Brazil: Selected Issues and Statistical Appendix

This paper analyzes several issues regarding fiscal sustainability and fiscal adjustment in Brazil during 1990 and searches for econometric evidence of a monetary dominant regime during some subperiods. The following statistical data are also presented in detail: macroeconomic flows and balances, industrial production, consumer price index, relative public sector prices and tariffs, minimum wage statistics, financial system loans, monetary aggregates, exports by principal commodity groups, direction of trade, detailed balance of payments, total external debt, central government operations, and so on.

Abstract

This paper analyzes several issues regarding fiscal sustainability and fiscal adjustment in Brazil during 1990 and searches for econometric evidence of a monetary dominant regime during some subperiods. The following statistical data are also presented in detail: macroeconomic flows and balances, industrial production, consumer price index, relative public sector prices and tariffs, minimum wage statistics, financial system loans, monetary aggregates, exports by principal commodity groups, direction of trade, detailed balance of payments, total external debt, central government operations, and so on.

II. Medium and Long-term Fiscal Sustainability in Brazil: 2000–101

A. Introduction

1. A main challenge facing Brazil in the medium term is to consolidate the fiscal adjustment effort undertaken since 1998Q3 and ensure the conditions for sustainable balanced growth. To achieve that goal, fiscal policy should be crafted to allow a smooth and nonnegligible decline of the debt to GDP ratio during the decade, in order to provide the necessary flexibility in the management of monetary policy and reduce the vulnerability of the country to adverse macroeconomic shocks—while leaving room for the continuation of the process of identification and recognition, in nondestabilizing fashion, of some “contingent” liabilities of the Treasury generated during the high inflation period of the 1980s and early 1990s. Actively pursuing a declining path of the debt to GDP ratio is also warranted on account of the fact that this ratio has increased substantially since 1995, particularly during the high interest rate period of 1997–99, and because the current composition of public debt, being far from ideal (short-term structure and biased toward floating rates), can sometimes generate enhanced vulnerability concerns among market participants regarding roll-over risks, which are often translated into higher risk premiums on both domestic and foreign financing.

2. Brazil has, however, recently taken several steps in the direction of institutionalizing fiscal discipline, notably through the approval of the Fiscal Responsibility Law (FRL), which, among other things, establishes that no permanent expenditure item can be created without prior identification of a new permanent source of revenue or permanent cut in previous expenditure items; sets ceilings on personnel expenditures and on public debt by level of government; curtails moral hazard by eliminating the scope for bailouts of subnational governments; and increases transparency and accountability in the management of public finances. Further, Congress has recently approved an ancillary piece of legislation to the FRL that stipulates the judicial and administrative sanctions to be impose upon those that violate the precepts of the FRL.

3. From a consolidated primary deficit of around 1 percent of GDP in 1997 and an even balance in 1998, Brazil has successfully managed to post fiscal primary surpluses in excess of 3 percent of GDP in 1999 and so far in 2000 (see Table 2.1 and Figure 2.1). However, given the short-term rigidities of several expenditure items, the fiscal adjustment effort carried out during 1998–00 was necessarily biased toward increases in revenues, many of which nonrecurrent (concessions, tax rate hikes such as the CPMF and Cofins, and incentives to collect tax arrears), and a substantial real retrenchment in discretionary public spending. From now on, the desirable fiscal effort should be anchored in structural fiscal reforms, which should lead to a more equitable distribution of the tax burden, enhanced microeconomic efficiency of the economy (e.g., tax reform), and the achievement of higher efficiency and effectiveness of public spending (e.g., social security reform).

Table 2.1.

Brazil: Primary Surplus 1997-2000

article image
Source: IMF staff estimates.
Figure 2.1.
Figure 2.1.

Brazil: Nominal and Primary Deficit (% of GDP; 12-months); 1997M1-2000M9

Citation: IMF Staff Country Reports 2001, 010; 10.5089/9781451805901.002.A002

Source: Central Bank, and Fund staff estimates(-) Surplus, (+) Deficit.

In the words of Blanchard et al (1990);

“… sustainability is essentially about whether, based on the policy currently on the books, a government is headed towards excessive debt accumulation.”

