Developing the Budget Declaration

The 2018–20 draft Budget Declaration, submitted to the Cabinet of Ministers on June 1, is a major step forward in terms of establishing a medium-term orientation to budget planning. For the first time, important elements of a MTBF were included in the budget documentation, including the presentation of detailed medium-term macroeconomic and fiscal forecasts and expenditure ceilings for 2018–20.

Nevertheless, there are several areas that can be enhanced in future years to better support fiscal policy decision-making. These include: a more comprehensive description of the main drivers of the macro and fiscal outlook, clarifying the characteristics of expenditure ceilings and explanation of their distribution among key spending units (KSUs), separately disclosing the baseline and target scenarios, quantifying the budgetary impact of new policies at the KSU level, reporting progress against key performance indicators, broadening the discussion of fiscal risks and their analysis, and reconciling differences between successive medium-term macro-fiscal forecasts and ceilings. Importantly, the draft Budget Declaration does not provide a sufficient level of information to demonstrate whether the expenditure ceilings for the general fund are consistent with achievement of the fiscal objectives. This should be addressed by including presentation of fiscal aggregates for the state budget (general and special fund) as well as for the general government sector.

Implementing the MTBF

The ability of the Ministry of Finance (MoF) to enforce compliance with ceilings during the annual budget will be critical to the success of the pilot exercise. The MoF will need to develop a robust report explaining the ceilings for cabinet discussion and for presentation to the legislature. It will also need to be empowered to execute the budget challenge function effectively by scrutinizing spending proposals to ensure consistency with the government’s overall policy and fiscal objectives. To ensure transparency, any changes to expenditure ceilings made as part of the annual budget process should be explained in the budget documentation.

The pilot exercise revealed some challenges that will need to be addressed to strengthen the roll-out of the full-fledged MTBF. The macroeconomic parameters and forecasts were finalized too late in the process, compressing the time available to settle the expenditure ceilings and negotiate these with line ministries (LMs). Policy costings were estimated by LMs on a constant price basis and there was no preparation of bottom-up baseline expenditure estimates to inform fiscal space available at the KSU level. There was also limited discussion between the MoF and LMs on expenditure ceilings to minimize any points of difference before their submission to the cabinet. The mission made several recommendations to address these challenges, including:

• preparation of two macro-fiscal forecasting rounds timed to best support the preparation of the Budget Declaration and the annual budget;

• the inclusion of standardized templates for line ministry submissions in the budget circular, that request a greater level of detail on cost drivers for existing programs; require LMs to report on progress against key performance indicators for existing programs; and to provide a stronger rationale for new policy proposals;

• adoption of a revised budget calendar, that clarifies the inputs required by the various departments and agencies involved in the budget process and ensures sufficient time is dedicated to bilateral negotiations between the MoF and LMs.

The Budget Code of Ukraine will also need to be amended to facilitate the roll-out of the full-fledged MTBF and the authorities have committed to doing so by the end of 2017. The April FAD technical assistance mission provided guidance on the nature of these amendments. In addition, the amendments to the Budget Code should include a more-finely tuned budget calendar; the core components required for the Budget Declaration, as well as a clear definition of the circumstances and rules for adjusting ceilings in response to macroeconomic developments.

Improving Fiscal Risk Disclosure

The authorities have been taking steps to improve their understanding and disclosure of fiscal risks and have made significant progress with the inclusion of a summary fiscal risk statement in the draft Budget Declaration. Expanding the fiscal risk statement, as the authorities intend, will require current information gaps to be addressed. As a priority, steps need to be taken to validate the existing data on SOEs to correct for errors and inconsistencies. In addition, systems need to be established or improved to collect information on a wider set of fiscal risks (such as legal cases and obligations under public private partnerships); monitor fiscal support for individual SOEs and identify and quantify their quasi-fiscal activities. To reduce duplication and enhance the consistency and reliability of SOE data, it will be important that a central process is established for the compilation of their financial information and fiscal support.

The institutional structures to support fiscal risk analyses and disclosure are also yet to be established. The Budget Code should be amended to formally assign responsibility for fiscal risk monitoring and assessment to the MoF and to require publication of a comprehensive fiscal risk statement with the Budget Declaration. The mandate of the Fiscal Risk Management Division will also need to be expanded and the responsibilities of departments and agencies involved in preparation of the fiscal risk statement should be clearly specified in the form of a cabinet resolution.

