specially attention should be given to loans assets, other accounts receivable and payable stock values. The authorities were encouraged to double check the data, and ensure it includes all assets and liabilities, as well as ensuring liabilities data is aligned with DMU debt data. Further work may be needed before the balance sheet data is suitable for inclusion in the IMF GFS database.

11. DMU compiles and disseminates PSDS to the joint World Bank/IMF Quarterly Public Sector Debt database. There was no progress with including accounts payable to BCG debt statistics as proposed in the previous mission. A thorough review of the debt numbers extracted from Commonwealth Secretariat Meridian System (CSM) is needed, before dissemination to the World Bank/IMF database, to ensure data is consistent with actual debt figures. In the longer term, the authorities were encouraged to add other instruments and sectors such as Category A corporations (see Box 1 and Appendix 5) and ECB debt currently available.

12. The action plan for implementing GFSM 2014 in the next two years was updated. The priority recommendations that the mission left with TWG are shown in Table 1 below. The complete list of recommendations can be found in the Action Plan presented in the next section of this report under Table 2.

13. The mission thanks the Kingdom of Eswatini’s authorities for this opportunity to work the country’s fiscal statistics. The Mission met MOF senior staff during the closing meeting of the TA mission. They appreciated the mission scope and ongoing IMF TA and agreed to provide support to the TWG and work comprehensively towards improving compilation of GFS and PSDS that will support the fiscal policy formulation, analysis, and decision-making process in the country. The authorities showed great interest in further training and TA mission and have provisionally agreed to host a follow-up mission within a year. The mission would like to thank Ms. Nompumelelo Dladla, Mr. Xolani Dlamini both from Budget and Economic Affairs Department (BEAD), and Mr. Lihle Dlamini from Debt Management Unit (DMU), for facilitating the work of this mission, for being very patient and motivated to receive this TA mission.

Table 1.

Priority Recommendations

article image
article image

Detailed Technical Assessment and Recommendations

A. Progress Since the Last Mission and Revised Action Plan

14. Limited progress was achieved in implementing recommendations from the previous TA mission. Appendix III presents feedback received from the Government of Eswatini (GoE) in relation to the progress of the priority action/milestones from the previous TA mission held remotely in March 2021.

15. The majority of the recommendations from previous mission were described as work in progress. The mission understood that only one recommendation from previous mission, related to the debt system generating nominal valuation was concluded. The majority of the recommendations were “work in progress” and the mission maintained most of them and in many cases with revised text in the updated action plan.

16. There was “no progress” in two recommendations:

  • (i) GFS TWG to advise Departments on what an appropriate definition of payables would be, to reduce the incidence of confusion of using the term “arrears”. The Treasury Annual Financial Report for the year ended March 31, 2021, explains that accumulation of expenditure arrears resulted from recent expansionary fiscal policy and the volatile nature of the Southern African Customs Union (SACU) revenue collection. On the other hand, accounts payables are mainly expenditure due for payment without any deadline determined that would transform these into arrears. The mission understood that there is a good understanding of what constitutes payables, but there is still no clear definition for arrears. The mission recommended and the authorities agree on a formal definition of arrears, perhaps enshrining this in legislation, and creating reporting requirements on arrears may be one way solution to address this issue.

  • (ii) GFS TWG to compile headline figures on the impact of COVID 19 on the debt and deficit statistics. While the mission did not cover this topic, BEAD agreed and committed to work on it in the next year.

17. The revised priority actions plan for implementing GFSM 2014 and PSDSG 2011 below represents the main recommendations from the recent mission, taking into consideration recommendations from previous missions that were in progress or had no progress at all in the course of last year.

Table 2.

Action Plan for Implementing GFSM 2014 and PSDSG 2011

article image
article image
article image
Legend: High = H / Medium = M / Low = L

B. Coordination of GFS and PSDS Compilation Process

18. The Kingdom of Eswatini has a GFS TWG created in 2019 with representation from the MOF, the ECB and CSO. While Terms of References for the group have been prepared, the implementation of the TWG tasks and responsibilities have been delayed due to COVID 19 pandemic that imposed restrictions, and limited members’ interaction. TWG are expected to organize routine meetings to ensure coordination of the GFS and PSDS compilation to ensure fiscal data consistency. This would imply developing and implementing a working plan, as well as coordinating data sharing.

