Chapter 3. Auditing for more strategic and sustainable budgeting in Brazil

TCU could encourage more meaningful medium-term perspectives in the budget by (a) auditing overall performance of multi-year sectoral plans, and (b) highlighting key findings on the alignment of annual budgetary allocations with multi-annual priorities, emphasising implications for entity-level planning.

Decades of OECD experience in working on good budgetary governance show a predominant trend in seeking a greater link between strategy and budget. Upstream allocation decisions of the budgeting process, including those decisions meant to be taken in the interests of citizens, determine which government priorities will or will not be funded. A mismatch between policy objectives and annual resource allocations can lead to unfulfilled strategic goals and objectives (OECD, 2014a; 2015b; 2016b).

Governments, particularly institutions within the Centre of Government (CoG) responsible for planning, often elaborate a vision or prioritisation that is relatively short-term. Medium- and long-term planning requires resources, capacity and skills that the CoG may not have. A long-term strategy has the benefit of setting priorities that span electoral cycles and thus that provide a general guideline against which governments should pursue, regardless of political platform. Brazil’s political actors have been able to achieve broad consensus on long-term policy objectives that has been useful in poverty reduction, growth stimulation and depolarization of policy conflicts that came about during prosperous economic times (Bertelsmann Stiftung, 2016). Yet, Brazil does not have a long-term plan at the federal level. Medium and long-term plans exist within select sectors, ministries and regions. Brazil’s commitment to the achievement of the Sustainable Development Goals for 2030 provides broad aims for the government over the long-term.

One of the major developments in budgetary governance since the turn of the century has been the increased focus on the multi-annual dimensions of budgeting. Medium-term perspectives in budgeting refers here specifically to the adoption of Medium-term Fiscal Frameworks (MTFFs) or Medium-term Expenditure Frameworks (MTEFs) that strengthen the ability of the government collectively, and the Ministry of Finance (Ministério da Fazenda, MF) in particular, to plan and enforce a sustainable fiscal path.3 MTFFs are increasingly relevant in a context where the multi-annual character of policies require an extended time horizon, such as large capital projects, new programmes, and organisational reform and restructuring (OECD, 2014a; 2015b, 2016b).

If properly designed, an MTEF should force stakeholders to deal with the medium-term perspective of budgeting and budgetary policies, typically covering a period of three to four years. It should also facilitate stakeholders in identifying the policy choices and trade-offs that will be necessary in light of the estimates of what would happen in the following three to five years based on unchanged policies. From the point of view of managers, in-line ministries and agencies, medium-term frameworks put them in a better position to plan their policies and operations, to the extent that they have some level of visibility regarding the likely level of funding beyond the next budget (OECD, 2014b).

Brazil’s PPA is not an MTEF4 , but shares characteristics with MTEFs of OECD member countries (OECD, 2014b). The aim of the PPA, like that of an MTEF, is to improve the quality and certainty of multi-annual fiscal planning (OECD, 2015a). The PPA includes a strategic dimension intended to guide the preparation of thematic programmes, which describe the government’s agenda in different areas, such as social security and public investment. In setting entity-level objectives, public managers are encouraged to link their entities’ goals to the main themes of the PPA. According to stakeholders within and outside of TCU, the PPA does not provide sufficient guidance or criteria for entities to plan.

The current PPA (2016-2019) seeks to enable the government to better demonstrate and facilitate an understanding of how the overall government strategy connects to objectives and goals at the entity level (MP, 2016a). The MP has played an important role, together with the Ministry of Finance, in trying to foster a link between planning and budget despite budget rigidity. However, in practice, there are misalignments that do not transfer the intended benefits of the PPA, as set out in the constitution, into reality. These limit the extent to which Brazil’s budgetary framework can allow for a medium- or long-term strategy and sustainability. Amongst other effects, the misalignment also makes it difficult for managers to know their programme targets and to accurately project performance and results (TCU, 2016a).