A. Background

1. The government has launched a pilot medium-term budget framework (MTBF) for the 2018 budget. Implementation of strategic planning and the MTBF is a core component of Ukraine’s PFM reform strategy.¹ Integral to this process is the preparation of the Major Areas of Budget Policy (referred to as the Budget Declaration), which the authorities have prepared for the first time for the period 2018–20. The Budget Declaration outlining the medium-term macro-fiscal projections, budget priorities, and expenditure ceilings was submitted to the Cabinet of Ministers of Ukraine (CMU) on June 1, 2017. The preparation of the Budget Declaration was supported by this mission, and an earlier FAD technical assistance (TA) mission in April 2017.²

2. The development of the Budget Declaration within a relatively short timeframe is a significant achievement. Until this exercise, medium-term macroeconomic and fiscal forecasts were not well integrated with budget planning which was mostly incremental and annual in scope. Furthermore, fiscal policy formulation lacked strategic direction in setting priorities and objectives and there was no “top-down” approach to guide it. The new Budget Declaration represents an important step forward in addressing most of these issues, given the tight timeframe for its development and the challenging political and macroeconomic environment. The authorities intend to build on it over time.

B. Expanding the Content of the Budget Declaration

3. Important elements of a MTBF were included in budgetary documentation, for the first time. The draft Budget Declaration was prepared within the boundaries of the current requirements of the Budget Code for the “Major Areas of Budget Policy.” This mission reviewed early drafts and proposed improvements to include more detailed discussion on the macro-fiscal context, an explanation of how the overall expenditure ceilings were derived, elaboration of key fiscal risks and further details of strategic objectives and new policies. Some of these recommendations were adopted in the 2018–20 Budget Declaration, which now includes:

• Some discussion of the macroeconomic outlook. The Budget Declaration presents the forecasts for GDP (and some components, such as exports), prices and employment and the assumptions underlying them, and summarizes the drivers of the previous year’s economic outcomes. In addition, it provides two alternative macroeconomic scenarios (optimistic and pessimistic) to highlight uncertainty around the baseline. But, only limited information is provided to set the context for the evolution of medium-term fiscal projections.

• A statement of fiscal objectives and the main policy indicators consistent with achieving them. This includes projections for 2018–20 for revenue, expenditure, and the budget balance. It also sets out the budget policy rules and key policy parameters (e.g., minimum wage, minimum subsistence level, floor for capital investment).

• Medium-term projections for revenue and main components. The revenue forecasts include detailed forecasts for revenue, broken down into the main tax categories (personal income tax, corporate income tax, value-added tax, and excises) and non-tax revenue, setting a basis for explanation of changes in medium-term revenue estimates to be presented in future iterations.

• A description of the main fiscal risks. For the first time, the authorities presented the key fiscal risks that could impact the budget. The mission worked with the authorities to develop a more fulsome narrative on the sources of fiscal risk and their significance to raise awareness of their potential impact on the budget. Highlighting the risks and uncertainty around the medium-term forecasts informs fiscal policy decision making, enhances fiscal credibility and helps frame fiscal policy discussions with the legislature. The initial summary fiscal risk section of the Budget Declaration focused on a discussion of macroeconomic and public debt risks as well as fiscal risks related state-owned enterprises (SOEs), guarantees and the financial sector.

• Description of new policy proposals and strategic goals. A list of new sectoral policies designed to enable implementation of the Sustainable Development Strategy “Ukraine-2020” is presented only in qualitative terms for the pilot³

• Inclusion of expenditure ceilings for key spending units (KSUs). These ceilings distribute the overall general fund expenditure envelope by each spending unit, setting the high-level framework for preparation of the detailed annual budget.

4. Nevertheless, there are several areas that can be further enhanced to better support budget planning and decision-making.⁴ It was not possible for the authorities to address these areas for this year’s declaration given the timeframe and availability of information, but as a priority, the next Budget Declaration (2019–21) should aim to include the following elements:

(i) A more detailed explanation of the medium-term macroeconomic and fiscal outlook.

• An expanded discussion of the key drivers of the medium-term macroeconomic forecasts to provide a better context for the fiscal forecasts;

• A discussion of the key drivers of the fiscal aggregates over the medium term, including the impact of macroeconomic developments, policy decisions, underlying cost drivers (such as the number of public sector employees and pensioners) and other one-off factors on the main fiscal aggregates as well as a discussion of the outcome for the previous year, the updated budget forecasts for the current year (which serves as the basis for the projections), and three-year projections for revenue, expenditure, and the budget balance;

• A description of the revenue outlook and its key components, including a more detailed disaggregation of the main drivers of revenue and further elaboration and quantification of the impact of specific tax policies during the three-year period;

• Reconciliations between successive vintages of macro-fiscal forecasts and ceilings clearly differentiating between adjustments arising from macroeconomic and demographic developments and those arising from policy changes;

• Separate presentation of the baseline (no policy change) and target (incorporating the effect of new policies) scenarios to help define consolidation needs or identification of possible fiscal space; and presentation of the key fiscal aggregates for the state budget (general and special fund), the consolidated general government sector, and social security funds.⁵,⁶ While local government budgets are not part of the MTBF, their disclosure is nevertheless important given that fiscal targets are currently defined on a general government basis under the IMF program, and to facilitate fiscal policy coordination.

