Technical Assistance Report-Supporting Implementation of the Expenditure Rule Through Public Financial Management Reforms

This Technical Assistance Report discusses the recommendations for supporting implementation of the expenditure rule through public financial management reforms in Brazil. To address rigidities in the budget and to create more fiscal space, authorities need to establish processes to build flexibility and improve efficiency. These include reviewing mandatory expenditures and indexation practices. Other options include a rolling program of spending reviews for large and fast-growing programs and efficiency dividends on ministries’ operational expenses. Authorities also need to begin to develop a medium-term budget framework to support strategic decision making.


This Technical Assistance Report discusses the recommendations for supporting implementation of the expenditure rule through public financial management reforms in Brazil. To address rigidities in the budget and to create more fiscal space, authorities need to establish processes to build flexibility and improve efficiency. These include reviewing mandatory expenditures and indexation practices. Other options include a rolling program of spending reviews for large and fast-growing programs and efficiency dividends on ministries’ operational expenses. Authorities also need to begin to develop a medium-term budget framework to support strategic decision making.

I. Background and Reform Progress

1. The 2000 Fiscal Responsibility Law (FRL) remains the backbone of Brazil’s fiscal framework. This law sets the framework and procedures for fiscal management and the budget process. With the objective of containing fiscal deficits and rising debt, the FRL introduced overall targets (primary balance) and limits on the wage bill, debt and guarantees for all levels of government. The primary balance targets are set every year at the beginning of the budget process.

2. The primary balance target is specified in the PLDO for the current year and, indicatively, for two additional years. The PLDO is sent to Congress for approval in April. It includes a statement of the budget’s goals and the rules governing cost control and evaluation of programs. The PLOA is submitted by end-August and contains the more detailed statement of annual budget programming compatible with the PLDO.

3. During the 2000s this framework, assisted by high levels of economic growth, was successful in providing fiscal discipline at both the Federal and lower levels of government. The law enabled a sizeable reduction of federal government net debt and control of payroll expenditures. During this period of high economic growth, however, buoyant revenues masked underlying weakness in the system.

4. Even prior to the recession, a number of cracks were emerging with the framework. These included unrealistic budgets, based on optimistic revenue forecasting and Congress’ use of a Constitutional loophole to correct “errors and omissions” to further inflate revenue forecasts. This created artificial headroom for spending priorities, while allowing Congress to remain within the bounds of the primary balance target. Fiscal slippage was averted using Presidential Decrees to contain expenditure execution. Thus, fiscal discipline has come at the cost of smooth and predictable budget execution. Other weaknesses include widespread earmarking of revenues, large mandatory spending items, generous indexation practices and a fragmented budget process with a short-term focus. Brazil possesses various aspects of a medium-term framework; however, these too are fragmented and do not guide the budget process nor do they drive strategic decision making on fiscal policy and budgeting.

5. The government currently faces a very challenging and complex economic and political environment. In 2016 the overall fiscal deficit was projected to reach 9.7 percent of GDP with a non-financial public sector (NFPS) primary deficit of 2.6 percent and a public sector gross debt of 74.8 percent of GDP.1 In addition to adopting structural measures, mainly social security reform which is ongoing in Congress, the government has also adopted reforms to the fiscal framework.

6. Within the past year solid progress has been made in introducing major reforms to the fiscal framework, which are complementary to the FRL. These include the new medium-term expenditure rule (MTER) enshrined in the Constitution (see Chapter II), and the establishment of the IFI in December 2016. The IFI is based in the Senate. It has a four-member board (one of whom has been appointed) and five permanent staff members. Its remit includes regular analysis of the government’s fiscal performance; elaboration of medium-term forecasts of macro parameters, fiscal indicators and fiscal impacts; and advice to the Congress on fiscal issues. The Treasury is improving the quality of its fiscal forecasts and new models have been developed. It is using its advanced capacities to expand the LDO’s fiscal risk annex to include fan charts and more scenarios.

7. Many other initiatives are ongoing. A new public finance law to replace the existing 1964 legislation has passed the Senate and will soon be placed before the lower house. Its goals are to modernize and standardize budgetary processes in line with international best practice, and to bring a medium-term perspective and greater realism to budgeting. In addition, several committees have been set up by MOF to oversee evaluation of major investments, major existing expenditure programs, and tax expenditures. The Federal Budget Office (SOF) has started to conduct spending reviews.

8. These reforms are in line with key recommendations from the two most recent Fiscal Affairs Department (FAD) missions. A March 2016 mission focused on Developing a More Resilient Fiscal Framework, and the June 2016 mission conducted a Fiscal Transparency Evaluation. The government has acted swiftly in implementing some recommendations from these reports. As discussed above, in the last 12 months the authorities have adopted an IFI and an expenditure rule and have made important improvements to the government’s Fiscal Risk Statement.

9. Moving ahead the government’s focus is on implementing the new expenditure rule. It hopes this rule will help address some of the weaknesses in the system. Mainly that the rule will make the budget more realistic by imposing a multiyear ceiling on expenditures and improve expenditure prioritization during the budget preparation phase. The government now must combine the use of both primary balance targets and expenditure ceilings in its fiscal planning strategy. The authorities consider that under the current scenario, the primary balance targets will be more binding until 2019. Come 2020 though, due to the expected trajectory of mandatory expenditure, it will be difficult for the authorities to comply with the expenditure rule without some additional structural adjustments.

10. Building on previous FAD work, this report focuses on institutional and procedural reforms that are needed to support implementation of the new expenditure rule. The rule necessitates a more medium-term focus for managing fiscal policy and budgeting. The remainder of this report is divided into five chapters: Chapter 2 describes the new expenditure rule in detail and the challenges with implementation; Chapter 3 discusses sources of budget rigidity and options for improving spending efficiency; Chapter 4 looks at strengthening medium-term frameworks to support fiscal sustainability and strategic budget decision making; Chapter 5 examines how the new rule can be embedded within the budget process; and Chapter 6 considers reporting and procedural measures to ensure compliance with the new rule.

II. Implementing A New Medium-Term Expenditure Rule in Brazil

11. In December 2016, Brazil enshrined a “New Fiscal Rule” in its Federal Constitution. Constitutional Amendment No. 95, adopted on December 15, 2016 establishes a new expenditure rule for the next twenty years (see Table 2). This MTER essentially stabilizes real primary spending (with some exceptions) at its 2016 level. This chapter looks at (A) the rationale for the adoption of such a rule, (B) its contents, and (C) the potential implementation challenges.

Table 2.

Articles Included in Constitutional Amendment 95

Constitutional Amendment 95 (2016), which institutes the New Fiscal Regime, adds nine articles to the Transitory Constitutional Provisions Act. This amendment allows for a time-bound implementation.

article image
Source: mission.

A. Rationale for the Adoption of a Medium-Term Expenditure Rule

12. The new fiscal regime and the MTER are primarily aimed at improving fiscal discipline and rebuilding fiscal credibility. The rapid collapse in tax revenues and ever strong expenditure growth amidst a deep economic recession have made both year-to-year and in-year revisions of the fiscal targets necessary, casting doubt on the credibility of the fiscal framework. Structural reform is needed to deal with the growing weight of social security and pension liabilities, which in the medium to long term will seriously endanger Brazil’s fiscal position if no action is taken. In this context, the MTER adopted in December 2016 is a clear and ambitious signal to the public and the markets that the government is committed to achieving fiscal discipline: (1) clear, because the rule enjoys the highest legal status; and (2) ambitious because introducing the structural reforms and overcoming the many rigidities of public expenditure in Brazil will not be an easy task.

13. The expenditure rule is also intended and expected to change the political economy of budget preparation, approval, and execution. The rule looks to reduce several inefficiencies in the budget procedure, that are specific to Brazil.

  • The setting of fiscal targets in the LDO will be guided more by the availability of resources than in the past. Since primary expenditure is now constrained, reliable revenue projections will be crucial in setting credible and sustainable primary balance targets, especially for the budget year.

  • Then, in the annual budget preparation process, the expenditure ceiling should bring about more realistic budget proposals from line ministries which take better account of overall fiscal constraints. They should get a clearer sense, earlier in the year, of the envelopes they will ultimately receive.

  • In Brazil, the Congress has traditionally inflated revenue forecasts in order to accommodate their proposed new expenditure items. Since the rule now caps the level of expenditure, there will be less incentive for such budget-inflating amendments. The debate should shift to reallocation of resources within ceilings.

  • Finally, the execution phase has been characterized by significant sequestrations in order to bring the budget back to a more realistic footing. By improving the credibility of the budget, sequestration requirements should decrease. And applying similar ceilings to both appropriations and payments, should contribute to limiting the formation of new arrears.

14. The expenditure rule finally represents a significant shift to a more medium-term perspective. This is the first time a binding medium-term fiscal rule is implemented in Brazil. Targets set for the primary balance of the non-financial public sector and of the federal government in the annual LDO are still valid. This procedural rule, however is only binding for the budget year, as in practice, the targets for the forward years are modified in the subsequent LDOs. The expenditure rule is meant to remedy the traditional short-term focus of budget preparation, potentially shifting the debate to medium-term fiscal planning.

B. Key Features of the New Fiscal Regime

15. The new expenditure rule forms the core of the new fiscal regime (see Table 2 for a list of all its provisions). The rule sets ceilings on actual primary expenditure for the next twenty fiscal years, with a mid-term review process. The trajectory is defined as follows: the base is the total of primary expenditures paid in 2016; the annual growth rate then applied to this base is the 12-month inflation rate, for the period ending in June of the year prior to the budget year.2 In other words, the rule freezes real primary expenditure roughly at its 2016 level. The ceiling is applied to actual payments including RAP, or unpaid commitments carried over.3 The ceiling also applies to appropriations voted in the annual budget. The rule applies to all branches of federal government—the Executive branch, the Legislature, and the Judicial branch—each branch will have its own ceiling within the overall expenditure ceiling. It excludes extraordinary credits used for urgent and unpredictable events, such as natural disasters, war and civil unrest. The rule is binding from the fiscal year 2018.

16. The rule precisely defines the scope of primary expenditures that fall under the ceiling (see Figure 1). The constitutional amendment explicitly excludes a series of specific federal spending categories, namely constitutional transfers (essentially to states and municipalities) and bailouts of non-financial state-owned enterprises.

Figure 1.
Figure 1.

Coverage of the New Expenditure Rule

Citation: IMF Staff Country Reports 2017, 292; 10.5089/9781484321218.002.A001

Source: IMF Staff. Note: The proportions used here are roughly representative of the shares of each type of expenditure.

17. While changing them, the new rule stays true to the spirit of preexisting constitutional provisions targeted at specific sectors and branches.

  • The protection of spending for education and health defined in the Federal Constitution is maintained. However, the mechanism has changed. Previously, education and health spending were earmarked as a minimum percentage of revenue (respectively 18 percent of net tax revenue and 15 percent of net current revenue). From 2018 on, the new rule defines spending floors for education and health; expenditure for each sector should be superior to their respective shares in the 2017 total primary expenditure ceiling.

    Additional funds can be allocated beyond these amounts, as long as it is within the overall expenditure ceiling.

  • For the first three years of the new rule, the legislative and judiciary branches are given the ability to go above their respective expenditure ceiling for a maximum cumulative total of 0.25 percent of the Executive branch’s ceiling. This excess should be offset by a corresponding decrease in the Executive branch’s budget.

18. The new rule sets safeguards to ensure compliance.

  • Ex ante, the regime includes the new obligation to accompany any legislative proposal for new mandatory expenditure or new tax expenditure with an assessment of its fiscal and financial impact. The regime also allows the Congress to block for 20 days and review any new proposed bill for which there is suspicion that it may not be compatible with the rule.