The PPA’s thematic programmes around important areas such as environment and regional development are meant to provide a framework for planning across sectors, such as in the National Education Plan (PNE). This reflects the strides made in recent years to adopt a more cross-cutting and forward looking view into its PPA – now structuring each 4-year PPA cycle around thematic outcomes. The Constitution spells out that the 3 budget laws – the PPA, the LDO and the LOA – should be aligned. In the absence of an explicit long-term plan at the federal level, long-term ambitions and indicators in Brazil are detailed through comprehensive sectoral plans, such as the National Education Plan (PNE). The PPA’s thematic programmes provide policy goals over the medium and long term. However, in practice, there are misalignments that do not transfer the intended benefits of the PPA, as set out in the constitution, into reality. These limit the extent to which Brazil’s budgetary framework can allow for a medium- or long-term strategy and sustainability. Amongst other effects, the misalignment also makes it difficult for managers to know their programme targets and to accurately project performance and results (TCU, 2016a).

Heavily funded programmes, which were incongruent with austerity measures, were evidence of a need to strengthen the link between planning and budget, even before the economic downturn (OECD, 2015c). The sustainability and strategic nature of Brazil’s budgetary framework is influenced by the rigidity and large proportion earmarked to particular expenditures. A 2009 assessment using the Framework for Public Expenditure and Financial Accountability (PEFA) found that during the execution of the budget, Brazil’s Executive restricts the release of funds for discretionary commitments and spending in a way that is often inconsistent with PPA. PEFA (2009) gave Brazil a grade C in ‘frequency and transparency of adjustments to budget allocations’. In-year adjustments, according to TCU, mean that the LOA ends up leading to adjustments in the PPA itself, instead of the other way around. Finally, TCU has detected a lack of synergy between individual sector-based plans and the PPA.

Medium-term budgetary planning yields a number of benefits for governments, ensuring that policies and programmes deliver results for citizens. For instance, one benefit of effective medium-term budgetary planning is that it should work to mitigate particular activities that may be considered as classic budgeting “games”, such as: i) capital projects being launched with changing operating costs; ii) programmes coming into effect late in the budget year, thus not exposing their full costs in the initial year; and iii) programmes’ spending implications not being fully reflected under the circumstances prevailing during the budget year, but will become so in out-years.

Another benefit of effective medium-term budgetary planning is in the flexibility it grants in spending, allowing governments to fund programmes that require multi-year funding in order to realise policy goals. Brazil’s budgetary framework allows for carryover of unused funds or appropriations from one year to another in order to facilitate multi-year budgeting for discretionary, operational and investment spending (OECD, 2014b). This flexibility in expenditure is important for multi-year programmes, when managed within reason, particularly in view of Brazil’s otherwise rigid budget. Brazil is the only country in OECD’s 2013 survey of select Latin American and Caribbean countries to allow line ministries to carry over unused funds without thresholds for both operational and investment expenditures (OECD, 2014). Of 32 OECD member countries, carry-over without threshold for operating expenditures is allowed in 34% of countries, and for investment expenditures in 43% of countries’ (OECD, 2014b).

TCU has been active in auditing budgetary planning, including the PPA, Budgetary Guidelines and law (LDO and LOA). As these budgetary instruments are created as laws, and not programmes, TCU’s corresponding work does not comment on the priorities chosen or merits within those laws, nor the recipients of the 35% ‘priority’ spending (OECD, 2003). Moreover, as an SAI, TCU does not audit the judgement behind the design of laws or policies. However, TCU can assess the reliability and achievement of objectives set within those laws. Many peer SAIs, like TCU, have been active in this space.

An OECD survey of 10 SAIs showed the extent to which they were active in assessing auditees against international budgetary standards. Nine of ten had assessed the adequacy of budgetary planning processes, including: to facilitate an alignment between budgeting and strategic objectives; to ensure reliability and quality of tools that underlie the budget; to manage public debt; to assess long-term sustainability; and to guide the multi-annual process of resource allocation. Ten out of ten SAIs surveyed had assessed the effectiveness of the procedures in place for managing, monitoring and overseeing financial allocations, including: the compliance and consistency of in-year budget reallocations; the adequacy of in-year budget execution reports (OECD, 2016a).