(ii) An explanation of the expenditure ceilings.

• A description of how the overall expenditure envelope for the three-year period is derived (as a residual of forecasted revenue minus the fiscal target agreed under the IMF’s Extended Fund Facility);

• Clarification on the nature (fixed or indicative) of both the aggregate and KSU ceilings for the budget year and the two outer years and a broad explanation on how the individual ceilings were defined (i.e. which factors underlined the distribution of the overall ceiling);

(iii) Quantification of the budgetary impact of new policies at the KSU level.

• Quantification of strategic goals and new policies, including identifying the contribution of each KSU in their implementation. This would provide a more transparent justification of the expenditure ceilings;

(iv) Expanded discussion of fiscal risks.

• Presentation of a more comprehensive fiscal risk statement (FRS) alongside the declaration covering all material fiscal risks and deeper analysis of their significance (see Chapter III).

5. In future years, additional components could be added to further enhance transparency and better support policy-making. This could include:

• Explanation of differences between the government’s economic and fiscal projections and those of independent forecasters;

• Presentation of expenditure on an economic classification for the state budget and general government to facilitate its economic assessment;

• Inclusion of performance information with strategic objectives to hold KSUs accountable for achieving the objectives of the planned policies (see Chapter II, Section D); and

• Expanded quantification of fiscal risks, including assessment of their significance and likelihood of materialization.

C. Recommendations

Recommendation 1.1. Build on the 2018–20 Budget Declaration in future years to enhance the transparency and credibility of the MTBF.

• For the 2019–21 Budget Declaration, provide a discussion of the macroeconomic and fiscal outlooks in both qualitative and quantitative terms, clarify the ceilings’ characteristics and disclose a justification for the KSU distribution, separately disclose the baseline and target scenarios, quantify the budgetary impact of new policies at the KSU level, broaden the FRS to cover all material fiscal risks and strengthen analysis, present fiscal forecasts for the general government, and explain the changes to the previous vintage of macro-fiscal forecasts and ceilings (June 2018);

• For the 2020–22 Budget Declaration, and subsequent versions, introduce a comparison with forecasts of independent forecasters, improve the quantification of fiscal risks, disclose key performance indicators (first for outputs and later also for outcomes) for government policies (June 2019).

A. Background

6. An approved action plan to gradually implement a more complete MTBF for the 2019 budget has been developed. Over the course of 2017, the authorities will introduce new MTBF features and improve some of those implemented in the pilot, drawing on the lessons learned during that exercise and consistent with the constraints of the IMF program, which sets the basis for Ukraine’s current fiscal framework. The April 2017 FAD TA mission recommended some specific actions toward that objective, namely:

• Defining contingency margins for the overall expenditure ceilings to help ensure their credibility;

• Integrating the Special Fund into the MTBF and presenting medium-term fiscal aggregates for the general government;

• Gradually moving toward more fixed ceilings for overall and KSU expenditure as the system matures;

• Expanding the scope of the strategic discussion of fiscal policy, aided by a rules-based fiscal framework and enhanced information;

• Developing and implementing a technical amendments model that specifies clear rules for the adjustment of ceilings following macroeconomic changes; and

• Developing and disseminating a methodology for forward baseline estimates to inform the MTBF discussions between the MoF and KSUs.

7. Implementation of these improvements will need to be combined with procedural changes to enhance the credibility of the MTBF. The remainder of this chapter provides guidance on the steps required to implement the current pilot in the 2018 budget formulation process and provides guidance to help strengthen implementation of the fully-fledged MTBF based on the lessons learned thus far from the pilot exercise.

B. Next Steps to Implement the Pilot MTBF

8. Negotiating and agreeing to the expenditure ceilings will present a major challenge for successful implementation of the pilot MTBF. Line ministry (LM) participation in the preparation of the draft budget declaration was primarily limited to the provision of requests for additional funding for existing and new programs, which were costed using existing parameters. Updates of expenditure estimates to include revised parameters and compilation of the final aggregate ceiling was performed by the MoF with limited discussions with LMs (the discussions that took place mainly focused on clarifications of requirements of the March 2017 circular). Consequently, the presentation of the Budget Declaration to the CMU will be the first time LMs are informed of their expenditure ceilings and accepted policy priorities.