4. To gauge whether fiscal developments during the period 2000–10, as far as we can anticipate them, will lead to unsustainable debt accumulation and to assess the margin of maneuver the government might have in fiscal management without rekindling unsustainable debt dynamics, we will in this section construct a minimalist above-the-line scenario for the evolution of the main revenue and expenditure items of the central government. With this, we will then derive a path for the primary surplus of the consolidated public sector based on a stylized conjecture for the evolution of the primary surplus of the subnational governments and the state-owned public enterprises at all three levels of government.2 Based on a macro framework construct we will then discuss the implied net public debt profile of the central scenario and perform stress tests by varying key macro variables that impinge on the debt dynamics and the primary surplus, such as the level of real interest rates, real GDP growth, the depreciation of the real versus the U.S. dollar, and the real rate of growth of federal discretionary expenditure and personnel expenses.

B. Primary Surplus of the Central Government: Central Scenario

5. For the central government (CG) we will assume that, apart from a few well defined items, tax revenues will show a unitary elasticity to nominal GDP. Revenue items projected separately involve the CPMF tax collections, concession revenues, dividends, the Oil Account surplus (PPE), and the revenues of the social security system for private sector workers (INSS). The assumption of unitary elasticity of revenues has not been validated empirically, at least in the short run, as there is nonnegligible heterogeneity of the tax burden across economic sectors. For instance, exports are usually not taxed as final products—but are taxed in the intermediate stages of production by cascading contributions,—and the agriculture sector is lightly taxed when compared with the industrial sector. From the income side, there are also sizable discrepancies in the taxation of wages, profits, and other income. These discrepancies tend to be stronger during periods of sharp acceleration or deceleration of economic activity, and are of material importance for short run projections, being less important for long-term scenarios. Here, as a simplification, and without loss of generality, for the long term scenario we assume balanced growth across sectors on average during the decade, that is, we assume an expansion of the taxable base in line with nominal GDP.

6. In the baseline (central scenario: no policy action) we assume a financial transactions tax (CPMF) rate of 0.30 percent from June 2000 to June 2002, and zero afterwards. Approval of the Poverty Fund, involving a constitutional amendment currently under review in congress, might lead to an increase of the tax rate back to 0.38 percent (coupled with a surcharge of the federal VAT tax (IPI) on luxury goods). However, we assume that the approval of the Poverty Fund’s earmarked sources of revenue will lead to a concomitant increase in expenditure, leading to, in essence, no impact on the primary deficit. As such, since the Poverty Fund has not yet been approved, in our scenario, CPMF revenues are still expected to decline in 2001 an 2002 from the level observed in 2000, reaching zero during 2003–10 (see Table 2.2).

Table 2.2.

Brazil: Baseline (Central Scenario); 2000-2010

article image
Source: IMF staff estimates.

Real growth (drift) of 2.0 percent per year from 2002-2010.

Real growth (drift) of 4.0 percent per year from 2002-2010.

Real growth of free discretionary expenditure (excludes education and health care) of 0.0 percent per year from 2005-2010.

Assumes that the tax collections of the states and municipalities will remain constant at the 9.4 percent of GDP collected in 1999.

7. Concession revenues are derived separately since they tend to be bulky and nonrecurrent. In 2001 the government expects to get R$8.7 billion (0.7 percent of GDP) in revenue from this source3. A residual amount of around 0.2 percent of GDP is expected to accrue in 2002. In the period 2003–05 we expect again significant concession revenues from the auction of the high speed 1.8 GHz data, voice, and image transmission frequencies (third generation frequencies, GSM-Global System for Mobility technology). The GSM concessions could generate around 0.3 percent of GDP per year during 2003–05.

8. The Oil Account (PPE) surplus, projected at 0.53 percent of GDP in 2001, is expected to be reduced to more moderate levels from 2002 onwards. As a working assumption, we presume that the approval of an explicit taxation system for oil products and it derivatives after 2001, as a substitute for the current cross-subsidization scheme, could yield around 0.35 percent of GDP per year on average during 2002–10.

9. The distribution of dividends to the Treasury by federally-owned enterprises (financial and nonfinancial) has been magnified in recent years by the balance-sheet effect of the floatation of the real in early 1999 on the profits of the federal banks and by exceptionally high international oil prices. With the continuation of the government’s federal privatization program, the sale of noncontrolling stakes in some enterprises, and more moderate levels of crude oil prices, we expect this revenue item to reach around 0.14 percent of GDP on average from 2003–10 (see Table 2.2).