TCU recently released a comprehensive audit of Brazil’s budgetary system using OECD principles espoused in the Recommendation of the Council on Budgetary Governance (OECD, 2015a). One main component of the audit focused on principle 2 of the Recommendation, on the need to “Closely align budgets with the medium-term strategic priorities of government”. TCU found a weak alignment between the 3 budgetary documents (Box 3.1). The alignment, or misalignment, is found primarily in the lack of integration of the PPA’s thematic programmes in the LOA. The low degree of implementation calls into question the ability of the programme goals to be met in by the year end, and thus by the end of the current PPA. Other key findings of this recent TCU audit in this area are summarised in Box 3.1.

To help the government address these challenges, TCU can develop its audit portfolio to improve budget processes and planning, and provide meaningful insights for budgetary decision making in the Congress. Given the governments focus on short-term priorities, as discussed, TCU’s audits and findings could continue to emphasise a perspective that looks beyond immediate objectives and challenges. This is particularly relevant in the coming years, as Brazil takes on the challenge of effectively designing and implementing policies to achieve the Sustainable Development Goals (SDGs). The SDGs, by their nature, have a medium to long-term horizon for the achievement of outcomes. In addition, the PPA is not as effective as intended in guiding annual allocations, making sectoral plans more critical for economic and social progress over the medium and long-term. Where sectoral plans exist, such as the National Education Plan (PNE) spanning from 2014 to 2024, they could feature more consistently as the focus of audits. As a result, TCU’s work on medium-tern budgeting would not solely be tied to the PPA.

Building on TCU’s existing work in this area, TCU could offer meaningful medium-term perspectives for improved budgetary planning in the following ways:

• TCU could assess overall performance of medium and long-term sectoral plans, focusing on 1) adequacy and execution of annual allocations for their goal achievement, 2) the impact of any relevant thematic programme goals set out in the PPA, and 3) the implications of performance for achievement of related SDGs; and

• TCU could continue to assess and communicate key findings on the alignment of multi-annual priorities with annual budgetary allocations, emphasising the importance of entity-level planning with a longer time horizon.

TCU could help to improve the reliability of evidence and information for making performance-based budgetary decisions through assessments of key areas, such as expenditure programmes funded by earmarks and the effectiveness of operational cost measurement.

Difficult budgetary decisions require objective and factual information on what works well and what does not. In 2015, the OECD recommended that Brazil’s policy makers review the earmarking of public revenues and mandatory spending floors with a view towards making public spending more responsive and flexible to the policy priorities (OECD, 2015c). These proposals reflect international standards and good practices. For instance, OECD’s recommendation on good budgetary governance stipulates that performance, evaluation and value for money be integral to the budget process (OECD, 2015a), including:

• evaluating and reviewing expenditure programmes – including associated staffing resources as well as tax expenditures – in a manner that is objective, routine and regular to inform resource allocation and re-prioritisation both within line ministries and across government as a whole (Principle 8d); and

• taking stock, periodically, of overall expenditure (including tax expenditure) and reassessing its alignment with fiscal objectives and national priorities, taking account of the results of evaluations; noting that for such a comprehensive review to be effective, it must be responsive to the practical needs of government as a whole (principle 8g).

In recent years, OECD and other countries have increasingly sought to develop a focus on the results achieved with the appropriations allocated through various approaches under the heading of “performance budgeting.” Performance budgeting frameworks are widely found in OECD countries and are generally flexible in nature. In general, the different policy sectors aim to use performance information for setting allocations for programmes, line ministries and agencies, for strategic planning, for proposing new areas of spending, and for a reduction in spending.

In reality, there is currently no settled consensus on the optimal way of using performance information within the budgeting context. The issues arising range from the level of information being used, the problems inherent in measurement of public performance and concerns that performance budgeting would displace other mechanisms for accountability and monitoring of performance.

A 2013 OECD survey showed that, generally, when performance targets are not met, the consequences for managers of ministries can be limited. This suggested that performance information was not easily transformed into action, as was the case in Brazil (Table 3.2). More recently, case studies of 5 OECD member countries (Canada, Ireland, the Netherlands, the United Kingdom and the United States), demonstrated that the performance public managers are increasingly evaluated with use of performance information in mind (OECD, 2016b).