9. The MoF will need to explain and defend the ceilings during the CMU discussion, and this will require compelling evidence. The preparation of a robust briefing explaining the determination of the overall expenditure ceiling and its distribution across KSUs would support this. Furthermore, the MoF must be prepared to produce quick costings of new policies brought to the table during the cabinet discussions, not only for the budget year, but also for the two outer years. Any policies that take expenditure above the proposed overall ceiling, will need to be compensated for by reductions elsewhere or be proposed for rejection.

10. In addition, the entrenched practice of introducing new policies during the formulation of the annual budget may pose additional difficulties. Although the MTBF process was explained to all key stakeholders, there is a risk that LMs or the Verkhovna Rada will seek to introduce new policy measures during the process (including resubmitting those proposals already put forward, but not incorporated in ceilings in the lead up to the budget declaration’s approval). Mitigating this risk will require that the MoF’s budget officers develop and be effective in their budget challenge function (Box 1), and make any policy change dependent on a consensus obtained within CMU, rather than a bilateral discussion between the MoF and LM. This will also be important for implementation of the full-fledged MTBF.

11. The ambiguity regarding the nature of expenditure ceilings under the pilot may result in pressure to revise ceilings for the annual budget. Between now and the approval of the annual budget, there may be developments that result in changes to the macroeconomic and demographic assumptions underpinning the macro-fiscal forecasts. No clear rules have been disclosed about how such developments will be addressed, which could add pressure to ceilings above and beyond those resulting from changes in these parameters.

12. To ensure transparency, any revisions to ceilings between the approved Budget Declaration and annual budget should be explained. The annual budget should include a summary reconciliation, detailing the main components of any changes to ceilings (using the template provided in Table 1.1. of the March/April 2017 TA Mission report). In addition, the MoF should disclose in an annex to the 2018 Budget Proposal, explanatory notes on the detailed adjustment of each ceiling.

Table 1.1.

State Guarantees

Table 1.1.

State Guarantees

		2011	2012	2013	2014	2015	2016
Total Guaranteed Debt		115.9	116.3	104.6	153.8	237.9	278.9
Percent of GDP		8.9	8.3	7.1	9.7	12.0	11.7
Domestic Debt		12.3	16.2	27.1	27.9	21.5	19.1
Percent of GDP		0.9	1.2	1.9	1.8	1.1	0.8
	Debt securities issued on the domestic market	5.8	10.0	21.1	21.6	16.4	16.0
	Due to banks and other financial institutions	6.5	6.2	6.0	6.3	5.1	3.1
External Debt		103.6	100.1	77.4	125.9	216.4	259.8
Percent of GDP		8.0	7.1	5.3	7.9	10.9	10.9
	Loans from International Financial Institutions	61.5	40.6	16.2	40.1	140.8	190.9
	Loans from Governments of foreign countries	1.5	2.0	2.0	3.8	4.7	4.0
	Loans from foreign banks and other foreign financial institutions	16.8	29.3	31.0	51.6	68.2	62.0
	Outstanding securities in foreign markets	22.8	27.2	27.2	28.5	0.0	0.0
	Other categories	1.0	1.0	1.0	1.9	2.7	3.0

View Table

Table 1.1.

State Guarantees

		2011	2012	2013	2014	2015	2016
Total Guaranteed Debt		115.9	116.3	104.6	153.8	237.9	278.9
Percent of GDP		8.9	8.3	7.1	9.7	12.0	11.7
Domestic Debt		12.3	16.2	27.1	27.9	21.5	19.1
Percent of GDP		0.9	1.2	1.9	1.8	1.1	0.8
	Debt securities issued on the domestic market	5.8	10.0	21.1	21.6	16.4	16.0
	Due to banks and other financial institutions	6.5	6.2	6.0	6.3	5.1	3.1
External Debt		103.6	100.1	77.4	125.9	216.4	259.8
Percent of GDP		8.0	7.1	5.3	7.9	10.9	10.9
	Loans from International Financial Institutions	61.5	40.6	16.2	40.1	140.8	190.9
	Loans from Governments of foreign countries	1.5	2.0	2.0	3.8	4.7	4.0
	Loans from foreign banks and other foreign financial institutions	16.8	29.3	31.0	51.6	68.2	62.0
	Outstanding securities in foreign markets	22.8	27.2	27.2	28.5	0.0	0.0
	Other categories	1.0	1.0	1.0	1.9	2.7	3.0