10. Regarding the revenues of the INSS, barring large changes in the share of formal/informal labor relations or a change in the current contribution rate structure, we assume that revenues will average around 5.1 percent of GDP during this decade. Clearly, this requires that the real wages in the formal sector of the economy grow with real GDP, thus maintaining the share of formal sector labor income in total income (see Table 2.2).

11. Finally, transfers to the states and municipalities are expected to remain constant as a percentage of GDP since both income tax and the IPI (the two main shared taxes) are expected to track the evolution of nominal GDP.

12. Hence, with the termination of concession revenues in 2005 and CPMF collections in 2002, and with the decline in dividend collections and the PPE surplus to more moderate levels after 2001, total revenues of the federal Treasury are expected to drop by over 2 percentage points of GDP after 2005 from the level expected in the budget proposal for 2001 (see Table 2.2).

13. With regard to nonfinancial expenditure, it is a well known stylized fact that, apart from a fraction of the other current and capital expenditure item (OCCs), in the short run, the margin for maneuver is limited.

14. Despite the government’s policy of granting no generalized wage increases for the last five years, expenses with personnel (active and inactive) have nevertheless, in recent years, shown a significant nominal drift on account of factors related to the age-structure dynamics of the public service (affecting expenses with retired civil servants), the restructuring of some careers’ pay structure, the normal career progression of individuals (career drift), selective wage increases granted to a few specific occupations, and extra payments derived from judicial decisions. Taking this into account, in the baseline scenario we assume that the nominal wage bill will grow between 4 percent and 5 percent, respectively, per year on average from 2002 until 2010. Notice that this assumption does not necessarily preclude real wage gains in the public sector since we are assuming away any other potential savings on personnel expenses. If, in the future, the assumed nominal growth of personnel expenses turns out to be smaller than the one assumed, or if, through further streamlining and efficiency gains the workforce is reduced, for the same wage bill, these savings/gains could be passed on to the remaining public servants in the form of real wage increases, not across the board, but, as the government has done in the past, concentrated on careers whose pay profiles show the biggest discrepancy with respect to equivalent occupations in the private sector.4 We will also derive an alternative scenario (Scenario B) where the total wage bill of the federal government grows with nominal GDP and analyzes the implications of such a policy for debt sustainability. Regarding social security contribution rates of public sector workers, we assume a baseline scenario where the current contribution rates on both current and retired civil servants are unaltered. However, it would be highly desirable for fiscal and equity reasons to broaden the tax base by including both civil retirees and the military.

15. Regarding the evolution of the INSS benefit payments we are assuming that benefits are updated by past inflation and the average historical drift of around 4 percent per year, (the result of demographic factors and the increase in the reference real wages for the calculation of the retirement benefit). As such, we conservatively assume a deficit of the INSS of around 0.8 percent of GDP on average during the decade. In this scenario, increases in the minimum wage (and benefit payments) beyond past inflation are assumed to be coupled with the identification of additional permanent sources of revenue, or with a reduction of other permanent sources of expenditure, as mandated by the recently approved Fiscal Responsibility Law, in order to preserve fiscal transparency and long-term solvency of the public sector.

16. Finally, the OCC’s (Other Current and Capital expenditures) are subdivided between nondiscretionary spending—unemployment benefits, and other small items such as the LC/87 transfers that compensate the states for revenue losses derived from the exemption of state-level VAT (ICMS) on exports—and the discretionary component (subject to appropriation limits). The nondiscretionary component is expected to stay constant as a percentage of GDP after taking into account that LC/87 transfers to the states will peak at 0.31 percent of GDP in 2003 dropping to zero thereafter. A peculiarity that should be taken into account is that under the discretionary component, around 50 percent of the total is allocated to health care and education. Further, the recently approved Health Care Act demands that health related expenditures at the federal level should grow with nominal GDP, while the Constitution mandates that a fixed proportion of tax revenues should be spent on education. Therefore, these two important items can grow more, but no less, than nominal GDP, suggesting that there is no scope for potential real savings on them. For the remaining 50 percent of free discretionary spending, we will construct three alternative real growth scenarios, ranging from no real growth (baseline scenario), 2.5 percent real growth (Scenario A, Table 2.3), and the same growth of real GDP (Scenario B, Table 2.4). The 2.5 percent real growth per year represents an interesting proposition since, although below expected real GDP growth, it is approximately double the average population growth rate. As such, it allows for better and/or higher provision of public goods to the population while leading to a smooth decline of such expenditure categories as a percentage of GDP throughout the decade, since in the steady state real GDP is expanding at a faster rate.