Brazil uses elements of performance-based budgeting (OECD, 2014b; IBP, 2016; Bertelsmann, 2016). However, the quality of the information used as inputs into the budget (underlying assumptions) and the results reported against the budget are questionable, according to TCU. There is oversight from constitutionally appointed bodies, such as TCU and the Ministry of Transparency, Supervision and Control (Ministério da Transparência, Fiscalização e Controle, CGU), as well as mechanisms that legally guarantee citizen engagement and feedback on programmes and services. In addition, there is a law (Complementary law 101/2000) which establishes a council to assess the performance of agencies and actors in the formulation of the budget, and the Senate is obliged to conduct annual evaluations.

In practice however, there are concerns with the quality of information and its meaningful use in making budgetary decisions (IMF, 2014; TCU, 2016a). Moreover, Brazil’s LOA and LDO do not identify the budgetary impact of new policies, making it difficult to anticipate implications for the budget (IMF, 2014). When performance targets are not met, consequences for managers of ministries can be limited, as is the case in Brazil (OECD, 2014b). A recent OECD report on trends in performance-based budgeting found that general programme evaluations can be useful, but lack a budget focus. It suggests:

In decentralised systems, it is rare for a central budget authority to conduct evaluations themselves and instead rely on evaluation of ministries and programme managers. Thus, if evaluations are not producing results that are comparable, the evaluations become less useful from a budgetary perspective. To date, reviews that have a specific budget focus – such as spending reviews – tend to provide more focused evidence that can be input into the budgetary process (OECD, 2016b).

A further issue is that evaluations often lack commentary on efficiency or effectiveness, as well as the success or shortfalls in meeting programme expectations. Instead, evaluations are prone to equivocal explanations that, again, make it difficult for budget officials to take a horizontal, comparative view on performance implications for the budget (OECD, 2016b). Such expectations and pitfalls of performance-based budgeting frameworks will be useful for Brazil’s government to keep in mind in this first year of the “Committee on Monitoring and Evaluation of Federal Public Policies” (CMAP), established by the federal government in April 2016. Established as an inter-ministerial committee, CMAP seeks to improve actions, programs and public policies, as well as the application of resources and the quality of public spending (Internal ordinance Number 102) (MP, 2016b; 2016c; 2016d; 2016e). It aims to generate results from across the administration that can better inform performance, as well as to serve in the elaboration and execution of the budget. Learning from OECD member countries, the CMAP may benefit from integrating budgetary-focused reviews into its basket of evaluative tools.

SAIs have been attempting to link results of policies and programmes to the budgetary assumptions and decisions that underlie them in order to promote more informed and evidence-based decisions. In a survey of ten SAIs, eight had assessed whether there are adequate mechanisms to generate and capture quality performance information during the phase of budget execution (OECD, 2016a). TCU’s own work on governance reflects this approach. One example is TCU’s systemic report of the labour function (FISC-Trabalho) that looked at the allocation of resources to labour policies, the subsequent monitoring of indicators and progress, or lack of, with regards to previous audits. The audit pointed to sustainability issues with the Support Fund for Workers (Fundo de Amparo ao Trabalhador), and an imbalance between revenue and expenditure (TCU, 2015c).

TCU has also looked at elements of performance and value-for-money elements in budgeting through other activities. For instance, TCU undertook a monitoring report on the procedures for drawing up the PPA for (2008-2011), with emphasis on the analysis of performance indicators linked to the various programs that make up medium-term budget. The work aimed to map out a diagnosis of the main problems and challenges faced by the planning system involved in the PPA design task (TCU, 2009a).