  • Ex post, the regime defines a wide range of adjustments to be automatically applied in case of non-compliance with the rule. These adjustments include freezes on new mandatory spending items, on salary increases and new hires, on the expansion or creation of credit lines, programs or tax incentives and on increases to mandatory spending that go beyond inflation.

C. Implementation Challenges

19. Even though the rationale for the rule is clear, Brazil faces challenges and risks that may impede the effective implementation of the rule. The rule poses: (i) expenditure policy challenges, as it requires structural reforms, increased expenditure flexibility and improved macroeconomic fundamentals for Brazil to fully benefit from it; (ii) political economy challenges, as the stakeholders may resist the necessary changes in budget procedures; and (iii) a risk that the rule is not effective in the end, due to creative evasion mechanisms.

20. The major policy challenge for Brazil will be to introduce structural reforms and reduce the rigidity of its expenditure. As indicated in Table 3, even with the passage of the new proposed social security reform, the government will need to do future adjustments to reach a balanced budget in the medium term, as the financial benefits from social security reform will not emerge immediately. Also, come 2020, given the trajectory of mandatory expenditure, the government already projects that without additional adjustments, it will be difficult to keep to the expenditure rule. Without finding a way to overcome the rigidity of spending (See Chapter III), either (i) the rule will not be complied with or it will be amended; or (ii) the rule will be complied with but at the expense of public investment.

Table 3.

Medium-Term Projections for Primary Revenue and Expenditure of the Federal Government

(R$ BN)

article image
Source: Independent Fiscal Institution (IFI).Note: The IFI carries out its own macro-fiscal forecasts. Here, revenues are projected to grow in line with nominal GDP, while expenditures grow with inflation (i.e., reach the expenditure ceiling), under the assumption that the social security reform has taken place.

21. Compliance with the rule may prove difficult if economic growth does not improve steadily. Indeed, expenditure rules are especially efficient in good times, because they constrain countries into countercyclical fiscal policy; though revenue may grow, expenditure will not. However, in times of slow or negative economic growth, expenditure rules can limit the capacity of the government to engage in fiscal stimulus to the economy through expenditure, even temporarily. This is a risk for Brazil as the new expenditure rule does not contain an economic “escape clause”. An economic escape clause would define the strict conditions for temporarily suspending the rule due to extreme economic situations and the process for going back to the rule when the situation ends.

22. The MOF and the Ministry of Planning and Budget (MPB) should be proactive in preparing all stakeholders in the budget process for the consequences of the rule. The drafting of the expenditure rule has involved the highest authorities in the Executive, as well as the Congress and the Senate. Explanations about the strategy of the Executive to comply with the rule have been scarce since its adoption. Line ministries may however not be fully aware of what this rule may entail for their budgetary envelopes in the next few years. This potentially puts the budget negotiations at risk. The financial ministries (Finance, Planning) and the Presidency have a responsibility to work towards greater coordination with line ministries and clearer strategic communication to the public, in order to alleviate the political cost involved with fiscal consolidation.

23. International experience shows that budget stakeholders may be naturally tempted to look for ways around rules and their objectives. As compliance with the rule may prove to be difficult, there may be attempts to work around it or to transform its purpose.

  • As previously shown in Figure 1, the expenditure rule only covers a limited part of the non-financial public sector expenditure. Fiscal discipline may be hindered by increases in spending items excluded from the ceiling, or by the creation of new tax expenditures, in spite of apparent compliance with the MTER. This is especially true if the primary balance target is adjusted every year to make room for these measures.

  • Expenditure rules contribute more to fiscal sustainability when they are combined with other types of medium-term rules. There is no medium-term debt anchor or target complementing the current rules. Thus, there is no mechanism which would ensure that when working around the expenditure rule (as described above), Brazil does not risk a ramp-up in public debt. Medium-term expenditure ceilings usually go together with debt rules, as evidenced by international experience (see Box 1).

  • The Treasury has been mindful in monitoring and successful in reducing the stock of RAP in the last few years. However, even if the rule prevents appropriations from crossing the expenditure ceiling, nothing prevents from theoretically providing line ministries, at the sequestration stage, with more commitments than payments, which could lead to arrears.

International Experience in Adopting Expenditure Rules

The most common combination of fiscal rules is a debt and a budget balance rule. As shown on Figure 1, in 2015 out of 96 countries covered by the IMF Fiscal Rules Dataset, close to 80 implement either one or both of these rules. Expenditure rules have grown more common since the 2008-09 global financial crisis, as a complementary way to constrain discretionary spending and ensure better fiscal discipline. As a result, in most countries debt rules have predated the implementation of expenditure rules. Out of the 38 countries currently implementing an expenditure rule and a debt rule, only 4 countries have implemented the expenditure rule on its own before adopting a debt rule later.

Figure 1.
Figure 1.

Global Implementation of Fiscal Rules (1985-2015)

Citation: IMF Staff Country Reports 2017, 292; 10.5089/9781484321218.002.A001

Source: IMF fiscal rules dataset (2017). The dataset covers 96 countries.

The option to adopt an expenditure rule with a simultaneous debt rule is usually preferred. 84% of countries implementing an expenditure rule also implement a debt rule. Figure 2 shows 36% of all countries implement a debt rule, a budget balance rule and an expenditure rule at the same time.

Figure 2.
Figure 2.

Combination of Fiscal Rules in 2015

Citation: IMF Staff Country Reports 2017, 292; 10.5089/9781484321218.002.A001

It is also true for countries implementing aggregate medium-term expenditure ceilings. Only 24% of countries which set medium-term expenditure ceilings at an aggregate level are doing so without also having a debt rule.

24. PFM and institutional reforms are needed in support of the new expenditure rule to ensure that it reaches its objectives. Besides the need to improve budget flexibility to free up more fiscal space for discretionary policies (See Chapter III), the effective implementation of the rule will rely on the fulfillment of a series of preconditions: setting up strong medium-term strategic planning and allocation processes (Chapter IV); integrating the new rule with the budget process (Chapter V); and ensuring compliance with the new rule in a transparent manner (Chapter VI).

III. Improving Budget Flexibility - Reducing Rigidities

A. Sources of Rigidity

25. There are three major sources of rigidities in government spending: earmarking of revenues, mandatory spending, and constitutional provisions for indexation. Rigid expenditures amounted to around 78 percent of the budget in 2016 and their growth rate is determined by legally mandated indexation. As a result, the room for new policy initiatives is limited. The combination of growing obligatory expenditures and recession-constrained revenues prompted a sizable decline in public investment both in nominal and real terms during the last two years.4 Therefore, reducing budget inflexibility is critical not only to conform with the rule but also to regain space for priority spending, including investment.

26. Almost 80 percent of the central government revenues are earmarked. Article 212 of the Federal Constitution establishes a mandatory education spending floor equivalent to 18 percent of net current revenue for the federal government. In addition, the states, the federal district and the municipalities must assign no less than 25 percent of their revenues including federal transfers to education. In terms of the minimum expenditure on health, the Constitution mandates an increasing percentage of net current revenues to finance health appropriations. This percentage raises from 13.2 percent in 2016 to 15 percent in 2017. As discussed in Chapter II, the basis for calculating the floor has changed with the expenditure rule, however there remains a floor.

27. Social security expenditures are showing a significant expansion and, as a share of total expenditures, have increased from 35.5 percent in 1998 to 41.3 percent in 2016. These outlays are expected to grow as a share of total central government expenditures by 3.9 percentage points per annum and reach a level of 63 percent of the total budget in 2023.5 Two elements are driving the increase in Social Security expenditures. First, the number of beneficiaries is growing at a 3-3.5 percent annually. Second, some 64 percent of the social security payments are linked to the minimum wage.6 Last year’s rate of inflation and two years lag real rate of economic growth are the variables used in Brazil to adjust the minimum wage. Increasing the minimum wage by the rate of inflation or a rate to maintain purchasing power is a constitutional provision, however the real GDP growth increase is in legislation.

28. The use of inflation as a reference point in negotiations and automatic progression have significantly increased public employees’ compensation. In the 2013-2016 period, the growth rate of personnel and social charges was 2.2 times the growth rate of total revenues. The government and the public sector trade unions agreed to an annual salary increase of 5.5 percent for the 2017-2019 period. In addition, automatic personnel progression drives a 3 percent increase in aggregate salaries annually. Therefore, an 8.5 percent annual increase in personnel expenditures is envisaged for the next two years.

29. As a result of the high and growing budget rigidity, it is increasingly difficult to comply with the fiscal rules without endangering public investment. To meet its primary balance target, the government engages in frequent in-year expenditure adjustments. Given the size of mandatory spending there is limited room to maneuver and the focus is on discretionary spending mainly public investment adjustment. Planned public investment, as a share of planned total PLOA expenditures, has halved between 2013 and 2017 (Figure 2), with the effect even more pronounced during budget execution.

Figure 2.
Figure 2.

Compulsory and Discretionary Expenditure and Investment

Citation: IMF Staff Country Reports 2017, 292; 10.5089/9781484321218.002.A001

Source: Resultado Fiscal do Governo Central.

30. The government has used one mechanism to increase flexibility: the Release of Union’s Revenues (Desvinculação de Receitas da União) known as DRU. In 1994 an initial budgetary reaction against the problem of excessive earmarking was implemented. To help the Plan Real to stabilize the economy, a Social Emergency Fund (Fundo Social de Emergência) was launched and funded by the financial release of 20 percent of the earmarked revenues. In 2000 the name of the fund changed to DRU. Most of DRU’s receipts stem from Social Security Contributions and are discretionarily assigned as financing sources to expenditures decided by the MOF.

31. The government needs more options for reducing rigidities, given that the demographic and structural drivers of key programs will require declining real expenditure trajectories. When the new expenditure rule is considered against the extensive mandatory and legislative formulae for spending programs, even near term compliance with the new expenditure rule requires creating fiscal space through a range of available policy measures.

B. Improving Spending Efficiency

32. The motivation for creating additional fiscal space goes well beyond the need to comply with the new expenditure rule. Some high priority areas—in particular, public investment with its generally positive spillovers—have been curtailed due to the growth of mandatory or legislated spending items. Improving budget flexibility through spending efficiency is about giving the government more space and discretion to fund contemporary national priorities.

33. Maintaining steady pressure on public sector spending growth would have wider benefits. Regular and comprehensive reviews of programs to ensure they are achieving their objectives is likely to achieve better value for money and quality of public service provision. This is even the case, if the savings from these reviews are reinvested, as they would likely be in the areas of health and education.

34. It is not the purpose of this report to examine the range or merits of major reform options but rather to offer some general suggestions on ancillary mechanisms, linked to the budget cycle. These may help achieve quick, more certain, and admittedly modest results than major structural reforms which are needed.

Review of mandatory spending and indexation practices

  • A gradual yet important approach to building flexibility could come from slowing the rate of growth in some mandatory spending or legislatively-indexed spending items, while remaining within the limits defined by the Constitution.

  • The key items where this may be relevant are the annual adjustments to the minimum wage and to public sector salaries. Using lower indexation factors could produce significant savings over time.

  • A central agency led process, to review other areas of mandatory and legislated spending, would help to identify other areas where similar gradual adjustments could be made (e.g., subsidies).

Spending reviews linked to the budget process

  • Spending reviews generally refer to processes established by governments to pursue explicit savings targets to be delivered by the relevant ministries or programs for the budget process.

  • For example, the United Kingdom initiated a series of rolling reviews (e.g., once every two to three years) which contributed to reducing the spending to GDP ratio by five percentage points in five years, with further reductions in proposed over the forward estimates period.

  • In the more general “business as usual” case, savings targets might be more modest and can be used to either build fiscal space or to allow headroom for new priorities.

  • Spending reviews involve periodically examining the baseline or recurrent spending of each ministry, to assess whether desired outcomes are being achieved.