C. Strengthening the Process for the Full-Fledged MTBF

13. Though not complete, the pilot exercise has thus far highlighted some challenges that will need to be addressed to strengthen the roll-out of the fully-fledged MTBF. Some of these arise from design or technical issues, while others are more procedural in nature. The main challenges relate to: the timing and preparation of the macroeconomic forecasts; the lack of separation between baseline and new policy scenarios; the role of LMs and sequencing of tasks; the documentation requested from KSUs as input to the budget declaration; and, the supporting legal framework. The remainder of this section provides further guidance to address these challenges. In addition, it details some of the issues that will need to be considered in the design of the technical amendments model, which will be critical for the implementation of the fully-fledged MTBF.

Information to Support the Process

22. The information requested from LMs for the preparation of the budget declaration has some limitations. The March budget circular requested information from LMs on costings of new policy initiatives (on a constant price basis), additional funding required for existing programs resulting from changes in legislation or regulations as well as suggested savings. In addition, not all templates required were provided along with the budget circular, and, as a result, submissions differed across LMs (for example, review of submissions provided by the Ministries of Health and Education showed substantial differences in the presentation of information).

23. The preparation of LMs’ submissions and bilateral discussions would be enriched by a revised information template. This should include a comparison of the estimated actual LM expenditure ceiling with the ceiling decided in the previous medium-term Budget Declaration. This would reduce the resources currently devoted to the preparation of unrealistic budget bids or lower priority initiatives. Table 1 provides an illustration of the information that could be provided by LMs to justify their expenditure proposals, and Annex 2 provides a template setting out the information requirements for new policy submissions.

Table 1.

Proposed Template for KSU Expenditure Ceiling Proposal

¹The template is set at LM level. LMs collect the required information from the KSUs.

²The starting point is the existing ceiling decided in the previous medium term Budget Declaration.

³To reach the baseline estimate, the existing ceiling is updated to take into account of i) new commitments that the Verkhovna Rada or the Government has made, ii) technical adjustments within the budget and between the budget appropriations resulting from e.g. the changes in the structure of the budget line items iii) changes in volumes and costs that result from changes in the respective updated macroeconomic indicators. The changes in volumes concern e.g. the changes in the number of citizens entitled to benefits or allowances according to the existing social legislation. The changes in costs result e.g. from the indexation of social benefits to prices or wages.

⁴New policy initiatives are added line by line on top of the baseline to reach the proposal for the actual expenditure ceiling. More detailed calculations, further arguments or explanations and any other relevant information to support the assessment of the fiscal and broader economic impact of the proposed new initiatives should be presented in a separate annex.

Table 1.

Proposed Template for KSU Expenditure Ceiling Proposal

Ministry […]1
		2016 actual	2017 budget	2018 ceiling	2019 ceiling	2020 ceiling
Actual/Budgeted		1000	1050
Ceiling acc. to Previous MT Budget Declaration2				1000	1000	1000
New commitments3.i, of which				100	100	100
	by VR			50	50	50
	by Government			50	50	50
Structural adjustments within the Budget3.ii
Automatic adjustments3.iii, of which				20	35	50
	volume factors			10	20	30
	cost factors			10	15	20
Estimated baseline				1120	1135	1150
New policy initiatives, of which4				70	100	30
	Initiative I			70	70
	Initiative II				30	30
	…
Overall impact of baseline and policy changes				1190	1235	1180

¹The template is set at LM level. LMs collect the required information from the KSUs.

²The starting point is the existing ceiling decided in the previous medium term Budget Declaration.

³To reach the baseline estimate, the existing ceiling is updated to take into account of i) new commitments that the Verkhovna Rada or the Government has made, ii) technical adjustments within the budget and between the budget appropriations resulting from e.g. the changes in the structure of the budget line items iii) changes in volumes and costs that result from changes in the respective updated macroeconomic indicators. The changes in volumes concern e.g. the changes in the number of citizens entitled to benefits or allowances according to the existing social legislation. The changes in costs result e.g. from the indexation of social benefits to prices or wages.

⁴New policy initiatives are added line by line on top of the baseline to reach the proposal for the actual expenditure ceiling. More detailed calculations, further arguments or explanations and any other relevant information to support the assessment of the fiscal and broader economic impact of the proposed new initiatives should be presented in a separate annex.

View Table

Table 1.