Linking outcomes to budgetary decisions is a challenging pursuit, but doing so is critical for developing a clear picture of Brazil’s federal performance year-on-year (TCU, 2016a). More reliable and relevant performance information, demonstrating what is working and what is not, would further help Brazil to determine trade-offs based on reliable evidence. This also helps to protect the budgeting processes from special interests, thereby creating trust in government spending decisions. TCU’s audit reports can serve as critical inputs into the process. To effectively inform budget deliberations and resource allocations, TCU’s report should be published in a timely, relevant and objective manner, in accordance with international standards (INTOSAI, 2013). In addition, TCU could further help to improve the reliability of evidence and information for making performance-based budgetary decisions. Areas TCU could consider as a focus of future audit work to this end include the following:

• The usefulness of budget reviews and evaluations for decision-making: TCU could look at existing reviews or evaluations for their link to the budget, as a separate endeavour to assessing evaluations for their reliability in general (discussed in Chapter 5). A 2014 IMF review of Brazil’s budgetary framework recommended the introduction spending reviews that would encompass mandatory expenditures (IMF, 2014). TCU could also assess the extent to which CMAP’s provisions for gathering budgetary related results will yield informative and reliable findings.

• Reliability of information of policies related to earmarking of public revenues: The government has produced ad-hoc evaluations on tax expenditures, but the government does not provide systematic reviews or an overarching view of results obtained through these increased expenditures. In 2011, Brazil’s tax revenue was highest amongst Latin American and Caribbean countries, amounting to 35% of GDP (OECD, 2014b), but much of it earmarked and limits the flexibility of government to allocate spending between areas (OECD, 2015). These tax-related expenditures are approved by Law but by different legal instruments and at different times than the LOA. According to TCU officials, these are also not under same scrutiny by Congress. In 2015, the OECD recommended that Brazil’s policy makers review the earmarking of public revenues and mandatory spending floors with a view towards making public spending more responsive and flexible to the policy priorities (OECD, 2015c). As mentioned, the IMF recommended that Brazil introduce spending reviews that encompassed mandatory expenditures (IMF, 2014).

• Reliability of measurement of operational costs: Another area for focus could be the periodic assessment of government’s evaluations of its operational costs. This requirement was foreseen in the fiscal responsibility law of 2000. The Ministry of Finance (MF) began developing a system for measuring costs of activities as was predicted approximately 4 years ago. While the implementation has been gradual, there may be an opportunity for TCU to assess the quality of any existing reports as well as of the underlying review process itself. This could act to accelerate the process, anticipated in 2000. This process will be an important means of supporting budgetary trade-offs made to operational expenditure. TCU may be inspired by the Netherlands Court of Audit for instance (Box 3.5), which found that a merger, intended to produce savings for the government by cutting operational costs, would fail to materialise because of faulty projections on which the decision for the merger was based. This is a good example of SAIs using foresight to highlight the need for quality, objective evidence for more sustainable decisions.

More consistent evaluation by TCU in these areas would help to make off-budget expenditures or expenditures under different oversight processes, more transparent. Due consideration should be given in advance to the oversight procedures that have been established, and whether there is an opportunity for this to be strengthened before TCU engages.

TCU may be further inspired by the SAIs of Canada, Korea and the Netherlands in which the SAI assessed workforce sustainability of government plans to manage population ageing, which will also become an issue for Brazil (Box 3.6).

TCU could enhance its contributions to the dialogue on fiscal discipline by periodically assessing in-year progress reports to discern how adjustments to fiscal targets and plans affect the achievement of goals.

With the introduction of the Fiscal Responsibility Law (Lei de Responsabilidade Fiscal, LRF) in 2000, Brazil moved towards a more established pattern of mid-term fiscal management that would provide more stability.5 Further specifications on fiscal rules and responsibility were made in the supplementary Law 229, originating in 2009 and being updated in 2015, which outlines norms on planning, budget control and public accounting for public sector management (Brazil Senate, 2015). The requirements outlined in the LRF essentially act as fiscal rules, which are concrete indicators of fiscal management.6

Fiscal rules can play an important role in credibility of national budgeting by constraining politically-motivated spending to improve macroeconomic stability in the short-term, while contributing to long-term fiscal sustainability. Further, fiscal rules should make it easier to understand and to anticipate the government’s fiscal policy course throughout the economic cycle (OECD, 2015d). Brazil’s expenditure commitments, budget allocations, and revenue projections tend to be adjusted fluidly in year, weakening the institutional link between planning and budgeting in practice that is set out in the Constitution (IMF, 2014; PEFA, 2009; TCU, 2016a). The International Monetary Fund (IMF) found “the Constitutional powers of Congress to increase revenue estimates and the President’s powers based on the LRF to issue decrees to reduce expenditures regularly lead to an inefficient and time-consuming budget process” (IMF, 2014).