  • Such reviews can be led by line ministries, who are requested to identify their preferred savings within a specific envelope (the incentives are best aligned if the level of ambition is modest and the ministry can reinvest in new priorities); or central-agency led, which is usually the model for larger or more thematic exercises (e.g., reducing the size of government).

Tax expenditures review, sunset clauses, one-in-two-out methodology

  • The extensive use of tax expenditures in many countries presents some other options for clawing back fiscal space. Also, as noted in Chapter II in some countries tax expenditures have been used as means to bypass the expenditure rules.

  • A broad-ranging review of tax expenditures (which is proposed and a committee is being established in the MOF to conduct this review) could consider the scope to gradually wind back or phase out those concessions which have no standing legal basis, or which may no longer serve their intended purpose.

  • Other general policy mechanisms can help contain the growth and reach of tax expenditures. These include a rule that all tax expenditures must be subject to regular review, or even a fixed term (sunset clause). In some cases, rules apply that no new tax expenditures can be introduced without removing one or more of the same or greater cost (offsets).

  • Some of the space created by such an exercise could be put towards facilitating more general tax reforms that improve the overall efficiency and competitiveness of the tax system.

Efficiency dividends

  • Some governments have opted to apply across-the-board constraints on the growth in the operating expenses of ministries, leaving the ministries to determine the best approach to finding savings.

  • One example of this mechanism would be to reduce growth in the available operating funding for ministries by a small amount each year (e.g., 0.5 percent below what would otherwise be the indexation factor for operational expenses).

  • The level of efficiency factor is essentially arbitrary, but might be informed by other factors such as reasonably expected annual productivity improvements; gradual reductions in headcount; or some degree of general wage restraint as discussed above.

  • These can be useful mechanisms to gradually shift behaviors and cultures towards continual improvement, but if overused or left in place for too long, they can have unintended consequences on smaller ministries or those with less flexible budgets.

35. A joint effort with SOF, active cooperation by the relevant line ministry, and the involvement of external experts would all help to facilitate the process for spending reviews, together with the authority of the President’s Office in establishing initial savings targets. The policy options and savings generated by these processes should be brought forward to ministers early in the budget process as this will be critical input in determining compliance with the expenditure rule and the space available for other priorities.

C. Recommendations

36. The following recommendations are elaborated on in the Action Plan in Annex 1.

Reducing rigidities and improving spending efficiency

  • The MOF should initiate a review of mandatory spending and indexation practices.

    • Reviews of mandatory spending and indexation practices in excess of limits defined by the Constitution should commence as a priority.

    • Additional changes to legislation and norms to disconnect spending and indexation could be implemented in parallel to the 2018 budget.

  • The SOF and MOF should establish and coordinate a suite of policy processes that can build budget flexibility over time including:

    • Spending reviews linked to the budget process with specific savings targets for ministries and/or large and fast-growing programs;

    • Reviewing all tax expenditures to consider the scope for winding back or phasing out certain expenditures, introducing sunset clauses and/or applying offset rules; and

    • Applying efficiency dividends to ministries’ operational expenses.

  • Medium-term forecasting of the public sector wage bill should be strengthened. In addition, regular benchmarking and comparisons of public and private sector wages should be introduced.

    • Terms of reference could be prepared for an independent report on public sector compensation. The recommendations could be proposed to the Cabinet for approval, with legislation to follow if required.

IV. Strengthening Medium-Term Fiscal and Budgetary Management

A. Medium-Term Frameworks

37. There is a strong case for fiscal frameworks which shift the fiscal debate to the medium term, and even more so in the presence of fiscal rules. Most government policies have implications for many years ahead. Restraining decision-making to the annual timeframe leads to incremental budget allocations that are guided by shortsightedness and do not look for an optimal use of public resources. A more strategic, forward-looking approach to decision making on fiscal policy priorities and objectives contributes to better budgeting. In the case of fiscal rules, a medium-term planning effort is required to ensure compliance with them. Medium-term frameworks are necessary to support the implementation of fiscal rules.

38. There are different types of medium-term expenditure frameworks, based on the level of detail in the multiyear planning of expenditure. Medium-term frameworks are sets of institutional arrangements in the budget process governing the requirement to present certain medium-term fiscal information at specific times, procedures for making multiyear projections and plans for revenue and expenditure, and obligations to establish numerical expenditure limits beyond the annual budget horizon. They usually cover three to five years, including the budget year.

  • A medium-term fiscal framework (MTFF) sets aggregate fiscal targets (revenue, expenditure, balance, debt) and broad expenditure ceilings consistent with medium-term resource availability.

  • A medium-term budget framework (MTBF) adds the strategic prioritization of expenditure by sector, ministry and/or program level to the MTFF. This process can be fully top-down (the central financial agency sets expenditure ceilings for spending agencies) or informed by bottom-up information (the spending agencies provide their own vision for the allocation of resources).

  • A medium-term performance framework adds a performance dimension to the MTBF. It links the budget with medium-term performance indicators, targets and results.

It is important to note that a medium-term framework is not just a technical instrument designed to forecast fiscal aggregates in the medium term, but is mainly an institutional framework for setting multiyear fiscal objectives and ensuring they are respected in budget formulation, approval and execution. Figure 3 shows how the medium-term framework is linked to the wider fiscal framework and the budget.

Figure 3.
Figure 3.

How Medium-Term Frameworks Fit Within the Broader Fiscal Framework

Citation: IMF Staff Country Reports 2017, 292; 10.5089/9781484321218.002.A001

Source: IMF Staff.

39. Brazil has developed several medium-term instruments over the years, but they are not sufficient or coordinated enough to constitute a medium-term framework.

  • The Multiyear Plan (PPA), published during the first year of a government, defines the main strategic targets and programs of the federal government for a four-year period. It includes a detailed macro-fiscal scenario and a broad estimate of the resources allocated to each activity or project during the four-year period. In practice, though, the PPA relies on optimistic assumptions regarding both resource availability and implementation capacity, and thus does not provide a policy anchor for the budget process.

  • The Program for the Acceleration of Growth (PAC) is a prioritization process for investment projects over the medium term. It has been successful in setting strategic investment priorities and allocating space for projects. However, in the last few years, due to the economic downturn, spending under the PAC has been cut to comply with the primary balance targets.

  • Annual budget preparation is informed by the Budget Guidelines Law (LDO) submitted every year to Congress. It includes economic assumptions, fiscal targets for the primary budget balance, debt levels, and revenue forecasts for the year of the budget and two years forward.

The PPA and the LDO contain elements of a MTFF, but the PPA does not rely on a realistic macro-fiscal scenario and does not guide the budget process, and the LDO does not provide much information on the projections and policy choices required to reach the fiscal targets. They do not constitute MTBFs as neither document presents expenditure estimates at the program or ministry level.

40. Building on existing initiatives, it is important for Brazil to develop a fully-fledged MTFF to support the implementation of the new rule. The new expenditure rule is intended to be a binding, multiyear constraint that will guide all aspects of budget planning, preparation and execution. A MTFF is not only multiyear projections, but also refers to the principles, objectives, rules, targets and practices (formal or otherwise) that instill discipline and transparency in fiscal policy and budget planning.

41. In Brazil, building a MTFF will entail strengthening macroeconomic forecasting and developing a MTFS to guide the budget process. This important initiative needs to commence rapidly. In addition, and simultaneously building on existing work take to develop a MTBF.

42. This chapter outlines the main recommended steps for Brazil. The following sections will discuss how Brazil should improve the reliability of its macro-fiscal forecasting processes (B); implement a fully-fledged MTFF with more credible medium-term forecasts and a MTFS that can frame the budget debate and guide the budget process (C), and take immediate steps towards the implementation of a MTBF (D).

B. Strengthening Macro-Fiscal Forecasting and Processes

43. Medium-term frameworks should be anchored by solid macro-fiscal projections for their implementation to be effective. Reliable forecasts should allow the government to consider whether fiscal targets set by MTFFs are realistic, to assess the available fiscal space, and to allocate resources over the medium term in a credible manner. This section presents a series of recommendations for Brazil to step up its short term and medium-term macro-fiscal forecasting processes.

Macroeconomic forecasting

44. The Secretary of Economic Policy (SPE), within the MOF, is responsible for the macroeconomic forecasts underlying all budget and fiscal documents. The SPE is in charge of maintaining the macro-econometric forecasting models, some of which have been used for the last fifteen years. They are mainly used for the current year and the budget year. Forward year forecasts rely on simpler extrapolation methods. Macroeconomic forecasts for the current year are updated every other month in order to underlie the new revenue projections and expenditure targets set in the bimonthly execution reports. Macroeconomic forecasts for the budget and forward years are finalized in April for inclusion in the PLDO, and subsequently revised for the PLOA.

45. The optimism bias that has appeared in macroeconomic projections in recent years calls for improvements in the forecasting model. Between 2011 and 2015, the average deviation between one-year-ahead real GDP growth forecasts and outturns has been systematically negative, reaching −3.9 percentage points on average over the period (−0.7 percentage point over 2001-2010). Though reflecting in part the unexpected steepness of the economic downturn, it also reflects an optimism bias relative to the market projections. There is a possibility this recent bias is due to outdated specifications in the model. More extensive analyses on how to explain forecasting errors in the recent years should be carried out to identify the potential sources for bias. The ex post reconciliation between forecasts and outturns should also be published, possibly in the Presidential message attached to the PLOA, in order to improve the credibility of the forecasts.

46. A wider selection of projected macroeconomic indicators, over a longer period, should be shared with the Executive and released to the public. Only a limited number of macroeconomic indicators are communicated to the departments in charge of fiscal and budget forecasting (Treasury, Federal Revenue Secretariat, SOF): mainly real GDP growth, GDP deflator, inflation, wages, payroll and consumption of a selection of products (e.g., cigarettes, cars). For instance, the demand or supply breakdown of the GDP growth forecasts is neither sent to other departments and agencies nor released to the public. Such forecasts (provided they are reliable), if they were released for the budget year as well as for the forward years, could be used to better explain the forecasts for the more aggregate indicators, to find new revenue forecasting methods, and to support better targeted fiscal policy decisions.

Macro-fiscal linkages

47. The assessment of macro-fiscal linkages is especially important in the context of consolidation. Fiscal consolidation measures do not all have the same macroeconomic impact. For instance, tax increases are usually considered to have a more negative impact on economic growth than expenditure cuts. However, all expenditure cuts are not equivalent: a reduction in public investment may have a cascading effect on growth and productivity, while the macroeconomic effect of the suppression of an inefficient program may be non-existent. The macroeconomic effect of new fiscal policies, already decided or planned, should be studied carefully at the time of the decision. An ill-advised expenditure cut that impacts GDP negatively may lead to fewer tax revenues and, consequently, to the need for extra measures to comply with the primary balance target.

48. Macro-fiscal linkages are currently not taken into account in the macroeconomic forecasts in Brazil. Except for a few ad hoc studies on the macroeconomic impact of ideas for new measures by the SPE, there is no structured mechanism to ensure that the economic effect of new fiscal policies is reflected in the macroeconomic indicators produced by the SPE, and consequently in the revenue and expenditure forecasts.

49. The SPE should progressively build on the existing macroeconomic model to include these linkages. Steps would include systematic analysis of the macroeconomic impact of new fiscal policy measures, based on macro-econometric simulations and the estimation of the main elasticities of macroeconomic indicators to a change in the fiscal environment. The SPE should also be invited to take part in the proposed Technical Committee (see Chapter 4C), so that the macroeconomic impact of the fiscal strategy is systematically part of the debate.