Proposed Template for KSU Expenditure Ceiling Proposal

Ministry […]1
		2016 actual	2017 budget	2018 ceiling	2019 ceiling	2020 ceiling
Actual/Budgeted		1000	1050
Ceiling acc. to Previous MT Budget Declaration2				1000	1000	1000
New commitments3.i, of which				100	100	100
	by VR			50	50	50
	by Government			50	50	50
Structural adjustments within the Budget3.ii
Automatic adjustments3.iii, of which				20	35	50
	volume factors			10	20	30
	cost factors			10	15	20
Estimated baseline				1120	1135	1150
New policy initiatives, of which4				70	100	30
	Initiative I			70	70
	Initiative II				30	30
	…
Overall impact of baseline and policy changes				1190	1235	1180

¹The template is set at LM level. LMs collect the required information from the KSUs.

²The starting point is the existing ceiling decided in the previous medium term Budget Declaration.

³To reach the baseline estimate, the existing ceiling is updated to take into account of i) new commitments that the Verkhovna Rada or the Government has made, ii) technical adjustments within the budget and between the budget appropriations resulting from e.g. the changes in the structure of the budget line items iii) changes in volumes and costs that result from changes in the respective updated macroeconomic indicators. The changes in volumes concern e.g. the changes in the number of citizens entitled to benefits or allowances according to the existing social legislation. The changes in costs result e.g. from the indexation of social benefits to prices or wages.

⁴New policy initiatives are added line by line on top of the baseline to reach the proposal for the actual expenditure ceiling. More detailed calculations, further arguments or explanations and any other relevant information to support the assessment of the fiscal and broader economic impact of the proposed new initiatives should be presented in a separate annex.

24. There is also room for improvement in the performance information used to support the MTBF. While performance-related information was requested from KSUs as part of their submissions under the pilot exercise, the indicators submitted were largely drawn from LMs current budget passports and primarily consisted of input and output indicators. Furthermore, as in previous years’ budget exercises, the performance information provided did not contain analysis of whether programs were on track to achieve performance targets set in previous years. The use of LMs medium-term budget submissions for policy prioritization, would be greatly enhanced with the inclusion of: (i) measurable performance indicators; and (ii) assessment of performance against the targets set in previous years. As the MTBF is more solidly established, consideration could be given to the development of more outcome orientated indicators over the medium term (Annex 3 describes the link between program objectives, outcomes, and outputs and provides an example of program indicators based on outcomes). Reporting progress against performance indicators will contribute to more effective allocation of resources and enhance accountability in the use of public funds.¹¹

D. Recommendations

Recommendation 2.1. Reinforce the credibility of the MTBF, by strengthening the pilot exercise’s controlling mechanisms. Specifically:

• Develop a robust explanatory note supporting the ceilings proposed in the Budget Declaration (June 2017);

• Build budget challenge capacity of Budget Officers and empower them to enforce that capacity during the 2018 Budget formulation (July 2017); and

• Transparently disclose ceiling revisions and their main drivers in an annex to the 2018 Budget P ro posal (September 2017).

Recommendation 2.2. Improve and expand the inputs for the MTBF (June 2018)

• Produce two macro-fiscal forecasts to inform planning, the baseline scenario (existing policies) at the beginning of the process (March) and an updated scenario including new policy initiatives decided by the CMU (May-June).

• To facilitate preparation of baseline scenarios, include macroeconomic parameters in the March budget circular and request LMs to submit information on their implications for existing policies;

• Design and circulate common templates for the LMs to explain deviations and submit information on factors driving the costs of existing programs and to support rationale and costings for new programs;

• To support the bilateral negotiations between the MoF and LMs, require the LMs to produce the first draft for the KSU ceilings under their branch.

• To strengthen the information base for policy prioritization, require LMs to report progress against key performance indicators in their budget submissions and end-year reports.

Recommendation 2.3. Clarify the roles and interaction of the stakeholders in the process for the MTBF.

• Ensure sufficient time is dedicated to bilateral negotiations between the MoF and LMs on KSU expenditure ceilings (April 2018).

• Give LMs a stronger mandate to plan and manage portfolio finances, within the approved LM ceiling, by requiring that they prioritize the requests of lower level spending units and negotiate with the MoF on their behalf (May 2018).