Periodic in-year reporting on budget execution and revision of budget estimates has helped OECD member countries to detect and manage the impact of economic developments and other factors on the budget in a timely manner. Such reports identify any changes to the assumptions underlying the budget, as well as other relevant events that have transpired during the fiscal year. Thus, quality in-year reporting becomes important to ensure that entities can manage in accordance with fiscal adjustments, and continue to work towards the achievement of goals and objectives. In Brazil, executive entities provide reports on fiscal targets and compliance with fiscal rules bimonthly and every quarter, marking points of control, or “pontas de controle” (MP, 2016b). The LRF requires the Executive to report to the CMO every 4 months to explain the evolution of fiscal objectives. They are sent to CGU for review. The LRF also mandates that Brazil’s TCU (and TCEs) audit the enforcement of the law’s procedural rules.

There are fewer TCU examples targeted to this area of budgetary governance when compared to the other areas explored in this chapter. TCU touches on fiscal discipline as part of its accounts examinations process. It also featured tangentially in the review of budgetary governance, explored at length in recommendation 3.1.

Given the importance of in-year reporting and the potential of in-year adjustments to affect strategic goals, TCU could audit in-year progress reports to understand how adjustments to fiscal assumptions are affecting entity and programme outcomes at the end of the year. Specifically, such audits could help TCU and audited entities to understand how the quality and decision making related to in-year reporting and adjustments to fiscal targets affect the annual or multi-annual goal attainment of the specific entity or programme. This work would allow for real-time feedback on a key governance mechanism, so that additional adjustments could be made, as needed, to ensure the success of policies and programmes.

TCU would be well placed to contribute further to the discussion on fiscal discipline and responsibility. It can do so with both Congress and the Executive. Indeed, part of TCU’s impact in strengthening more sustainable budgeting rests on improving communication particularly with the CMO. Further, TCU could consider more frequent communication of key findings with central entities involved in budget and planning (MP and MF).

TCU could further assess whether adequate institutional procedures and safeguards are in place to ensure the quality, reliability and sustainability of budget forecasts and medium-term projections.

Public fiscal sustainability is the ability of public governments to maintain public finances at a credible and serviceable position over the long run, in light of the prevailing mix of spending and revenue policies. Prudent macroeconomic assumptions, sensitivity and risk analyses, as well as appropriate fiscal rules, are factors that can help to orient spending and revenue policies within sustainable levels in the short and medium term.

Long-term fiscal projections can play a useful role in identifying the expected future costs and associated debt of current policies in light of forecasted demographic and economic developments. In most OECD countries, the Central Budget Authority (CBA) is responsible for long-term fiscal projections. In some countries this responsibility is located in other departments of the Ministry of Finance other than the CBA or in other core ministries. Some OECD governments also present long-term fiscal challenges in a cross-generational perspective, in order to distribute benefits and cost in a reasonable way over generations.

In Brazil, projections are prepared jointly by the MP and the MF. Any ex-ante comparison of the government’s macroeconomic and fiscal forecasts with those of other institutions goes unpublished (IMF, 2014). Past assessments of Brazil’s preparation of multiyear fiscal forecasts and functional allocations have identified room for improvement (PEFA, 2009). Despite the strong institutional framework for creating projections, adjustments are made to the revenue projections and discretionary spending during the Congressional approval process (Hallerberg et al., 2009; PEFA, 2009; IMF, 2014; TCU, 2016a). The IMF has recently suggested that Congress should be required to use forecasts prepared by independent parties, which are perceived to be more objective in the event the Congress rejects the estimates of the MP and MF. This could also help to ensure that increases in revenue estimates by Congress, in exercise within its constitutional powers, are measured and evidence-based.