50. Ultimately, an iterative forecasting process between the macroeconomic and the fiscal sides should be set up to fully integrate macro-fiscal linkages in the forecasts. Currently, when the fiscal forecasters receive the macroeconomic projections, they only have a few days to take them into account. A longer process, ideally over a few weeks, with back-and-forth between macroeconomic forecasts produced by the SPE and fiscal/budget forecasts produced by the MOF and SOF, would ensure that the interactions between the real sector and the fiscal sector are anticipated and taken into account.7

Revenue forecasting

51. Revenue forecasting is mainly the task of the Federal Revenue Secretariat (RFB), the agency in charge of tax policy and revenue collection. The RFB forecasts all tax and customs revenue items they are in charge of collecting (i.e., close to two thirds of total nonfinancial federal revenue).8 For each type of tax or customs revenue, the RFB applies the macroeconomic indicator which it considers the most appropriate to the base of the revenue item. The impact of new revenue policy measures is then added to the totals. Depending on the measure, the RFB can execute simple extrapolations or use microsimulations based on individual tax information. The RFB is also in charge of assessing the impact of tax expenditures. The main forecasting exercises are achieved ahead of the PLDO and the PLOA. Forecasts for the current year are revised ahead of the bimonthly execution reports based on revenue collection observed in the first months of the year.

52. Improvements to the RFB’s revenue forecasting methods could help reinforce the credibility of the annual forecasts. Revenue forecasting errors have been significant in recent years, in particular due to the economic downturn.9 To build on its current forecasting methodologies:

  • The RFB should identify the macroeconomic indicators that it would like to use as first-best assumptions for its revenue forecasts and communicate them to the SPE. These first-best indicators should be looked for by studying historical revenue collection data. In some cases, the SPE may be able to cater to the requests, while in other cases, it might not be possible, because the indicator is either not covered by the SPE’s model, not forecast, or its forecast is unreliable. Sources other than the SPE could be used for certain indicators, as long as this is done in a transparent manner.

  • The RFB should assess the impact of major tax and customs measures decided in the past ten years in order to distinguish them from the natural evolution of taxes, but also to review the costs and yields of past policies and reflect on possible policy changes in the future.

  • The RFB should conduct detailed analyses of the forecast errors, in order to distinguish the error due to the macroeconomic indicator used in the forecast (which is the responsibility of the SPE) from all others including wrong estimation of the impact of a new measure and variations in revenue collection, etc. All revisions to the forecasts presented in the bimonthly execution reports should be explained to the public.

53. Other stakeholders in the fiscal and budget process should be more involved in the revenue forecasting process, so that the budget benefits from different sources of expertise. While the RFB’s forecasts benefit from detailed micro-level knowledge on the revenue it collects, the Treasury’s expertise would be suited to a more macro-oriented, econometric look at the forecasts. The SOF could also have an expert say on the revenues that feed into the budget and on distribution and earmarking mechanisms. Discussing the forecasts and their underlying assumptions in an annual meeting ahead of the PLOA would help structure the exchanges of information between the different agencies and shape a storyline for the budget. The meeting would result in decisions made by the MOF on the forecasts that will ultimately be proposed in the budget bill.

C. Implementing a Medium-Term Fiscal Framework: Strengthening Medium-Term Forecasts and Developing a Fiscal Strategy

54. The new multiyear expenditure rule requires developing a MTFF, improving medium-term forecasts and developing a fiscal strategy to guide the budget process. The new rule is intended to be a binding, multiyear constraint that will ultimately guide all aspects of budget planning, preparation and execution. As such, “upstream” aspects of the MTFF would benefit from further developed and adopting a more medium-term orientation including the overarching fiscal objectives, the fiscal strategy, fiscal stance, and associated government policy priorities. These initiatives also require improved coordination across government.

Medium-term fiscal forecasts

55. Tools that are needed to support strategic decision making include fiscal aggregate projection methodologies. Fiscal aggregate projections—medium- to long-term estimates of revenue, expenditure, budget balance and debt—can help to illuminate the fiscal implications of existing policy settings and the wider set of risks and pressures to fiscal sustainability. Such top-down analytical tools can help decision-makers gain a better appreciation of the longer-term fiscal effects of underlying structural or demographic trends.

56. Medium-term fiscal forecasts should remain mainly under the responsibility of the Treasury. The larger the time horizon, the higher the macro-fiscal uncertainties. The Treasury has the expertise and skills to achieve such medium-term projections. In particular, medium-term revenue forecasts, being less reliant on revenue collection and much more on the macroeconomic projections, could primarily fall under the responsibility of the Treasury rather than of the RFB (though input on measures or revenue collection trends would be useful).

57. The Treasury should build a medium-term projection tool aimed at a swift assessment of the available fiscal space. Such a tool, used and maintained by the Strategic Fiscal Planning Unit, would be of use for fiscal policy decision-making. It would allow discussion of different macro-fiscal scenarios ahead of taking decisions. The Treasury is currently stepping up its medium-term fiscal projections. In particular, on the revenue side, the Department of strategic fiscal planning is currently working on a simplified revenue forecasting model based on econometric evaluations of tax elasticities to GDP. A similar work could be achieved on the expenditure side for the most rigid items. However, this tool would not bar the need for (i) more detailed macro-fiscal analyses by other departments or units, using a range of other tools, such as macro and micro-simulations; and (ii) a full-blooded assessment of debt sustainability and fiscal space.10

Medium-term fiscal objectives

58. Even Brazil’s traditional fiscal objective, the primary balance target, could be buttressed by stronger medium-term commitment mechanisms. Commitment mechanisms refer to the extent to which the government of the day is actually bound by its fiscal targets. For example, while the within-year commitment to the primary balance target appears strong (absent significant external shocks), the target is set only for the budget year with its evolution in subsequent years being indicative. As such, it does not provide an unequivocal statement of the desired pace of fiscal consolidation or the resulting trajectory of debt. Clearer forward guidance in either or both areas is desirable.

59. A medium-term debt anchor would complement the expenditure rule. A debt anchor would help guide fiscal policy and the expenditure rule. It would make the pathway to debt stabilization clearer and reinforce fiscal sustainability efforts. The FRL allows for the establishment of a debt target for the federal government, however in practice it has only been established for sub-nationals. A further extension of the framework would involve a requirement for the government to identify—either in abstract or detail—the policy measures it has undertaken to ensure fiscal outcomes remain in line with its strategy.

Medium-term fiscal strategy statement

60. The regular publication of a medium-term FSS at the start of the annual budget cycle would help bring the elements of Brazil’s MTFF together in a more integrated way. One of the key benefits would be a stronger strategic alignment between fiscal policy and the budget process, with important decisions taken up front to guide the more detailed process. The FSS could initially be presented as a statement to Congress by the Finance Minister that clearly conveys the government’s main fiscal objectives. Including a set of more detailed (indicative) multiyear fiscal targets, projections and economic assumptions; provides a transparent and detailed assessment of recent economic and revenue forecasts and outturns; and considers the main fiscal risks and alternative scenarios.

61. There are various choices surrounding what is included in a FSS and its substantive form. The mission’s proposal is that the FSS be initiated as a non-legislative instrument accompanying the PLDO. It should be replicated, or updated as necessary, ahead of the introduction of the annual budget law. In this form, it could become a major plank of the budget documentation, without being itself legislative in character or requiring Congressional approval or amendment. An account of the experiences of Australia and New Zealand is included in Box 2, and the proposed scope of a FSS for Brazil is detailed in Annex 2.

Fiscal Strategies: The Cases of Australia and New Zealand

The Australian Government’s Fiscal Strategy is published in Budget Paper No.1 each May. The Fiscal Strategy and Outlook (Statement No.3) documents the strategy, reports on each of the main fiscal aggregates (actuals vs forecasts and projections for up to 10 years), and reconciles changes in fiscal aggregates since the previous budget. Other statements in Budget Paper No.1 provide a detailed account of the annual budget’s goals and objectives; survey the macroeconomic outlook, forecasts and assumptions; present detailed estimates of revenue, expenses and investment; and assess forecast performance, alternative fiscal scenarios and risks.

The New Zealand Fiscal Policy Report (FPR) is published in May and the Budget Policy Statement (BPS) in December. The FPR announces the government´s goals, communicates the fiscal parameters, explains the economic context and the fiscal priorities, presents past fiscal performance, and provides fiscal forecasts and projections. The BPS announces the government’s plans and the budget goals and priorities. It also includes fiscal forecasts and projections. Both reports share an analysis of the short-term fiscal intentions and the long run fiscal objectives. The short-term fiscal intentions included in the FPR are compared with the ones addressed by the previous BPS. In the same way, the BPS analyzes the short-term fiscal intentions and provides a comparison with the ones included in the previous FPR. This communication strategy allows for continuity and underlines the government’s commitment to sound fiscal policies.

Source: IMF Staff.

Improving coordination and governance

62. Additional aspects of the MTFF relating to governance and reporting could also be reconsidered. In particular, a clearer role for responsible ministers in framing the fiscal strategy, determining the budget envelope and setting priorities could be contemplated and supported by the already strong technical capabilities of their respective ministries. Strategic coordination of macroeconomic, fiscal and budget issues could be overseen by a Ministerial-level Fiscal Committee involving the Finance and Planning and Budget ministers and the President’s Chief of Staff (see Figure 3). The experience of Malaysia’s Fiscal Policy Committee is recounted in Box 3.

63. A standing technical committee of officials, to support the Fiscal Committee, would allow better coordination and help enhance the smooth running of the budget process. This should be led by the strategic fiscal planning function, involve representation from the macroeconomic and budget functions, and draw upon other areas, such as revenue and the debt office, as required. The impartiality of technical inputs—particularly as they may relate to macroeconomic assumptions or fiscal forecasts—is critical and needs to be carefully protected. Regular reporting on fiscal outturns against initial forecasts would provide positive reinforcement.

Malaysia’s Fiscal Policy Committee

Malaysia’s Fiscal Policy Committee (FPC) was established in 2013 as the premier committee for fiscal management in the country. It aims to strengthen public finances, and ensure fiscal sustainability and long-term macroeconomic stability. It is chaired by Malaysia’s Prime Minister.

The FPC’s mandate includes:

  • endorsing guiding principles for sound fiscal policy and fiscal strategy formulation;

  • approving Malaysia’s MTFS;

  • reviewing fiscal performance and tracking towards fiscal targets;

  • endorsing policies to meet fiscal targets and manage fiscal risks; and

  • ensuring alignment across ministries and agencies on fiscal discipline and compliance.

The FPC is supported by a secretariat in the Fiscal and Economic Division of the Treasury which coordinates development of the MTFF and MTFS; production of the macro-fiscal forecasts, parameters and projections; liaison with other government agencies; and periodic studies on long term fiscal trends and drivers of expenditure and revenue.

Source IMF Staff.

64. Brazil should seek to formalize additional aspects of its MTFF over time. The requirement to produce, update and report against the FSS could, in time, become a formal component of the FRL. Related to this, further consideration could be given to the range of additional formal targets or informal supplementary indicators against which progress could be measured.

D. Introducing a Medium-Term Budget Framework: Aligning Priorities Within a Multiyear Expenditure Ceiling

65. Taking steps to develop a medium-term approach to budgeting would complement the MTFF and support strategic decision making in light of the new expenditure rule. Whereas, MTFFs are a set of top-down tools that help guide fiscal policy and budget planning, MTBFs are focused on developing the processes and information required to ensure that annual budgets remain aligned with the intended fiscal trajectory over coming years. MTBFs relate to the production of multiyear financial information (forward estimates) at specific times in the budget cycle, which are formally reviewed on a rolling basis at each budget update.

66. MTBFs can help to transform the nature of the budget decision making process. Key in this regard are the introduction of several strategic decision points in the budget process (ensuring fiscal space and robust prioritization) and the setting of expenditure ceilings early in the process. This set of issues is discussed further below and in Chapter V.