A. Background

31. The government is taking steps to improve its understanding of fiscal risks and their disclosure. It has established a Fiscal Risk Management Division (FRMD) in the MoF with a mandate to oversee SOE risks, established a methodology for assessing SOE-related fiscal risks, commenced disclosure of fiscal risks in the Budget Declaration, and committed to further strengthening disclosure and capacity to assess fiscal risks in its PFM reform Strategy (2017–21). The April FAD TA mission provided an assessment of fiscal risks and mapped out a phased approach to the development of a comprehensive FRS (see Annex V of that report). This chapter builds on the recommendations of that mission, by further elaborating on the steps required to expand fiscal risk disclosure for the 2018 budget and subsequent years.

B. Expanding Fiscal Risk Disclosure

32. The MoF is well placed to build on the current fiscal risk disclosure. The April 2017 FAD TA mission recommended that the 2018 FRS focus primarily on SOE related fiscal risks, and include discussion of macroeconomic risks, public debt exposures, and guarantees. The Budget Declaration, largely fulfilled this, although not all the elements related to disclosure of SOE risks envisaged by the previous TA mission were fully incorporated. This largely reflected the summary nature of the FRS included in the Budget Declaration, an intention decision so that it aligned with the broader structure of that document. The authorities intend to expand disclosure of SOE related risks, and present this as a separate statement as part of the 2018 budget documentation. To guide this process, the mission prepared and shared with the authorities a draft FRS (Annex 4). It may not be possible to incorporate all of its components in time for the 2018 budget, but the authorities are well placed to build on its reporting in future years:

• The FRMD has developed a methodology to assess risks related to SOEs and has collected and begun to analyze data on 48 SOEs that pose the most significant fiscal risk;

• The MEDT’s macroeconomic forecasting department already includes alternative macroeconomic scenarios in its report to the CMU, which could be used as the basis for developing alternative macro-fiscal projections; and

• The debt department maintains records on government guarantees and has almost finalized its debt management strategy, which includes discussion of key risk exposures.

33. There are, however, some information gaps that will need to be addressed.

• There is a pressing need to validate the information collected by the FRMD on the group of 48 SOEs before it can be used as a basis for fiscal risk analysis, as a number of internal errors and inconsistencies in the data submitted by LMs were identified by the mission, along with some inconsistencies with other data sets (such as guarantees and budget subsidies). In addition, the FRMD indicated that forward looking information provided by SOEs on the basis of their strategic plans (2018–20) is not available for all 48 SOEs and does not provide a reliable basis for forward-looking assessment.

• The information on SOEs contained only a limited number of line items from the financial statements (e.g. total assets, debt and equity) and pre-calculated financial ratios. This limits the analysis that can be performed. The full set of financial information should be obtained to ensure data reliability can be assessed and to support more enhanced risk assessment methodologies over time.

• While aggregate data on guarantees and on-lending is available, the FRMD will need regular access to information on outstanding amounts, execution of guarantees, and loan performance for individual SOEs. This will likely require the debt department to augment its existing records on guarantees by creditor, to also include information and performance of guarantees by individual SOE.

• Information on quasi-fiscal activities of SOEs remains limited and unquantified, as identified in the March/April FAD TA report.

• For the future development of the FRS, systems will also need to be established to collect information and input on a wider range of fiscal risks – legal cases; public-private partnerships (direct obligations over their life cycle and contingent liabilities), local government liabilities by municipality and their contingent liabilities, and natural disasters -although this is a more medium-term exercise.

34. Formally establishing the necessary institutional structures in the MoF and other key agencies would smooth the coordination and preparation of the FRS. Whilst the FRMD must lead the process for monitoring and managing fiscal risks, they need to coordinate inputs from a number of stakeholders both within the MoF as well as other agencies. In particular, a close working relationship with the MEDT is required both with respect to the assessment of macroeconomic risks and also in assessing the risks arising from SOEs. The April FAD TA mission provided a suggestion for the allocation of responsibilities across departments and agencies, and recommended that these be formalized in a CMU resolution, but this is yet to be actioned. It also recommended amending the BCU to formally assign the role of fiscal risk monitoring and assessment to the MoF and expanding the mandate of the FRMD to cover all material fiscal risks as part of the amendments to be made later this year.

35. The FRMD will need to have a team dedicated to fiscal monitoring and management to fulfill this mandate. The FRMD currently has responsibilities for a large range of tasks (for example, in addition to overseeing large SOEs and those under the MoF, it also has responsibilities for procurement, regulation and state aid). It will be important that a dedicated team is assigned and accountable for fiscal risk monitoring, assessment, reporting and mitigation on a full-time basis, otherwise there is a risk these tasks are set aside by other competing demands. At the same time, there is a need to enhance the analytical capacity of FRMD over time, so that it can deepen its fiscal risk analysis.