19. Consistency between the different sets of macroeconomic statistics is of paramount importance. The GFS and PSDS should be consistent with each other, to the extent possible and with relevant data from other macroeconomic statistics, e.g., national accounts (GDP), monetary and financial accounts, and external sector accounts. This includes ensuring that all statistical compilers have a common understanding of the general government (GG) sector and sub-sectors, as well as ensuring stocks and flows can be reconciled. The mission reiterated the importance of MOF to meet regularly with the other compilers of relevant macroeconomic statistics, to discuss and resolve any fiscal data compilation and consistency issues to seek greater consistency of macroeconomic statistics in the country.

20. Documenting business processes for compiling GFS and PSDS was recommended. The development of a GFS and PSDS compilation manual (a guide) that represents a living set of working instructions used in the process of compiling the GFS and PSDS will be useful to ensure sustainability (create an institutional memory) and consistency in the compilation process. The authorities already have an existing MTFF Manual detailing the compilation process, so another suggestion was to include a chapter in the existing Manual that describes the process of compiling GFS and PSDS by the Kingdom of Eswatini GFS TWG.

article image

C. Institutional Coverage - The Public Sector

21. The mission reviewed the structure of the public sector in The Kingdom of Eswatini, updated the inventory of public entities and ensured each public entity was classified according to the guidance of the GFSM 2014. The public sector in the Kingdom of Eswatini is subdivided into GG and Public Corporations sector. The preliminary list of entities is presented in Appendix IV and the Figure 1 below shown understanding of public sector in the country that indicates that at the GG level, there are eighty-six (86) BCG entities, twenty-one (21) EBU and twenty-two (22) LG entities.

22. The classification of public entities was discussed and agreed with the TWG. The mission discussed the classification of public entities individually. The assistance of the Treasury department was helpful to clarify key features of some entities, including the operational independence and extent of their ability to survive without / dependence upon budget support. The mission also consulted the PFM Act 2017 that includes a list of agencies and Government business enterprises.

Figure 1.
Figure 1.

Public Sector in the Kingdom of Eswatini

Citation: IMF Staff Country Reports 2022, 356; 10.5089/9798400226960.002.A001

Source: MOF of the Kingdom of Eswatini

23. Within the BCG, there are four sovereign entities, nineteen (19) ministries, two (2) offices, five (5) governmental departments, eighteen (18) agencies and regulatory agencies, five (5) commissions, two (2) funds, fourteen (14) health and education public entities and seventeen (17) other entities such as boards, councils, public institutes, and services. A significant number of BCG units are classified as Category A entities which, according to the Treasury Annual Financial Report for the year ended March 2021, correspond to public enterprises that fall under different line ministries and that are governed and directed by a Board of Directors which is accountable to the Minister responsible. A list of the Category A public enterprises according to the referred Treasury Annual Report is presented in Appendix V.

Category A Public Enterprises

The Kingdom of Eswatini has several Category A public enterprises, originally established under the Public Enterprises (Control and Monitoring) Act, 19894.

This Act defined these entities as “a public enterprise or body which is either wholly owned by Government or in which the Government has a majority interest or which is dependent upon Government subvention for its financial support”, and originally identified 28 entities.

Today, there are 49 Category A entities, that are a mixture of public facing commercial operations and more governmental entities. Category A entities include entities that should clearly be classified as government units, such as the Eswatini Revenue Authority, the Eswatini Competitions Commission and the Eswatini Environmental Authority. Category A entities also include more commercial entities that operate in a more market facing way, such as the Pigg’s Peak Hotel and Casino, Eswatini Electricity Company and Eswatini Water Services Corporation.

About 45 percent of the Category A entities should be treated as BCG units, 27 percent as EBU, 22 percent as public nonfinancial corporations and the reminder public finance corporation and non-profit organization. A significant number of the Category A entities is not included in the BCG data, only any transfers (subsidies / grants) to them. Therefore, pursuing consolidation of the full GG could bring a different picture of fiscal data not seen through BCG currently being reported.