Table 2.3.

Brazil: Scenario A

article image
Source: IMF staff estimates.

Real growth (drift) of 2.0 percent per year from 2002-2010.

Real growth (drift) of 4.0 percent per year from 2002-2010.

Real growth of free discretionary expenditure (excludes education and health care) of 2.5 percent per year from 2005-2010.

Assumes that the tax collections of the states and municipalities will remain constant at the 9.4 percent of GDP collected in 1999.

Table 2.4.

Brazil: Scenario B

article image
Source: IMF staff estimates

Real growth (drift) of real GDP from 2002-2010.

Real growth of 4.0 percent per year from 2002-2010.

Real growth of free discretionary expenditure (excludes education and health care) identical to real GDP (4 percent per year) from 2005-2010.

Assumes that the tax collections of the states and municipalities will remain constant at the 9.4 percent of GDP collected in 1999.

17. In all, for the year 2001, our scenario is consistent with the budget proposal submitted to congress. While congressional revisions to both revenues and expenditures are likely, we assume that the bottom line primary surplus of R$28.1 billion (around 2.3 percent of GDP) will be preserved. Further, for 2002 and 2003 we assume that the federal government Budget Guidelines Law (LDO) target of a primary surplus of 2.2 percent and 1.8 percent of GDP will be observed. As such, the 2002–03 level of OCCs is calculated as a residual, that is, given the scenario for revenues and other expenditure items, a value higher than the one assumed will not be consistent with the primary surplus target unless additional revenue sources are found or other expenditure items reduced. In 2004, following two years of very low execution in discretionary spending, we assume that discretionary OCCs will expand to the level observed during 2000 (4 percent of GDP) and grow thereafter in line with the assumptions underlying the different scenarios.

18. In terms of the execution of fiscal policy, given the underlying scenario assumptions, we anticipate that 2002 and 2003 will be difficult years regarding the level of discretionary OCC spending if the primary surpluses of the LDO are to be met, as concession revenues, CPMF collections, and the surplus of the PPE are bound to decrease from the levels expected to prevail in 2001 (net revenues drop 1.8 percentage point of GDP from 2001 to 2003 and do not recover in the following years, see Table 2.2).5 After 2004, under the baseline scenario, the primary surplus starts to recover gradually from 1.6 percent of GDP in 2004 to 2.25 percent of GDP in 2010 aided by the erosion of free discretionary spending and personnel expenses as a percentage of GDP, which would provide a gain of almost 1 percentage point of GDP in 2010 in comparison with 2004, more than offsetting the expected decline in revenue during the same period (see Table 2.2, Baseline Scenario). Under Scenario B—with both personnel expenses and OCCs growing in tandem with nominal GDP—the federal government primary surplus will decline from 1.8 percent of GDP in 2003 to 1.3 percent in 2005 and 1 percent of GDP during 2006–10 (see Scenario B, Table 2.4).

19. In order to close the baseline primary surplus scenario for the consolidated public sector, we will make a conjecture regarding the evolution of the primary surplus of the state owned nonfinancial enterprises and the subnational governments.

20. The primary surplus of the state-owned nonfinancial enterprises (SOEs) is expected to decline to more moderate-than-recently-observed levels from 2002 onwards reflecting the continuation of the privatization program (smaller set of public enterprises) and the return of international oil prices to more moderate levels (reducing the surplus of the state oil-refining monopoly Petrobrás). Also, the SOEs are expected to bolster investment levels in order to deal with the projected stronger real expansion of GDP and to pursue economically viable investment opportunities, particularly in the oil and gas sectors. Accordingly, in the steady-state we assume that the SOEs should be generating primary surpluses no larger than 0.1 percent of GDP per year (see Table 2.2).