36. Current institutional arrangements result in duplication in the collection of information on SOE financial performance. The MoF, MEDT, and LMs are all collecting and compiling financial information related to SOEs―an onerous task, given there are more than 1,800 operational SOEs. In addition, the MoF is compiling information on various forms of fiscal support provided to individual enterprises. The various ministries should work to harmonize the process of collecting information, with the aim of establishing a single process for data collection that meets the requirements of all stakeholders, including the timeframe for its provision. The development of a single central excel database, available to relevant stakeholders, containing the financial information and state support for the most significant SOEs and verified by specialist departments and SOE governing authorities would be a good first step. Certain supplements may still be required for information only relevant to targeted SOEs, so as not to over-burden companies. Over the medium-term, implementation of a central, electronic reporting platform capturing both financial and non-financial information that is made available to all stakeholders would help reduce reporting burdens on enterprises and ensure consistency and reliability of data.

37. While the fiscal risk methodology developed by FRMD for SOEs provides a good base for risk assessment, there may be scope to enhance this overtime. The current methodology, set out in a CMU resolution, determines the process for selecting the SOEs that will be subject to more detailed analysis and the approach to ranking them into risk categories to assess the government’s risk exposure. In addition to the recommendations provided during the 2017 April TA mission to improve the methodology, consideration could also be given to:¹⁴

• Widening the set of ratios used to rank SOEs into risk categories, over time, to better pin point potential sources of risk specifically to include ratios to identify potential liquidity (e.g., current and interest cover ratios) or solvency challenges;

• Including, in the CMU resolution, the principles informing the selection of the SOEs for in-depth assessment (for example, macro-critical entities; those with liabilities above a certain threshold or those ranked among the highest risk categories). Further, the decree should provide the MoF with the discretion to require in-depth assessment of other entities, where potential fiscal risks are identified;

• Comparing actual outcomes to the forward-looking projections to determine whether there are persistent biases in the projections provided by SOEs (and therefore under or overstatement of risk assessments based on forward indicators);

• Collecting supplementary data on all accrued arrears (in addition to wages and taxes); cash flow projections, capital expenditures, borrowing plans, and any contingent liability exposures, to enable more in-depth analysis of specific sources of risk;

• Using historical information to identify trends pointing to a deterioration in the financial performance or position of an SOE over time; and

• Meeting with the SOEs, supervisory authorities and regulators to allow a deep dive into potential drivers of fiscal risk.

38. The methodology for categorizing SOEs by risk should also be reviewed against future outcomes to ensure that it is providing appropriate forward-looking guidance. Where unexpected deviations from budget forecasts are caused by SOEs, the most recent risk categorization of the relevant SOEs should be noted. If they were not already identified as high sources of potential risk (e.g. risk category 1 and 2), this could indicate weaknesses in the methodology’s predictive ability which should be addressed through refinements. Further, the methodology ranks highly leveraged, but profitable SOEs as more risky than unprofitable enterprises with lower leverage. This may well be appropriate, but it also seems plausible that some loss-making SOEs that are not highly leveraged may be more likely to generate fiscal risks than others that are more highly leveraged and have a track record of good profitability. In undertaking future reviews of the methodology, consideration should be given to introducing graduated thresholds for the ratios used to group entities into risk categories.

39. Further, relying on SOEs and supervisory authorities to self-evaluate fiscal risks may not generate accurate or consistent outcomes. Rather than requiring SOEs to self-assess risks, consideration should be given to soliciting specific information that would enable the FRMD to assess such risks (e.g. information on pledged assets; upcoming maturities of debts; or requiring SOEs to perform sensitivity analyses of their cash flows and balance sheet to changes in key parameters (e.g. exchange rate and commodity prices). Moreover, templates should be designed and provided to the SOEs that allow such information to be aggregated across all SOEs.

C. Recommendations

Recommendation 3.1. Verify the financial information already collected from SOEs and establish mechanisms to enhance the reliability of data.

• Verify the 2016 financial information for the 48 SOEs against their financial statements, and check the consistency with other data sources including the MEDT database for the top-100 SOEs, debt department data on guarantees and budget data on subsidies (July-2017);

• Streamline the collection of financial information from SOEs by establishing a single database (excel-based initially) for financial information, non-financial indicators and state support for SOEs to be made available use by all stakeholders (March-2018).

Recommendation 3.2. Approve the updated methodology for fiscal risk assessment of SOEs, based on a richer information set.

• Update and approve fiscal risk methodology, in line with the improvements suggested in this report and by the March/April FAD TA mission (June 2017);

• Review the risk categorization against realized outcomes (first review end-2019).