In Brazil, bodies of expertise that could provide insight on fiscal sustainability, apart from the MF, have come and gone. Establishment of an Independent Fiscal Institution (IFI) is one way in which OECD countries are trying to improve the credibility of budgetary forecasts and enhance independent oversight with regard to fiscal performance. A proposal to establish an IFI in the Brazilian Senate is currently incorporated into draft law number 65.7 The creation of IFIs reflects a trend of invest in resources and institutions that facilitate Parliamentary engagement in the budget process. TCU, and other bodies in Brazil, can look to OECD’s Recommendation of the Council on Principles for Independent Fiscal Institutions (OECD, 2014c), for both guidance in its establishment or criteria against which an eventual IFI could be evaluated.

An SAI is another institutional mechanism for supporting adequate institutional procedures and safeguards for quality economic and budgetary forecasts. OECD’s “Supreme Audit Institutions and Good Governance: Oversight, Insight and Foresight” (2016a) demonstrates how SAIs are contributing to budget sustainability. They are active in using foresight to assess the reliability of fiscal projections in order to ensure that underlying assumptions of the budget are sound. Nine of 10 SAIs, including TCU, are looking to the reliability and quality of tools that underlie the budget. Eight of 10 had assessed the adequacy of other mechanisms for determining the quality and reliability of budgetary forecasts and fiscal plans, and managing accordingly (OECD, 2016a).

TCU is responsible for assessing compliance of entities with the Fiscal Responsibility Law, and its more traditional role has been to monitor fiscal targets. TCU’s recent work in assessing the budgetary framework against OECD’s Recommendation of the Council on Budgetary Governance looked at the ability of public administration to achieve consistent and intact budgetary forecasts. Select examples of TCU’s work in this area include:

• Evaluation of the revenue forecast of the year 2013: Seeks to verify the adherence of the statements presented to the constitutional and legal requirements, particularly the Fiscal Responsibility Law (TCU, 2013b).

• Audit of financial position of government: Follow up report to check that the financial statements of Ministry of Finance reflect, in all material respects, the financial position at the end of year (TCU, 2015d).

• Audit on the reliability of the budgetary forecasts and fiscal plans like revenues, expenditures, fiscal targets, national economic forecasts and central government primary balance (TCU, 2015e).

• Survey report to identify the criteria used to forecast macroeconomic indicators that are used for preparing the Annual Budget Law, to compare the predicted values with the realized and to check for possible discrepancies and effects on the budget execution and government finances (TCU, 2014a).

• Evaluation of the revenue forecast of the year 2014: Seeks to verify the adherence of the statements presented to the constitutional and legal requirements, particularly the Fiscal Responsibility Law (TCU, 2014b).

• Compliance and monitoring report, Follow-up: Revenue forecast evaluation of the federal budget bill 2015, to assess the estimates of revenue contained in the Budget Bill of the Union for 2015. This report showed that projected revenue estimates did not match the market expectations (TCU, 2014c).

The recommendations are not meant to suggest a role for TCU in long-term projections or conducting fiscal forecasting, but instead how it can support their quality and use them for pointing to the sustainability of the programmes being funded. In fact, TCU’s future audits could delve further into the subject, to assess whether Congress specifically is equipped with the institutional capacity, including independent professional advice and critique, to meet OECD principle 10 to “promote the integrity and quality of budgetary forecasts, fiscal plans and budgetary implementation through rigorous quality assurance including independent audit” (OECD, 2015a).

In addition, future TCU audits could use fiscal projections to ascertain progress or advancements in the quality of forecasts over time, and to more systematically assess sustainability of programmes, as has been done by the SAI of Mexico (Auditoría Superior de la Federación), which assessed the assumptions on which decisions related to the deficit were made. As a starting point, ahead of the establishment of the IFI, TCU could analyse and collect key findings from audits already conducted in order to identify the processes and conditions that produce quality fiscal targets. In addition, according to external stakeholders, TCU could integrate these conditions as criteria into its audits. Other approaches include those of the SAIs of Canada, the United Kingdom and the United States (Box 3.7).