24. EBU sub-sector comprises twenty-one (21) entities. One (1) regulatory agency, two (2) commissions, five (5) boards, one (1) fund, eleven (11) health and education entities and one (1) other public service. Some of the EBU are also classified as Category A entities.

25. The local government subsector includes 22 entities. Six (6) town councils, ten (10) town boards, five (5) health and education entities and an additional public service. According to the Public Finance Act 2017 (PFM Act 2017), a LG authority may not borrow expect from Central Government and may issue securities.

26. Eswatini does not currently operate any social security funds, but there are developing plans to create a social security system. In November 2021, the Ministry of Labor and Social Security (MLSS) approved the National Social Security Policy5 and under this policy the authorities are considering the creation of a basic state pension scheme, health insurance fund and workers compensation fund. Entities created to operate social security schemes are typically classified as social security funds and classified within the GG sector unlike the existing National Provident Fund and Public Service Pension Fund, both of which are classified outside the GG as public financial corporations (pension funds). The authorities should monitor these schemes and seek to ensure units are classified correctly and GFS data made available from the outset.

27. As for the public corporation sub-sector, there are seventeen (17) entities, of which nine (9) are nonfinancial public corporations and eight (8) are financial corporations including the ECB, two (2) pension funds under the control of the MLSS and Financial Services Regulatory Authority (by convention, applying GFSM 2014 §2.158, financial supervisory agents are classified as public financial corporations). Some public corporations were also identified as “Category A” entities. The GoE is the principal shareholder in all “category A” public enterprises with shareholdings ranging from 75 percent to 100 percent.

28. The mission briefly analyzed the public enterprise sector for the purpose of evaluating their role as market producers. According to data provided in the Performance Model file managed by the Treasury, there is evidence that some of identified public companies such as the Royal Eswatini National Airways Corporation, the Eswatini Posts and Telecommunications Corporation, the Small Enterprises Development Company, and the Eswatini National Industrial Development Corporation, have reported operating losses over years, which is a sign that companies are in distress and most likely need constant support from the Central Government to survive, therefore should be submitted to market test. The market test aims to determine whether a public enterprise provides all or most of its output to others at prices that are economically significant and the extent to which sales of goods and services at economically significant prices are sufficient to cover the majority of production costs6. This is a criterion used to determine if a public enterprise is a nonmarket or market producer. This test should be carried out regularly, on an entity-by-entity basis to determine if an entity should be treated as a GG unit.

29. The Government of Eswatini is a minority shareholder in six (6) other companies. Public units are defined as part of the public sector when they are controlled by government, and this will typically not be the case where GoE only has a minority shareholding, and so these companies should be considered as private companies, absent some other indicators of government control. Thus, the shares that GoE owns in companies such as Maloma Colliery (25 percent of shareholding), Macmillan Eswatini (Pty) Ltd (30 percent of shareholding), Royal Swaziland Sugar Company (6.5 percent of shareholding), Africa Reinsurance Corporation (0.25 percent of shareholding), Nedbank Swaziland Limited (22.35 percent of shareholding), and Standard Bank Limited (25 percent of shareholding) should be recorded in the GFS as equity, in other words, financial assets and GoE should expect to receive dividends as shareholder.

30. A number of non-profit institutions (NPI) serving households that benefit from budget grants were identified. A total of sixteen (16) NPI were identified, among them associations, fund, foundations, schools and clinics. According to the GFSM 2014, paragraph 2.61, a NPI is a resident nonmarket entity that often is not controlled by government. Although in receipt of government funds, these entities should not be classified as part of the GG unless it is established that Government has control over them.

article image

D. Source Data

31. To assist the authorities, an inventory of source data and its respective providers was created to facilitate the collection process of source data needed for the compilation of GFS and PSDS. Table 3 presents a summary that details the type of source data and entities that can be used for the purpose of providing financial data.

32. Various source data have to be used to compile annual BCG GFS. Regarding the revenue, tax revenue can be obtained from Eswatini Revenue Services (ERS). Non-tax revenues are collected from Treasury and other taxes from both ERS and Treasury. Grants are provided by the Ministry of Economic Planning and Development (MEPD). BCG expense is collected from Treasury department. Below the line transactions and respective stocks are obtained from the Treasury Department and DMU.