67. MTBFs guide bottom-up preparation of forecasts of expenditure, revenue and investment. They can support a stronger focus on expenditure reallocation and efficiency with a multiyear orientation. The bottom-up, multiyear forecasts can be consolidated to prepare better quality estimates of key fiscal variables which can be reconciled to previous vintages, as well as improve the accuracy of top-down fiscal aggregate modelling.

68. A particular form of MTBF is initially proposed for Brazil which would focus on detailed expenditure forward estimates to guide budget planning and preparation. Focusing, for now, on the expenditure side of the budget appears reasonable given the uncertainties currently affecting the revenue side of the budget. The framework can be extended in various directions over time as the various pre-requisites fall into place (discussed below). Several prerequisites are being addressed in the short term and the new expenditure rule will also help to do this.


69. Experience in other countries has shown that the prospects for successfully implementing MTBFs are improved if a number of preconditions can be met that help to deliver greater budget stability and predictability over time.11 While making progress towards full achievement of the preconditions is important (see Box 4), Brazil would nonetheless benefit from rapidly developing more detailed expenditure estimates with a medium-term horizon. The expenditure rule will help address many of the prerequisites, nevertheless efforts will be needed to improve forecasting and to reduce RAPs in order to enhance budget credibility.

Preconditions for Successfully Introducing Medium-Term Budget Frameworks: A Brief Assessment for Brazil

Pre-conditions for successfully introducing a MTBF include a credible and predictable annual budget process, established fiscal objectives and rules, accurate medium-term forecasts and a comprehensive and top down budget process. The text below assesses these pre-conditions in relation to Brazil.

Credible and predictable annual budget process. There has been a tradition of Congress inflating revenue forecasts to provide the “headroom” for additional expenditures. Thus, adherence to the primary balance target, was managed by ever tighter controls over budget execution, producing an enduring gap between expenditure estimates and outcomes. This practice should be curbed by the new expenditure rule, and the return of economic stability which will drive more stable fiscal outcomes.

Established fiscal objectives and rules. The annual primary balance (with an indicative trajectory) has been a feature of the fiscal framework for some time, and the new MTER has been added. There is however no overarching MTFS. The interaction between the primary balance target, the expenditure ceiling and various budget rigidities could still present challenges and unintended consequences.

Accurate medium-term forecasts. The recent period of economic volatility has made economic and fiscal forecasting more difficult than usual. The Treasury is however, making efforts to improve its forecasting and engaging in more comparisons with private sectors forecasts. That said, the development of a forward estimates and policy costings framework focusing on expenditure should be feasible, despite the ongoing uncertainty around revenue and what this can mean for within-year budget execution.

Comprehensive and top-down budget process. Given significant rigidities, the budget allocation process can be characterized as more incremental than top down or priority driven. Implementing the new rule in the manner discussed in this report is expected to help produce a more top down, unified, and medium-term perspective take root.

Source: IMF Staff.

Design choices

70. Linked to the introduction of the new expenditure rule, authorities may consider what form the MTBF should take in practice. There is no single ideal model. Countries have adopted different approaches (see Figure 4) and there is instead a set of choices around the following dimensions:12

  • Coverage – for example, the coverage would need to align with that of the new expenditure rule, but may or may not exclude other items. Future versions of the framework may be even more comprehensive and linked to the primary balance.

  • Specificity – what level of disaggregation makes sense, in terms of producing estimates / applying limits by sector, ministry, program, or objective? Can more detail be built up over time?

  • Time horizon – how many years should the forward estimates cover? This might be informed by the length of the electoral and/or planning cycles (e.g., a rolling four-year framework).

  • Discipline – are expenditure limits binding or indicative, and how often are they updated? What reserve margin or contingency is built into the framework to manage uncertainty?

  • Other – are estimates nominal or real; what are the links to planning process; how is prioritization managed; what controls apply to multiyear commitments and carryovers; how are reporting and reconciliations handled?

Figure 4.
Figure 4.

Overview of Proposed Coordination Mechanisms and Responsibilities

Citation: IMF Staff Country Reports 2017, 292; 10.5089/9781484321218.002.A001

Figure 5.
Figure 5.

Selected Design Elements in Some Advanced Economy Medium-Term Budget Frameworks

Citation: IMF Staff Country Reports 2017, 292; 10.5089/9781484321218.002.A001

Source: IMF Staff.

71. MTBFs can be designed in various ways and the choices that Brazil faces will be informed by other aspects of its fiscal and budgetary frameworks. The expenditure rule applies only to the central government. That said, the central government’s fiscal exposures to other levels of government is a critical element that should be captured. The proposed scope of coverage would simply align with the current practice (which excludes certain public corporations), although this has been the subject of other recent review processes.

72. Detailed forward estimates of expenditure would initially be indicative rather than binding, in keeping with the existing legal framework for relevant fiscal aggregates. Authorities have indicated an intention to produce real rather than nominal estimates. And the budget would remain structured around single-year appropriations, in line with the provisions governing the preparation of the annual budget in the Constitution.

Steps for implementation

73. The implementation of a MTBF should include the four key features:

  • A standardized process for forecasting expenditure estimates over the forward years.

  • Sufficient programmatic information to distinguish “existing” and “new” expenditure as the basis for better management of “fiscal space.” This process is designed to ensure that, as far as possible, all proposals for new spending are assessed relative to the priorities established in the strategy setting process. It recognizes that the existing expenditure base is an accumulation of spending decisions over many years, and needs to be monitored to investigate whether the activities being funded and the level of resourcing (especially staffing levels) are still required or could be used in different ways.

  • The setting of binding aggregate expenditure ceilings for the current year early in the budget process. This involves a distinct “strategy setting” stage to decide expenditure aggregates and broad policy priorities, to be presented in the Fiscal Strategy Statement. In essence, this means that budget priorities are set in a cascading manner: a total expenditure level is determined before the allocation between main sectors/departments is made, and sectoral/departments ceilings are set before the detailed division of expenditure within each sector is discussed and decided.

  • Integration of the forward estimates with the annual budget process so that at the beginning of the process the outer years are used as the starting position for the following year’s budget process and that during the process outer years are systematically updated for the impact of budget year spending decisions and parameter changes. This feature uses the mechanism of “rolling forward estimates” to integrate the MTBF and the annual budget process.

74. It appears desirable to rapidly develop a bottom-up, multiyear approach to expenditure forecasting. This would be a quick first step in a more gradual evolution towards a fully-fledged MTBF. It would extend and complement the work being done by MOF on top-down fiscal aggregate projections, while potentially improving the quality of information and processes used to make budgeting consistent with the new expenditure rule over the medium term.

75. The preparation of expenditure forward estimates will require strong coordination mechanisms across various parts of government. Preparation of the forward estimates could be driven within the Budget Office, given its well-established linkages to the planning process and relationships with the various line ministries. An overview of the forward estimates preparation process is included in Box 5, with a more detailed account in Annex 3.

Broad Steps in Preparing Forward Estimates of Expenditure Under a Medium-Term Budget Framework

Parameters – a single, consistent and complete set of economic parameters is prepared by the MOF. When the time horizon for medium-term estimates extends beyond a couple of years the parameters for the later years may be more in the nature of technical assumptions, consistent with trend economic conditions. It is important to ensure that demographic and related labor market effects (ageing, participation, urban/rural differences) are captured, so that accurate “target population” estimates can be prepared for relevant programs.

Baseline – entities produce multiyear estimates for the costs of all programs/objectives within their competency, utilizing the updated parameters and assumptions provided and assuming no change in government policy. The “no policy change” assumption requires well-structured guidance to ensure consistent approaches in cases where programs may formally cease but might be renewed, or a significant policy change is mid-way through execution. The Budget Office would need to carefully guide participating line ministries through the process.

New Policy Costings – line ministries would be responsible for producing multiyear costings of new policy proposals involving new spending or savings. The methodology applied to such costings needs to be consistent with relevant baseline models, or must be developed on a case-by-case basis for novel situations. It is an important discipline over the process that the Budget Office endorses the modelling approach, assumptions and results.

Consolidation – additional steps and judgments are required on the part of the Budget Office to transform the bottom-up estimates by objective, program, or ministry into accurate estimates of the fiscal aggregates. For example, this may involve the elimination of inter-agency expenditures to avoid double counting or the incorporation of public debt interest expenses into the estimates (if nominal expenses or the nominal balance is being estimated).

Reconciliation – preparation of detailed forward estimates at each budget update allows for a relatively detailed reconciliation to occur, explaining the sources of changes between updates. The impact of changing economic conditions on fiscal aggregates and even key programs can be isolated from the impact of new policy decisions. This breakdown can be included in budget documentation and provides a sense of the extent to which economic developments or discretionary government policy is affecting fiscal outcomes.

Source: IMF Staff.

76. The initial MTBF development process would benefit from involving line ministries with existing capabilities such as Social Welfare and Security, Labor and Employment, Education, Health, Defense and Transportation. The initial goal is to have a well-developed set of expenditure forward estimates to guide internal budget decision making by early 2018. As experience and confidence with the new MTBF grows, it should be possible to extend participation to all ministries.

Future steps

77. Successfully embedding a new MTBF within Brazil’s broader fiscal framework will require ongoing progress across several complementary fronts. First, this requires ensuring that the budget process evolves to support the intended strategic and medium-term orientation, engaging responsible ministers at key points to ensure fiscal space is created and that prioritization occurs. This relates to initiating and integrating processes and reforms that build fiscal space and allow some policy flexibility during each budget cycle, as discussed in Chapter III.B.

78. Budget credibility and predictability will be enhanced over time as the expenditure rule promotes a convergence between the fiscal strategy, budget estimates and outturns. In order to bring in the more volatile revenue side of the budget and align the framework with the primary balance target, much will turn on continuing to improving macroeconomic and fiscal forecasts (as discussed in Chapter IV.B) and rounding-out the remaining aspects of the MTFF (as discussed in Chapter IV.C). Regular, transparent reporting of performance and future projections against the fiscal strategy and fiscal rules can also help the new MTBF take root.

79. In time, the MTBF could be extended to produce more detailed baselines and new policy costings for revenue and tax expenditures, and thereby support adoption of a firmer multiyear primary balance target. A reliable bottom-up projection of the primary balance makes a sharper focus on the medium-term debt trajectory more feasible. Once the process matures, the forward estimates and new policy costings could shift from an internal tool to support decision making to being published as part of the annual budget papers.

E. Recommendations

80. The following recommendations are elaborated on in the Action Plan in Annex 1.

Improving macro-fiscal forecasting

  • Assess the macro-fiscal linkages at play and set up an iterative macro-fiscal forecasting process in order to embed them into the forecasts:

    • Estimate the elasticities of key macroeconomic indicators to different types of fiscal events or scenarios;

    • Produce systematic analyses of the macroeconomic impact of new fiscal policy measures;

    • Release forecasts for the demand and supply breakdowns of GDP growth, both to the Executive and to the public;

    • Set up an iterative macro-fiscal forecasting process in order to embed macro-fiscal linkages into the forecasts.

  • Produce more analyses on in-year revisions and reconciliations of macro-fiscal aggregates in execution reports:

    • Explain macro-fiscal forecasting revisions in the bimonthly execution reports;

    • Carry out and publish analysis of the macroeconomic forecasting errors to identify sources for the recent optimistic bias.

  • Improve revenue forecasting methods and processes by assessing the cost or yield of previous tax measures and by setting up a forecast review process:

    • Assess the cost or yield of major tax and customs measures decided in the past ten years;

    • Set up annual meetings ahead of the PLDO and PLOA with the RFB, the Treasury and SOF to discuss revenue projections and decide on the forecasts that will be ultimately proposed in the PLOA.

Implementing a medium-term fiscal framework

  • Develop a fully-fledged MTFF by improving medium-term macro-fiscal forecasting and publishing a regular FSS to guide the budget process.

    • Build a medium-term projection tool aimed at a rapid assessment of the available fiscal space.