21. Fiscal adjustment at the subnational level has been a reality since 1997. The states and municipalities have posted a combine primary deficit of 0.7 percent of GDP in 1997 and 0.2 percent of GDP in 1998, shifting to a surplus of over 0.2 percent of GDP in 1999, and around 0.5 percent of GDP in the 12-months to September 2000. For the period 2000–10 the states and municipalities’ primary surplus is assumed to remain consistently in the black but declining over time. The baseline scenario assumes that the covenants imbedded in the debt restructuring agreements signed with the Treasury by almost all the states and the biggest municipalities will continue to be honored. Further, the fiscal and debt targets contemplated by the Fiscal Responsibility Law and its ancillary legislation will force many subnational governments to pursue further fiscal adjustment (especially with regard to personnel expenses and the imbalances of many of the subnational level-sponsored retirement programs), limiting the likelihood of observing deficits at these levels of government. However, as own revenues and transfers expand with nominal GDP, we expect debt service payments to lose weight as a percentage of net revenues (the renegotiated debts should be amortized over a 30-year period in fixed monthly installments).6 Further, the subnational governments will also be affected by the Health Care Act and will be hard pressed to boost investment in basic infrastructure. As a result, the primary surplus is expected to decline gradually until 2005, reaching afterwards a primary surplus of around 0.1 percent of GDP per year on average until the end of the decade, although, a slightly better performance cannot be discarded if cautious fiscal management prevails (see Table 2.2).

22. Under the assumptions outlined above, the primary surplus of the consolidated public sector is expected to peak at 3.4 percent of GDP in the year 2000, declining to its lowest level in 2004 (1.9 percent of GDP), recovering gradually until the end of the decade to the equivale0nt of 2½ percent of GDP in 2010.7 Under the assumptions of Scenario B—real wages and OCCs growing in real terms with GDP—the primary surplus will decline continuously until 2006 (reaching 1.2 percent of GDP) stabilizing thereafter. In conclusion, a more indulgent fiscal stance (baseline scenario versus scenario B) is expected to cost over 5 percent of GDP in accumulated lost primary surpluses over seven years and even more in terms of public debt (around 6 percentage points of GDP due to accrued interest).

C. Macro economic Framework and Debt Dynamics

23. According to our medium- and long-term macro scenario, in the steady-state, real interest rates drop to 7.5 percent per year, real GDP grows by 4 percent, inflation declines gradually, stabilizing at 2.5 percent per year in 2005, and the foreign exchange rate stays on average on the Purchasing Power Parity curve for a foreign inflation rate of 1.5 percent per year. Further, the incorporation of liabilities in the debt statistics (FCVS and other liabilities) is expected to reach 0.75 percent of GDP in 2002. In the period 2003–10 recognition of liabilities net of privatization proceeds is expected to average 0.5 percent of GDP per year (see Table 2.5).8

Table 2.5:

Macroeconomic Framework and Baseline (Central Scenario) Nominal Deficit, Primary Deficit, and Net Public Debt

(%GDP 2000-2010)

article image
Source: Fund staff estimates.

Calculated on the basis of the capitalized overnight daily interest rate prevailing during the period.

Average of the annualized monthly overnight interest rate deflated by the annualized monthly inflation rate (measured by general price index; IGP-DI) during the period.

24. Under these assumptions the primary surplus needed to stabilize the debt to GDP ratio at the 48.5 percent expected to prevail at the end of 2002, hovers around 1.5 percent per year. Alternatively, under the traditional present value method, the constant steady state level of the primary surplus whose infinite horizon present value equals the 48.5 percent of GDP debt stock of 2002 (i.e. it pays off the debt stock) should lie in the range of 1.1 percent to 1.2 percent of GDP per year, assuming zero net incorporation of liabilities into the debt stock.9

25. Under our macroeconomic and baseline scenario assumptions the derived profile for the consolidated primary surplus implies that net public debt will decline smoothly, at the rate of around 0. 7 percentage point of GDP per year on average, reaching a ratio of around 43.3 percent of GDP in 2010, with the PSBR declining to slightly over 1 percent of GDP.10 Assuming that public investment remains broadly constant as a percentage of GDP, this implies an increase in public sector savings of over 3.5 percentage points of GDP in the year 2010 in comparison with the year 2000. This increase in public sector savings should support the necessary balance of payments adjustment, and free additional funds for private investment during the decade (see Section 10 on External Sustainability). The improvement in the macroeconomic conditions underpinning the baseline scenario for the period 2000–2010 is certainly predicated on a relatively strong fiscal stance (consistency requirement), since weaker-than-assumed fiscal performances will almost certainly lead to higher implicit probabilities of default and higher sovereign interest rate spreads.11