Table 3.

Data Sources for GFS and PSDS Compilation Process

article image

33. The annual Statement of Assets and Liabilities produced by the Treasury Department are potential source data for financial assets and liabilities. This statement summarizes data from debt records, deposits, MOF supportive documents on receivables and payables, capital project or grants reports, financial statements from trading accounts. Considering that deposits and loans can be obtained directly from ECB and DMU respectively, this statement was regarded as potential good indicator for data on loans assets, other accounts receivables and payables. Financial assets such as currency and deposits should be always reconciled with ECB.

34. Regarding nonfinancial assets, Treasury has a working group that has been working to put together an inventory of tangible assets of the government. The result of this work could help TWG in the valuation of nonfinancial assets needed for the compilation a full balance sheet under GFSM 2014 framework.

35. The authorities expect that a new Integrated Financial Management Information Systems (IFMIS), currently under development, will improve the process of compiling data, though the authorities do not know when the system will be fully operational. The authorities also informed that they are in the process of adopting International Public-Sector Accounting Standards (IPSAS).

36. High frequency (monthly or quarterly) data was not made available during the mission. Although the MOF does have access to monthly budget execution fiscal data, the high frequency data was not made available to the mission. The source data is the same as the annual, the authorities expressed concern over the use of high frequency data due to preliminary nature of the data that results from the lack of closing periods. However, the mission strongly recommended that the authorities compile and disseminate high frequency data during the year, even if this data comes with a lag, clearly stating that it is preliminary data and updating it to final as the actual data is available.

37. Transactions of the GoE are a mixture of cash and non-cash recording. While revenue is recorded on a cash basis, some expense items and transactions in nonfinancial assets are recorded on a noncash basis. Below the line data however remains incomplete, and while the authorities are recording transactions in currency and deposit and loan assets, as well as transactions in loans and debt securities liabilities, there are no transactions recorded in other accounts payable, and this may explain, in part, the large statistical discrepancies between above and below the line data reported to the IMF GFS Database.

38. No data is currently available for LG. The Ministry of Housing and Urban Development (MHUD) is the entity theoretically responsible for collecting and compiling data on the operations of local government municipalities and town boards. While not as large as the EBUs, the LG sector does have some own revenues and expenditures. In 2018/19, two of the largest municipalities – Mbabane and Manzini - had a total income (excluding Government grants) of SZL130 million and SZL 88 million respectively, mostly consisting of, assessment rates (local property taxes). This revenue for just two municipalities suggests that the LG subsector could, in aggregate, have reasonably large revenues and expenditures that is worth including in GFS data. Therefore, the mission recommended that the TWG to follow up on this, and set some deadlines for MHUD to collect and compile GFS for LG.

39. EBU source data required for GFS compilation is available for some units. During FY 2019-2021 transfers from BCG to EBU represent on average about 50 percent of total grants expense (about 3 percent of GDP). Comprehensive GFS would turn these transfers into additional wages and salaries, or use of goods and services, providing a different picture of GoE expenditure compared to just looking at the BCG alone. Financial statements of some of the “Category A” public entities - mostly EBU and public corporations, are available in an excel file entitled “Public Enterprises Performance Model”, created by the Public Enterprise Unit (PEU) of the Treasury Department. The file presents main aggregates of financial reports such as total assets, operating surplus, operating revenue, fixed assets, current assets, current liabilities, total debt outstanding and others. The referred file does not present sufficiently detailed information needed to allow compilation of comprehensive GFS and the format of the data needed to be revised to facilitate compilation. In addition, the file presents indicators for the available Category A public enterprises without separating them according with subsectors of public sector indicated by the GFSM 2014 – EBU and public corporations. To better support transformation of financial data into GFS, the structure of the file needs to be revised and PEU should work toward adding detailed data from notes of public enterprises’ financial statements.