    • The FSS could become a major plank of the budget documentation without being legislative in form or subject to Congress approval.

    • The FSS could be released with the PLDO and replicated or updated to accompany the PLOA. Details on the proposed coverage of the FSS are contained in Annex 1.

  • Improve fiscal policy and budget coordination by creating a Ministerial Fiscal Committee (Minister of Finance, Minister of Planning and Budget and the President’s Chief of Staff).

    • This committee would agree primary balance and expenditure rule targets, the MTFS, budget priorities, and aggregate ceilings for ministries.

    • It should be supported by a Technical Committee representing the macroeconomic policy, fiscal planning and budget functions.

  • Adopt a credible debt path to guide fiscal policy and reinforce fiscal sustainability efforts. This has both short-term and longer-term dimensions:

    • Initially setting an internal debt target would help to guide fiscal policy and the expenditure rule.

    • Over time, it would make the pathway to debt stabilization clearer and consideration should be given to implementing it as a formal fiscal rule or permanent anchor.

Introducing a medium-term budget framework

  • Develop a MTBF to produce multiyear expenditure baselines for major sectors and guidelines on new policy costings:

    • The Technical Committee, with the Budget Office taking the lead, would need to develop the necessary frameworks and guidance, working with major spending ministries.

    • The initial goal would be to have the framework ready to employ ahead of the 2019 budget process. Coverage of the MTBF can be extended to all ministries and to revenue at a later stage.

  • As the MTBF matures, the forward estimates could be published in the budget papers and the updated baseline adopted at the start of each year would serve as the ceiling for the annual budget process.

V. Embedding the New Expenditure Rule in the Budget Process

A. Changes in Budget Procedures and Timelines

81. This section deals with some broad changes in budget procedures and timelines that would help to embed a more medium-term horizon for budget planning purposes. The changes will support earlier strategic decision making to implement the new expenditure rule. The proposed approach involves a deliberate shift in emphasis to engage the responsible ministers at critical points in the budget cycle. It is also intended to promote some changes in thinking and practices that underpin budget planning and preparation, and to bring a clearer focus to expenditure prioritization as line ministries’ budgets are produced.

82. The mission has not sought to provide a detailed review of the existing budget process and timeline in Brazil. Rather, it focuses on how the elements of its proposals could be combined and overlaid on the existing process (see Box 6). The general observations can be offered that the existing budget process is quite protracted (commencing in February with the fiscal year beginning the following January), with the key targets being locked in relatively early (April unless amended).

83. There are some additional tools which could be developed to support the engagement of ministers and the medium-term emphasis of the strategic decision making process discussed in Box 6. These primarily relate to the ability to distinguish between parameter and policy-driven changes in expenditure estimates and devices that allow real-time tracking against targets and the creation and use of available fiscal space.

84. Maintaining a multiyear reconciliation table is useful for strategic and technical purposes. This shows the movements in the primary balance and expenditure since the last budget, broken down into parameter variations and new policy. At an aggregate level this can be published in the budget papers to give a sense of the key drivers of changes over the forward estimates period. It is also helpful for the Technical Committee to maintain a more detailed version of this reconciliation table for internal purposes during budget preparation. For example, when a new expenditure baseline is prepared, it is useful to consider which ministries, programs or objectives have produced the largest variations due to changing economic parameters.

Indicative Timeline to Integrate Changes with the Budget Process

January – The Technical Committee oversees preparation of updated macroeconomic and fiscal forecasts. When the MTBF is operational, this will also involve ministries contributing detailed bottom-up estimates of the expenditure baseline. In the meantime, the top-down fiscal projections, or some blend of top-down and bottom up expenditure estimates, can be used.

February/March - Ministerial Fiscal Committee has its first strategic meeting for the cycle.

  • Considers previous outturn, latest fiscal forecasts and draft FSS.

  • Determines levels of fiscal aggregates to be used for budget planning purposes, consistent with the two fiscal targets (determines primary balance and expenditure targets; accepts technical advice on macroeconomic and revenue forecasts).

  • Approves allocation of pre-ceiling to ministries for PLDO preparation purposes and reserve a buffer for contingency purposes.

  • Discusses emerging policy priorities and sets savings targets for expenditure reviews and other related mechanisms.

April – FSS is released by the Minister of Finance when PLDO introduced. In 2017, it may need to be issued later, to accompany the PLOA.

April-June – Working meetings of the Ministerial Fiscal Committee.

  • Committee considers marginal allocations or reductions to accommodate new policy (spending and savings measures) and the results of spending reviews.

June – Technical Committee oversees preparation of updated forecasts and expenditure baselines.

July – Ministerial Fiscal Committee has its second strategic meeting.

  • Considers updated macroeconomic and fiscal forecasts (including inflation outturn for expenditure cap, and latest revenue forecasts).

  • Considers the aggregate impact of new policy measures, and any necessary adjustments or reallocations to be reflected in the final annual budget proposal.

  • Approves adjustments to aggregate and detailed expenditure ceilings if necessary.

August – The PLOA is introduced, accompanied by an updated FSS.

Source: IMF Staff.

85. The major estimates variations can be the focus of further scrutiny and possible remedial policy action later in the budget process. Updating this detailed reconciliation table after each meeting of the Ministerial Fiscal Committee (and to incorporate formal updates of the economic parameters or revenue forecasts) provides a rolling estimate of the primary balance and level of expenditure.

B. Impact on Legislative Amendments and Oversight

86. The new primary expenditure rule is a game-changer for the political economy of budgeting. In the past, the Congress systematically increased revenue projections included in the budget proposal to allow higher expenditures. Unrealistic budgets resulted from this political game and the absence of fully funded budget expenditures severely affected the cash management and paved the way for arrears and unpaid commitments. The new rule changes this since the aggregate spending to be voted by the Congress will be restricted by the rule´s ceiling. In this regard, the rule effectively addresses the common pool problem of budgetary design and approval. International experience reveals that fiscal discipline weaknesses can largely be explained by institutional deficiencies in the budget process that make the common source resource problem become more intense.13

87. The rule also introduces other changes which impact congressional operations. Since 2015, parliamentary amendments made to the budget bill are capped at 1.2 percent of the net current revenue, half of which should be assigned to the health sector. The new rule keeps this amendment reserve but from 2018 on, it will be calculated as a share of the 2017 ceiling, adjusted for inflation. In addition, it puts restrictions on creating new expenditures by requiring that the proposals for new mandatory spending or tax expenditures be accompanied by an estimate of the budget and any financial impact. Also, bills which are not compatible with the rule can be suspended for 20 days.

88. The effective Congressional oversight of the rule will have to take into account the link between the Lei Orçamentária Anual (LOA) aggregate envelope and cash execution. Whereas the rule-driven spending feeds into the budget process providing maximum expenditure authorizations, the compliance of the rule will be assessed on a cash basis (pagamento). Therefore, to explain the rule’s observance in a transparent way, a table linking the rule´s primary expenditure payments and the LOA budget envelope should be presented. In relation to this, and as it is analyzed in the next sub-section, not all the LOA primary expenditures will be executed since the execution of the budget must be designed in a way that makes room for the RAP to be paid in year. Consequently, a methodology to explain the discrepancy between the approved and cash disbursed budgets is required.

C. Linking the Rule to Budget Execution and Cash Management

89. The EC 95 establishes that the expenditure rule ceiling must be consistent with the budget. The EC 95 determines that the PLOA should show the aggregate level of budgetary expenditures compatible with the rule’s ceiling. In addition, it establishes that the primary expenditures included in the budget must not be higher than the rule’s ceiling. Since the budget must be consistent with the rule ceiling, the EC 95 aims at preventing any incompatibility between the budget and the financial executions.

90. To integrate the budget process and the rule, the realism of the budget has to be strengthened. The budget process is fragmented by the coexistence of two parallel budgets. On the one hand, the budget approved by Congress. On the other hand, the execution of a sizable amount of committed but not paid expenditures dating from previous budgetary years, not included in the current approved budget, and executed in parallel to the approved budget.14 The stock of committed but unpaid expenditures amounted to some 2.4 percent of GDP at the beginning of 2017.

A key aspect of the committed but not paid expenditures (RAP) is that it competes with the approved budget in the use of the financial resources available. RAPs are a longstanding element in Brazilian financial management.15 The execution of the budget appropriations (dotação orçamentária) includes three separate instances, each one subject to an independent treatment in the budgetary accounts: the commitment stage (empenho), the accrual stage (liquidação), and the payment stage (pagamento). There are two types of RAP. On the one hand, committed but not accrued expenditures denominated unprocessed RAP. On the other hand, accrued but not paid expenditures, denominated processed RAP. Both types of RAP are different in nature. The processed RAP corresponds to expenditures whose goods and services have been received whereas the unprocessed ones are similar to carryovers in the sense that they allow the transfer of non-accrued commitments from one year to the next.16

91. Although the unprocessed RAP can be explained as an instrument that accommodates the annuity of the budget with the multi-annuity of many expenditures,17 the stock of RAP represents some 12 percent of annual expenditures imposing a challenge to aggregate fiscal management. If the size of the carryover becomes large, the potential conflict between the spending priorities of the government and the action pursed by the budget manager could quickly escalate.18 Given that the expenditures have already been committed, their execution imposes a significant fiscal pressure on the budget and to the cash availability since the RAP compete with the current budget at the accrual and payment stages.

92. The Presidential Decree that distributes the budget appropriations among line entities also distributes the RAP to be paid during the year. Therefore, the Decree merges the two separate budgets into a single execution timetable. Since there are not enough financial resources to match the payments related to the approved budget and the RAP, the excess of commitments over payments is a permanent aspect of the budgetary process as the Table 4 reveals. It is relevant to note that the Budget Organic Laws of most Latin American countries includes a clause establishing that the amount of non-accrued commitments at the end of the year will be assigned to next year appropriations, an alternative that the Brazilian framework does not authorize.

Table 4.

Restos a Pagar

R$ billions

article image
Source: Tesouro Nacional, Avaliacao dos Restos a Pagar 2017 and FAD estimates

93. In the last two years, fiscal measures have significantly reduced the stock of RAP. It is useful to analyze the dynamics of the stock of RAP by comparing the rate of RAP creation with the rate of RAP elimination. The former is the ratio between the new RAP generated in a year and the stock of RAP at the beginning of the same year. The latter is the ratio of the sum of RAP paid and RAP cancelled in a given year and the stock of RAP at the beginning of the same year.19 The stock of RAP will increase, decrease or be constant if the rate of RAP creation is greater, lower or equal than the rate of RAP elimination. Figure 6 shows the RAP creation and elimination for the period 2006-2016. The rate of unprocessed RAP creation has significantly lagged behind the rate of RAP elimination in 2015 and 2016. The same happened for processed RAP in 2016.

Figure 6.
Figure 6.

Creation and Elimination of Restos a Pagar

Citation: IMF Staff Country Reports 2017, 292; 10.5089/9781484321218.002.A001

Source: Tesouro Nacional, Avaliacao dos Restos a Pagar 2017.

94. Three factors have impacted the stock and flow of RAPs in the past two years. First, the increase in RAP cancelled during 2015, some of them because the Federal Court of Accounts (TCU) intervened in determining the cancelation. Second, an increase in RAP paid during 2014 - 2016. Third, a tighter commitments (empenho) control applied in 2015-2016 that determined a reduction in the flow of new RAP.20

95. The design and monitoring of the rule requires ensuring that spending stays within the target on the execution phase, but there are coordination challenges in terms of budget execution. The EC 95 establishes that the PLOA will make it clear that the rule’s ceiling is consistent with the expenditures and that the budgeted primary expenditures will observe the target. Therefore, the LOA will provide to the Treasury and SOF a neat and consistent aggregate level of expenditures to guide the execution of the budget. This is a welcome aspect of the reform. However, in terms of transparency, the sizable amount of RAP that rolls along the budget years presents a challenge. As soon as the budget year starts, the Presidential Decree merges the approved budget with the RAP programmed to be paid. The overall consolidation is still consistent with the rule’s target since the Decree is guided by the rule in determining the overall execution ceiling communicated to the line entities.