40. Debt statistics source data are recorded and managed on the new CSM system. The CSM system has been operating for less than a year in the GoE and it incorporates advanced functionalities to address debt management requirements and takes advantage of the latest state-of-the-art technologies, with a reporting module with great search features, analytical reports with features that allow customization of reports. The System produces automated reports for disclosing debt. The current challenge is related with source data availability from the system and the need to clean up data recorded in the system. For example, reports have been configured to allow calculation of gross and net debt. In regards to the net debt, due to the lack of data on financial assets, reports are presenting gross debt equals net debt. Besides, there is no document defining terms used in the system. A Glossary of main terms will support understanding of the data being generated in the system.

41. All public enterprises have the same financial year commencing on April 1st and ending on March 31st of the following year, facilitating the compilation and consolidation of fiscal statistics.

article image
article image

E. GFS Compilation and Dissemination

i. BCG Statement of Operations on cash basis and MTFF

42. The PFM Act 2017 requires that the MOF should compile and publish GFS in accordance with best acceptable international standards. The referred ministerial department is compiling GFS for the BCG and disseminates to the IMF GFS database detailed revenue (GFS Table 1), expense (GFS Table 2) and transactions in assets and liabilities (GFS Table 3).

43. The mission discussed 2018-2020 BCG GFS data that was submitted to the IMF GFS database. For fiscal years ending 2020 and 2021, GFS presents large statistical discrepancies of > SZL 1 billion, (>1 percent of GDP) as it can be seen in the Table 4 below. Resolving these discrepancies will improve the overall quality of GFS compiled. In practice, statistical discrepancy is normal and promotes transparency when it is reported as below 1 percent of GDP. This is the case of the statistical discrepancy reported for FY 2018/2019 which is not considered very significant to impact the quality of the data.

Table 4.

BCG – Statement of Operations on Cash Basis

article image
article image
Source: MOF/BEAD

44. There are many reasons for such discrepancies. Firstly, statistical discrepancy is normally to be expected when using a variety of data sources which is the case for GoE where source data used to compile GFS comes from different platforms most likely not using the same criteria to compile the fiscal data. Secondly, the authorities are using different accounting methods to record transactions (cash vs. accrual), this is also a potential source of discrepancy, especially if above and below the line transactions are captured under different accounting methods of recording. Thirdly, the net acquisition of financial assets was a concern for the mission for two reasons: (i) Flows of government deposits reported by ECB and Treasury are not easily reconciled, and (ii) the stock and flows of loans provided could not be reconciled.

45. Furthermore, GFS is compiled from fiscal data that has been bridged to MTFF. Bridging MTFF to compile GFS creates risks of consistency and accuracy of the data as it becomes more subject to errors. Although MTFF is largely presented in the GFSM 2014 standards, it is intriguing to observe that output of the two presentations of government transactions have significant different results.

46. The mission reviewed the consistency and accuracy of GFS for FY 2018 – 2020, against the MTFF. Table 5 below presents the summary result of the comparative analysis. While revenue is aligned with insignificant differences, expenditure presents significant differences between fiscal data reported on GFS format when compared with data presented in the MTFF format. Net incurrence of liabilities is also reported with significant differences. The authorities explained that GFS data submitted to the IMF GFS database is preliminary data while MTFF reflects actual data, therefore GFS for FY 2018-2021 should be resubmitted.

47. While GFS data revenue data is reasonably well, though not perfectly, aligned with MTFF data, there are more serious issues with expenditure. MTFF expenditure, and especially MTFF expense data is consistently higher than expense in GFS data. Transactions in financial assets are well aligned, but there are more issues in liabilities and so coupled with the above the line differences in expenditure, the MTFF data has much larger statistical discrepancies. Considering that data source is the same, it might indicate that MTFF is not completely aligned to GFSM 2014, and the authorities should consider fully migrating its MTFF to GFSM 2014 standards as it seems to provide better fiscal results that could serve as a source of data for policy formulation and forecasting. In addition, the authorities should review the GFS compilation file, to ensure it is capturing all transactions correctly.

48. GFS compilation beyond BCG is still very dependent on availability of source data, and the focus should be on improving the quality of annual and high frequency BCG. Regarding high frequency, no monthly or quarterly BCG GFS source data was made available to the mission to enable compilation of high frequency GFS, however the mission extensively discussed ways to overcome issues related to the compilation of high frequency data.

article image

Table 5.

BCG – Cash flows from Operating Activities vs. MTFF

article image
Source data: MOF/BEAD

ii. Functional Expenditures (COFOG)

49. The mission analyzed expenditures classified according to the COFOG that are currently under development by the MOF. COFOG is still limited to BCG, and available source data from public entities outside BCG is not detailed enough or inexistent to allow compilation of COFOG for the entire GG. Table 6 shows the extent of the functional expenditures classified by COFOG, but also highlights the gap between functional expenditures and total expenditure in the GFS. There are still significant amounts of BCG expenditures that have not yet been mapped to COFOG, and the authorities have further work to do to capture and classify all BCG expenditures before starting disseminating COFOG data.

Table 6.

BCG – COFOG Sum of FY 2018–2021

article image
Source data: MOF/BEAD
article image

iii. Balance Sheet Data

50. Balance sheet is a core component of GFS analytic framework that enables a comprehensive view when monitoring and assessing economic and financial conditions and the behavior of public sector units. It records the stock positions of assets, liabilities, and net worth of the sector or subsector at the end of each reporting period.

51. The mission prepared a draft financial balance sheet in the GFS Framework using the existing annual Statement of assets and liabilities produced by the Treasury Department. This captured assets such as deposits and accounts receivable and liabilities such as debt securities, loans and accounts payable. Table 7 presents the preliminary financial balance sheet compiled during the mission.

52. It was practical impossible to reconcile currency and deposits between Treasury and ECB. Treasury appears to report deposits inflows and outflows separately, while ECB reports deposit as net figures of a fungible account. ECB presents government balances in local and foreign currency including accrued interest, indicating that ECB reports on accrual basis while Treasury uses a cash basis approach. This also creates differences in the deposit figures that need to be reconciliated. For the purpose of compiling financial balance sheet, the mission understands that the ECB can be a more reliable data source, but figures need to be reconciled to ensure that transactions above the line respond to the ones below the line.

53. The statement of assets and liabilities does not allow the split between domestic and external assets and liabilities. The debt liabilities reported by Treasury were hard to reconcile with the DMU figures. Treasury confirmed that data used in the referred statement was obtained from DMU, therefore, the mission suggested the use of the DMU figure for the purpose of compiling Finance Balance Sheet.

54. The Treasury statement also provides a potential data source for loans assets, other accounts receivable and payable. With support from Treasury department staff, mission tried to understand the content of each item reported in the statement of assets and liabilities and concluded that advances are mainly loans; capital and special fund, accrued liabilities, and all the items identified as 6218 and 6318 in the statement of assets and liabilities reported in the Appendix VI should be regarded as other accounts receivable and payable.

55. To sum up, Finance Balance Sheet was produced with currency and deposits from ECB; loans assets, other accounts receivable and payable from Treasury; and debt liabilities from DMU. A first draft was put together however, further work is required before there is an attempt to disseminate it through the IMF GFS database. Assets that should be included in the government’s financial balance sheet should include government equity assets, currently missing, and further work is needed to further interrogate the source data and ensure its correctly classified.

Table 7.

BCG – Finance Balance Sheet

article image
Source data: MOF/BEAD & Treasury
article image

F. PSDS Compilation and Dissemination

56. The Kingdom of Eswatini compiles and disseminates BCG debt statistics to the joint World Bank/IMF Quarterly Public Sector Debt database. The mission reviewed annual debt data that is compiled in accordance with PSDSG 2011 and found that coverage is limited to the BCG, although some debt data is available for some other public sector enterprises.

57. Debt data is provided by instrument, counterpart (external and domestic) and currency. The CSM has been well defined, it can automatically produce reports that allow debt by instruments, by maturity, by currency of denomination and maturity, type of interest, residency (domestic and external), by counterpart, debt services and others and in different time frequencies - daily, monthly, quarterly and annually

58. Expanding instrument coverage from D1 (debt securities and loans) to cover all debt instruments remains challenging due to the lack of available source data. Table 8 shows the summary of the annual debt figures for FY 2018-2021 at face value. Debt at nominal values is available in the CSM, but for years before 2021 they need to be properly reconciled between the valuation methodologies.

Table 8.