96. However, in order to do that, some of the LOA appropriations will need to be removed to compensate for the RAP payments programmed to be paid in the year. In other words, the same fiscal envelope will equal both the approved LOA aggregate expenditures and the sum of the expenditures to be executed and the RAP to be paid. This aspect will prevent the primary expenditures executed at the end of the year from being completely linked to the programs and activities approved by the LOA.

D. Changes in Spending Ministries

97. The EC 95 rule represents a challenge for line entities. Not all the entities are similarly exposed to the trends that drive expenditure costs. The most typical case is social security but, even considering the non-social security entities, the budget preparation stage will have to coordinate the top-down aggregate consistent with the rule with the differential growth of the budget base at the entities level. In this regard, both aggregate and ministerial aspects need to be considered. At the ministerial level, the programs must be re-prioritized to be consistent with the entities ceilings. At the aggregate level, the Ministerial Committee should further analyze how to equalize at the margin the social benefits of expenditure among entities under the budget constraint. In this regard, it is important to apply medium-term considerations as the economy is expected to gradually recover from the recession and with it government revenues.

98. The SOF has recently enacted the Circular (Portaria) 9/2017, which defines the new budget deadlines related to the preparation of the PLOA 2018. The circular instructs the line entities on the budgetary observance of the expenditure rule and flags the limited fiscal space available for the next years.

E. Recommendations

99. The following recommendations are elaborated on in the Action Plan in Annex 1.

Changes in budgetary procedures and timelines

  • Bring a more strategic approach to guide the budget process in two distinct stages: the first setting of ministerial expenditure ceilings earlier before the LDO is approved by Congress, and a second stage discussing adjustments or reallocations of any fiscal space to priority areas or allocation of reductions.

    • Tools will need to be developed to allow tracking of the fiscal position as decisions are made and to permit reconciliations of the main fiscal aggregates to be published in each budget.

Impact on legislative amendments and oversight

  • Provide regular information on the link between LOA’s aggregate appropriations and the rule’s paid expenditures

    • Develop a table, published in the bimonthly execution reports, linking the rule’s primary expenditure payments and the aggregate budget envelope approved by Congress

    • Develop a methodology reconciling the discrepancy between appropriations and payments, to be included in the annual execution report

Linking the rule to budget execution and cash management

  • Develop a plan to gradually reduce the stock of RAP to no more than 3 percent of the aggregate level of expenditures.

    • Develop a detailed financial plan, updated annually, to track towards this target. A focus could be on the investment projects pipeline to improve operational forecasting methods.

VI. Ensuring Compliance with the New Expenditure Rule

100. With any new fiscal rule, arises the question of how to ensure compliance with the rule, both ex ante and ex post. The constitutional amendment setting up the new fiscal regime contains limited provisions on monitoring and reporting obligations. Currently, the authorities plan to follow similar monitoring and reporting procedures as with the primary balance rule. However, the implementation of the new rule raises the following questions: (A) how to ensure transparent reporting of compliance with the rules; (B) how independent watchdogs should be involved in monitoring compliance; and (C) how to apply sanctions.

A. Transparent Fiscal Reporting

101. The public should have regular access to information on compliance with the rule. This is relevant: (i) ex ante, to explain how the fiscal strategy chosen by the government is coherent with the rule; (ii) in-year, to justify potential budget sequestrations; (iii) ex post, to make the government accountable for its fiscal policy and management choices. Information can be made available in fiscal reports, in budget documents (including the MTFS), in budget execution reports, but also in ad hoc communication devices decided by the Executive.

102. The ongoing plans by the Treasury to include new tables in existing reports are going in the right direction.

  • The Treasury intends to report past compliance with the new expenditure rule in its monthly fiscal bulletin, as well as in the four-monthly reports on compliance with the fiscal rules, in a similar format to what is published for the primary balance rule.

  • The Treasury will also use bimonthly execution reports to explain how proposed sequestrations on commitments and payments may allow for compliance with the rule.

103. In order to ensure transparency, an effort should be made to provide explanations of these new tables. Though the expenditure rule may look simple at first, specific explanations are needed to account for exclusions from the rule and differences with the scope of the primary balance target,21 and gaps between commitment and payment targets. Reports should, in particular, contain explanations on all revisions applied to the targets in relation with the rule, a technical note on the scopes of the different fiscal rules and aggregates, and a reconciliation effort between the various targets at play.

B. Independent Oversight: Independent Fiscal Institution and Federal Court of Accounts

104. The newly-established IFI has a wide mandate, relative to its size and resources. The IFI was established in December 2016 by a senatorial resolution. It comprises of six staff members, including the Executive Director. Its functions are fourfold: (i) produce its own macroeconomic forecasts as a basis for its fiscal scenario; (ii) assess compliance with fiscal targets; (iii) evaluate the fiscal impact of government policies, including monetary, credit and foreign exchange policies; and (iv) project the evolution of fiscal indicators that are relevant for long-term fiscal sustainability. These functions are typical of fiscal councils, as evidenced by international experience, but councils with comparable remits usually enjoy a larger number of staff (see Box 7). The IFI intends to publish monthly reports, containing macro-fiscal forecasts, assessment of compliance with the expenditure rule and ad hoc studies. The IFI will not provide normative analysis or recommendations.

The Remits and Resources of the Brazilian Independent Fiscal Institution Compared with International Experience

The Brazilian IFI enjoys a wide range of remits. As shown on Figure 1, the mandate of the IFI covers five of the six main functions usually performed by independent fiscal councils, including the fiscal costing of measures, which less than half of the fiscal councils cover. Only three fiscal councils in the dataset—the CPB Netherlands Bureau for Economic Policy Analysis (CPB) and the United Kingdom’s Office for Budget Responsibility (OBR) and the US Congressional Budget Office (CBO)—enjoy such a wide remit.

Figure 1.
Figure 1.

Remits of Independent Fiscal Institutions in the World

(in green, remits of the Brazilian IFI)

Citation: IMF Staff Country Reports 2017, 292; 10.5089/9781484321218.002.A001

The Brazilian IFI is however significantly understaffed for such a wide mandate, compared with other fiscal councils. Figure 2 shows that the wider the mandate, the more resources and staff in the fiscal councils. The policy costing is the most staff-intensive remit. For example, the CPB and the OBR respectively host 117 and 17 technical staff, compared with the Brazilian IFI’s 5.

Figure 2.
Figure 2.

Remit and Number of Technical Staff in Independent Fiscal Institutions

Citation: IMF Staff Country Reports 2017, 292; 10.5089/9781484321218.002.A001

105. Though the quality and breadth of its first two monthly reports is impressive, the IFI may run the risk of spreading its resources too thin. Considering the limited size of its staff, the IFI should strategically plan the time and resources devoted to each of its functions. Emphasis could be laid on the parts of its mandate which are key and may foster useful societal debate on fiscal policy.

  • The IFI is the only independent institution in Brazil whose mandate explicitly includes the assessment of compliance with fiscal rules. These assessments will be made on a very regular basis; it plans to have these around the time of the bimonthly execution reports. It should also be ahead of the PLDO and PLOA.

  • The IFI is currently the only institution which publishes long-term debt projections. This aspect should be further developed.22

  • In terms of studies, the IFI should focus its effort on the major fiscal reforms under discussion, in order to contribute as much as possible to the fiscal policy debate.

106. A cooperation process between the Executive and the IFI could be fostered. While preserving the IFI’s independence, the Executive should work with the IFI and ensure regular and timely exchanges of information. In return, the IFI should also offer a platform for the Executive to discuss its own forecasts when there are divergences.

107. The Federal Court of Accounts (TCU) also has the mandate to assess compliance with the expenditure rule and plans to do so. Article 71 of the Federal Constitution defines the mandate of the federal external audit institution, which includes the annual examination of the final accounts. On its own initiative or when requested by Parliament, it may conduct any inspection or audit “of an accounting, financial, budgetary, operational and patrimonial nature” in the administrative units of the three Branches. This is ground for the TCU to be involved in monitoring compliance with the expenditure rule. The TCU already provides an ex post analysis of compliance with the primary balance targets set in the LDO in its annual report. A similar analysis could be produced for the expenditure rule in the annual report. The TCU also intends to monitor the compliance with the rule on a bimonthly basis, which may duplicate the analysis of the IFI.

C. Applying Sanctions

108. Though the constitutional amendment details a series of automatic adjustments to be applied in case of non-compliance, it remains vague on how to trigger or end sanctions. The amendment explains that the adjustments are triggered by non-compliance of any of the branches of government with the rule. However, it is unclear who should trigger the mechanism, when it should be triggered and if the non-compliant branch should be allowed to provide explanations on its deviation from the targets. Similarly, the amendment does not specify when the adjustments should stop being applied and who should decide on this.

  • Who? In practice, the Treasury seems to be well-placed to do so, since it will regularly assess compliance with the rule in the fiscal reports. This should however be complemented with an independent opinion (i.e., IFI, TCU) to ensure transparency.

  • When? Possible options include to apply and end the adjustments as soon as there is reason enough to do so; another option is to wait for the LOA to apply and end the adjustments, on an annual basis, to avoid creating a gap between the budget and the actual payments.

  • How? The bimonthly execution reports should contain a section on the triggering and ending of the automatic adjustments.

109. There is little or no information on how to monitor the implementation of the automatic adjustments. Non-compliant branches should provide explanations on their strategies to go back to a compliant fiscal trajectory, in particular as part of the overall MTFS. Bimonthly execution reports should also monitor the implementation of this strategy.

110. The rule includes escape clauses for natural disasters, war, and civil unrest, however there are no details on the procedures for initiating and ending these escape clauses. International experience highlights the importance of having strict definitions of escape clauses, and defined procedures for initiating them and including plans on how the government plans to return to compliance with the rule.

D. Recommendations

111. The following recommendations are elaborated on in the Action Plan in Annex 1.

Improving reporting

  • Ensure transparent fiscal reporting on compliance by the National Treasury:

    • Include new tables and written explanations analyzing compliance with the expenditure rule in the Treasury’s monthly bulletins and four monthly reports on fiscal rules;

    • Explain revisions applied to commitment and payment targets in relation with the rule in each bimonthly execution report;

    • Attach to the four monthly report on compliance with fiscal rules a technical note about the scopes of the different fiscal rules and expenditure aggregates, and explaining how the reconciliation can be achieved.

  • Provide an annual performance assessment (tables and written explanations) against the rules in the MTFS and in the citizens’ budget.

Monitoring compliance

  • Focus IFI on the assessment of compliance with the fiscal rules, studies on major fiscal reforms and long-term sustainability analyses;

Applying sanctions and escape clauses

  • Consider developing procedures to trigger, implement and end automatic adjustments in case of non-compliance and to activate and end escape clauses.

Annex 1. Action Plan

article image
article image
article image
article image

Annex 2. Outline for a Fiscal Strategy Statement


This annex outlines the suggested elements of a FSS which could be prepared in Brazil to round out the MTFF and bring a more strategic, medium-term focus to budget planning and preparation.

It is intended to be neither prescriptive nor exhaustive, but rather to assist authorities to think through how the elements of such a Statement could come together in the Brazilian context.

The primary purpose of the proposed FSS is to serve as a strategic communication tool which helps to place the fiscal policy debate on a medium-term footing, but also as something to which the government is willing to be held directly accountable.

Fiscal policy and related targets and objectives

  • Specify the core fiscal targets – place the primary balance and expenditure targets in context.