BCG Gross debt – Face Value

article image
Source: MOF/DMU – CSM, July 2022

59. DMU is still validating data entries in the CSM and only able to validate debt numbers for FY 2020/2021. The mission discussed main challenges related to debt statistics and the authorities informed about issues related to comprehensiveness of the data recorded and available that reflect real debt situation, particularly related to disbursements for fiscal years before 2021. This certainly explain the very large discrepancies in 2018, 2019 and 2020 between loans at nominal and at face value. At the moment, the authorities are only able to validate FY 2021 numbers and working on CSM to clean entries and ensure debt data for fiscal years before 2020 reflect actual numbers.

60. The Meridian System produces debt figures valued under nominal, face and fair value. While GFSM 2014 proposes that stock positions be valued at market value, it recognizes that not all debt instruments can be valued at market value and suggest also valuation of debt instrument under nominal value. No market valuation is clearly available in the CSM; however, a fair value is reported. According to the GFSM 2014, paragraph 3.115, fair value is a market-equivalent value defined as the amount for which an asset could be exchanged, or a liability settled, between knowledgeable, willing parties in an arms-length transaction. It thus represents an estimate of what could be obtained if the owner sold the asset or the debtor settled the liability. Definition of the terms used in the Meridian system were not available to ensure if fair value would be a good proxy for market value, particularly for treasury bills and bonds that are tradeable instruments.

61. The mission carried out a basic stock-flow check, to ensure opening and closing debt stocks (at face value) reflected new drawings and amortization within the period reported in the statement of operations. An integrated balance sheet for debt liabilities was generated from the system. Transactions related with disbursement and amortization when integrated with stock explain in large closing debt figures as it can be seen in Table 9 below.

Table 9.

Integrated Balance – Annual Gross Debt (Face Value)

article image
Source : MOF/DMU – Meridian, July 2022

62. No Other Economic Flows (OEF) are recorded for debt securities, but are applicable to loans. Debt securities stocks and flows are well integrated in 2020 and 2021 – the change of stocks (at face value) is fully explained by the transactions reflecting no OEFs due to revaluation. On the other hand, transactions don’t fully explain the change in the stock of loan liabilities due to the fact that much of the external loans to Eswatini are denominated in foreign currencies. Per GFSM 2014, paragraph 3.31, OEF represent changes in the volume or value of assets or liabilities that are not generated by transactions. The OEFs are not transactions because they do not meet one or more features of transactions. For example, when the amount of a debt is expressed in foreign currency such as US$, the value in domestic currency can change to the impact of the exchange rate even if the stock of debt in US$ is unchanged.

63. The mission is assuming that “parity change” figures obtained from the CSM would be a proxy for OEF as it is shown as residual amounts and according to the authorities, they reflect mainly the impact of exchange rate movements on external loans denominated in foreign currencies. However, as mentioned before, the authorities admitted that disbursements of 2020 and previous years, might not be fully captured in the system, meaning that the total amount reported under Table 9, might not all be related to OEF.

article image

Appendix I. Names of Officials

article image

Appendix II. Technical Assistance Mission Agenda

article image
article image

Appendix III. Priority Action / Milestone from Previous TA Mission

article image
article image

Appendix IV. Public Sector Institutional Table

article image
article image
article image
article image
article image
article image
Source data: MOF/BEAD and other Government official documents and websites

Appendix V. Category A Public Enterprises

article image
article image
Source data: MOF/Treasury

Appendix VI. Statement of Assets and Liabilities

article image
article image
Source data: MOF/Treasury
2

See list of officials met during the mission in Appendix I

3

See final TA mission agenda in the Appendix II.

6

For more information, see GFSM 2014 §2.64-2.75

Other IMF Content

Other Publishers

  • Collapse
  • Expand
Cover IMF Staff Country Reports
Eswatini: Technical Assistance Report on Government Finance Statistics Mission (July 6-12, 2022)
Author:
International Monetary Fund. African Dept.
Volume/Issue:
Volume 2022: Issue 356
Publisher:
International Monetary Fund
ISBN:
9798400226960
ISSN:
1934-7685
Pages:
45
DOI:
https://doi.org/10.5089/9798400226960.002
Figure 1.

Public Sector in the Kingdom of Eswatini