  • Nominate any supplementary fiscal targets – these may or may not be legally binding, but are still important in framing the fiscal strategy (such as floors for education and health, as well as indicative guidance on desired allocations towards public investment).

  • Present the indicative primary balance target over a longer-term horizon - this communicates the intended pace of fiscal consolidation, absent economic shocks. Consider how this might be translated into a firmer target (e.g., rolling average, structural balance).

  • Present the indicative target levels of revenue and expenditure as a share of GDP – this communicates the broad policy directions regarding the overall size of government (declining as a share of the economy under the new rule) and the overall tax burden.

  • Show projections for the profile for growth in mandatory, non-discretionary and other expenditure which indicates the discretionary fiscal space for national priorities (e.g., health, education, public investment).

  • Show projections for revenue with estimates of relevant tax expenditures.

  • Assess the aggregate value of reforms required to deliver on the fiscal rules - this could be done both over the coming budget year and over the projection period and include some discussion of major policy themes which will be pursued.

  • Present a profile of gross and net debt – this may include an indication of the government’s desirable / tolerable levels of debt or annual borrowing requirements as well as help to illustrate the links to the nominal balance trajectory.

  • Estimate the size of primary surplus that would be targeted to help achieve fiscal objectives, such as stabilizing and paying down debt and as insurance against future economic shocks.

Macro-fiscal analysis

  • Analyze recent domestic and international economic developments and illustrate how changes in the economy have impacted upon the fiscal situation.

  • Track progress against the core and key supplementary medium-term fiscal targets.

  • Present a transparent assessment of recent economic and revenue forecasts vs outcomes.

Projections and risks

  • Include projections of main fiscal aggregates for 5 and ultimately 10 years (updated each year and compared to previous vintage).

    article image
    Primary and nominal balance

    article image
    Gross and net debt

    article image
    Public sector net borrowing

    article image
    Primary surplus required to stabilize and reduce gross/net debt

    article image
    Mandatory expenditures as a share of total

Additional elements which could be added over time

  • Formal debt anchor – implement the legislated targets for gross and net debt.

  • A more detailed list of the government’s preferred policy measures which will be pursued in the 1-3 years ahead to preserve discretionary fiscal space, help repair the budget and give more flexibility in prioritization and budget execution.

  • Alternative fiscal scenarios which show the impact of varying economic assumptions on the likely fiscal outcomes (e.g., a terms of trade shock; or a fall in global growth).

Annex 3. Basic Methodology for Preparing Forward Estimates

Producing expenditure forecasts or forward estimates at the disaggregated level involves a number of steps. These are focused on understanding the existing budget; understanding and applying the medium-term cost drivers; and aggregating the forward estimates and summarizing the overall sources of variations.

Each of these steps need to be identified, analyzed and ideally have the methodology agreed by the Technical Committee, as the forward estimates form the basis of future budget allocations. Taking the time to agree on the methodology and inputs at the beginning will simplify future budget negotiations, as the agreed baseline can be approved quickly, leaving more time to focus on the higher value activity of assessing and costing new policy initiatives and proposals, rather than negotiating over existing budget allocations.

In the early stages, the MPB could work with a subset of line ministries, with coverage to be expanded following the initial round. In the absence of agreement with line ministries, the expenditure estimates risk not being treated as legitimate and can be unrealistic. Given the indicative nature of the process, this is acceptable initially, however as the process matures, and more weight is given to the estimates in determining budget and medium-term allocations, formal agreement between the Budget Office and line ministries will be necessary in order to both improve the quality of the estimates, as well as establish their credibility.

An important factor to remember while first producing forward estimates is that there is a trade-off between complexity and tractability in setting up expenditure forecast models. The forward estimates are models of how expenditure will evolve over the medium term. They are an approximation of reality; and should not necessarily be thought of as extending the appropriations into the medium term. They do not need to be prepared at the same level as appropriations. To do so would make the models far too complex, and difficult to both develop and operate. Similarly, the drivers of cost pressures will represent approximations of what drives spending. Taking the time early on to get the right balance between complexity and workability will make the forward estimates far more robust.

The first step is to understand the current spending base, and allocating it against the key spending areas within each ministry, program or objective. This requires knowing where and what the ministry is spending its money on, and how that is linked to key program outputs. In many ways, the program level provides a sensible level on which spending can be forecast on, although in some cases where there are multiple objectives these may need to be further disaggregated.

Major Steps in Producing the Forward Estimates

article image

The second step is to identify the current level of service delivery, or the no-policy change baseline. One of the important features of forward estimates is to separate the cost of providing existing levels of service, and the cost of introducing new policies. Thus, for education the current level of service delivery can be defined as the cost of maintaining enrolment rates, no matter what happens to the population of school age children. If the population of children were to increase, “no policy change” would see the number of students enrolled increase, but the enrolment rate would remain fixed. This is in contrast to an increase in student numbers due to an increase in the enrolment rate, which would be considered a policy change.

The third step is identifying the major spending areas within the spending units. These are often set up on an economic basis, such as salaries and wages; goods and services; transfers etc. However, at this point, it can be useful to break them up into key elements, so in the case of education, goods and services will be separated into textbooks, food and utilities. Usually there will be an ‘other’ category. As a general rule, this should make up no more than 10 percent of total spending within the unit. Each of these spending areas within the spending unit will be modeled separately.

The fourth step is to identify any one-off expenditures that need to be taken out to adjust the base. For instance, major IT upgrades in a given year may not be expected to occur next year, so the additional amount allocated for it should be excluded from the expenditure base. Similarly, there may be some one-offs that need to be accounted for specific years over the medium term. Classic examples include elections and census collections.

The fifth step is to identify the price parameters that will affect spending. A number of factors can affect the price of providing services. These factors, such as wages, inflation, specific input factor prices (such as fuel and utilities) need to be identified and analyzed for their relationship with expenditure growth, as changes will not always be one for one.

The assumptions used need to align with the relevant macroeconomic parameter forecasts. Often the price parameters will simply be linked to the overall inflation forecast. However, in some cases, there will be difference between the forecast of overall inflation and specific price parameters, such as fuel, which will be driven by world oil prices, and will have little relation to overall inflation. Again, the degree of detail here represents a trade-off between complexity and tractability. In general, the rule in setting price parameters should be to keep it as simple as possible, and only use unique parameters if large difference with overall inflation occur.

The sixth step is to identify changes in the volume parameters that drive the cost of providing the services. A number of factors will affect the amount of services being provided over the medium term. These factors can be due to demographic change (such as the population of school age people, or higher utilization of health services by older populations); policy changes, where previously increased capital expenditures require a higher level of maintenance once those projects come on line; and macroeconomic factors, such as a higher unemployment rate that requires a higher level of unemployment or social security benefits. Like the price parameters, these must align with the macroeconomic and demographic forecasts and projections.

The range of macroeconomic and fiscal forecasts prepared may need to be increased in order to provide the required parameter inputs. As parameters are identified, this will need to be communicated to the units who prepare the macroeconomic and fiscal forecasts within MOF, to check that they can be identified, and reach agreement on what basis the forecasts will be made on. For instance, the Consumer Price Index (CPI) forecasts may only be prepared for the overall CPI. However, the forward estimates may require parameters for fuel price CPI, so the macro forecasters may need to develop a methodology for forecasting the lower level CPI series.

The seventh step is to apply the overall price and volume parameters to the base expenditure to yield an expenditure forecast for the spending area. This is done by multiplying the adjusted base (i.e., accounting for one-off expenditures) by both the overall price and volume parameters. Note that there will be some interaction in the process, as price variations are applied to volume variations.23

The final step is to aggregate across spending units and spending areas to yield a forward estimate for the ministry, and then across ministries for the budget as a whole. The expenditure forecasts for the individual spending areas within a spending unit should be summed, then the spending areas summed in order to give the overall forward estimates for the Ministry. The same is true for the price and volume parameter variations, which can be used to provide a full reconciliation of the changes in the forward estimates.


IMF Article IV 2016.


An inflation rate of 7.2 percent has been applied to the 2016 base to calculate the 2017 ceiling, so as to roughly stick to the 2017 budget that was about to be voted at the time of the adoption of the amendment.


Article 107, §11 states that in case there is a positive gap between the primary balance outturn and target, this margin can be used to process the payment of RAP accumulated as of the end of December 2015, without being applied to the expenditure ceiling.


Between 2013 and 2016 central government revenues increased by 11.5% whereas the wage inflation of the combined period was 26.3%.


Institução Fiscal Independente, Relatorio de Acompanhamento Fiscal, Março de 2017, N 2.


Institução Fiscal Independente, Relatorio de Acompanhamento Fiscal, op. cit.


Iterative processes are used in macro-fiscal forecasting in many countries, including the US, Australia and France. For more information on the iterative process used in fiscal forecasting, see, for example, Leal et al., Fiscal Forecasting – Lessons from the Literature and Challenges, Working Paper No 843, European Central Bank.


Non-administrative revenue, such as dividends, concessions or sales of assets, are projected by the Treasury.


Upward adjustments introduced by the Congress during the budget approval process have also contributed to these deviations.


For more information on fiscal space assessment as recommended by the IMF, see Assessing Fiscal Space – an Initial Consistent Set of Considerations, IMF Policy Paper, June 2016.


For a more detailed discussion of the relevant preconditions see Medium-Term Budget Frameworks in Advanced Economies in Cangiano, Curristine. and Lazare (2013).


See Medium-Term Budget Frameworks in Advanced Economies in Cangiano, Curristine, and Lazare (2013) for a more formal discussion of the pros and cons of various MTBF design choices.


Von Hagen, J, “Budget Institutions for Aggregate Fiscal Discipline,” February 1998.


The role of RAPs execution as a parallel budget is a well-known in Brazilian budgetary and academic communities. See Gasparini, F, and de Almeida, M, “Plurianualidade Orgamentaria no Brasil: Diagnóstico, Rumos e Desafios”, Brasilia, ENAP, 2017.


Art 36 of the Law 4320 (year 1964) defines the RAPs as committed (empenhos) but not paid (pagamento) outstanding as of December 31st.


RAP have been a source of controversy and confusion since the Decree 93.872, art. 115. Some consider the RAP as floating debt (dívida flutuante). This reflects the country´s budgetary terminology and the fact that across countries the budget concepts are not always the same.


Carvalho, Munique Barros. Restos a pagar e a anualidade orçamentária. VI Prêmio SOF de Monografias. Brasília: ESAF. 2014, cited by Gasparini, F, and de Almeida, M, op. cit.


Lienert, I, and Ljugman, G, PFM Technical Guidance Note. Carry-over of Budget Authority, IMF, 2009.


The processed RAP are cancelled (Prescrição) after 5 years of existence (Decree 93.872, art. 70). The unprocessed RAP are cancelled on June 30th of the second year of existence (Decree 7.654, art. 1, $2).


Tesouro Nacional, Avaliacao dos Restos a Pagar 2017.


On top of the differences in coverage explained in Chapter II, two different sources are used to assess payments considered for each rule. Payments used in the primary balance target are the outward flows from the treasury single account, while payments considered for the expenditure rule are those which have been recorded in the SIAFI database. There is on average a one-day difference in terms of timeframe between both rules, which may create a very slight gap.


This would complement the government’s own long-term debt projections, which should be published alongside the budget, as recommended by the Fiscal Transparency Evaluation.


For instance, for a spending base of 100, a 10 percent increase in both price and volume will lead to an expenditure forecast of 121, greater than if price and volumes were applied individually, which would lead to a forecast of 120. This difference is described as the interaction term, and needs to be accounted for when reconciling the change in expenditure between the base and BY+1.

Brazil: Technical Assistance Report-Supporting Implementation of the Expenditure Rule Through Public Financial Management Reforms
Author: International Monetary Fund. Fiscal Affairs Dept.