26. Further, not only is the net public debt to GDP ratio declining in the baseline scenario, but also other indicators of fiscal sustainability point to a moderate improving trend during the decade. The ratio of net public debt to total revenues of the central government improves slightly from 232 percent in the year 2000 to around 220 percent in 2010, after peaking at 243 percent in 2003 (see Table 2.2). In the same vein, the ratio of net public debt to total revenues of the general government (includes states and municipalities) improves from a peak of 165 percent in the year 2003 to 150 percent in 2010. While somewhat improved, this ratio will, nevertheless, still exceed the normative reference target ratio of 100 percent used by the Brazilian Treasury in the debt renegotiations with almost all the states and the largest municipalities. This points to the necessity of deepening the fiscal adjustment effort at all levels of government during this decade through reforms that improve the structural fiscal fundamentals that led to a subpar fiscal performance in the late 1990s. Finally, the indicator of the primary surplus effort of the central government (primary surplus as a share of gross revenues) does not point to a scenario where the share of revenues committed to the generation of the envisaged surpluses increases beyond what is expected to be observed in 2000–2002 (see Table 2.2).

27. Under Scenario A, where the discretionary component of OCCs grows in real terms by 2.5 percent per year, net debt to GDP will still decline, albeit at an uncomfortably slow pace, reaching 44.4 percent of GDP at the end of the decade. However, when both OCCs and personnel expenses grow with real GDP, (Scenario B) net debt stops falling around 2004 (notice that until 2003 strong primaries are guaranteed by the LDO targets) and it enters an unsustainable path (see Table 2.4). In conclusion, the debt dynamics projection point to the fact that while there might be some room for a moderate real increase in OCCs (still below real GDP growth) a more expansionary fiscal stance might rekindle fears of unsustainable debt dynamics and would leave basically no cushion for additional recognition of liabilities beyond the amount envisaged in the baseline scenario or for a deterioration of the macroeconomic picture that leads to higher real interest rates, lower real GDP growth, and eventually a more depreciated path for the nominal exchange rate. For this reason, Scenario B is a riskier proposition and points to the necessity of control over the real growth of the wage bill and discretionary OCCs during the decade and the need to push forward the structural fiscal reform agenda. Under Scenario B, since the fiscal effort of the central government decreases substantially after 2002 (as measured by the primary surplus share of gross revenues), all the other indicators of fiscal sustainability also point to a scenario of gradual deterioration of the fiscal stance, particularly after 2005 (see Table 2.4).

D. Stress Tests

28. In this section we will stress test the central scenario and the implied net public debt ratio path by varying the level of real interest rates, the nominal exchange rate, real GDP growth, and the real growth rate of OCCs and personnel expenses (see Tables 2.6 and 2.7, and Figures 2.2 and 2.3).

Table 2.6.

Brazil: Sensitivity Analysis of the Baseline (Central Scenario) to Real Interest Rates and Real GDP Growth.

article image
Source: IMF staff estimates.
Table 2.7.

Brazil: Scenario A and B, and Sensitivity Analysis of the Baseline to the Foreign Exchange Rate.

article image
Source: IMF staff estimates.
Figure 2.2.
Figure 2.2.

Brazil: Net Public Debt, Primary Surplus, and Nominal Balance

(PSBR)

Citation: IMF Staff Country Reports 2001, 010; 10.5089/9781451805901.002.A002

Source: Fund staff estimates
Figure 2.3.
Figure 2.3.

Brazil: Net Public Debt Sensitivity Analysis; 2000-2010

Citation: IMF Staff Country Reports 2001, 010; 10.5089/9781451805901.002.A002

Source: Fund staff estimates.

29. Net public debt shows heightened sensibility to both real GDP growth and the level real interest rates (see Table 2.6 and Figure 2.3).12 If the average real GDP growth between 2000 to 2010 turns out to be 1 percentage point lower, the path of net public debt will then show an increasing trend after 2003 and will reach a high level (55 percent of GDP) by the end of the decade. The same is true for an increase of average real interest rates of around 200 basis points from 2001–2010 (see table). The stress tests performed show that the critical points beyond which unstable debt dynamics seem to kick in are around 75 basis points of lower average real GDP growth or 130 basis points of higher real interest rates.

30. Further, the debt dynamics show that the debt to GDP ratio could support well a nominal exchange rate depreciation shock of over 10 percent—ceteris paribus, the declining path of debt to GDP is not derailed by such a shock (see Table 2.7).

31. In the case of a shock to real GDP growth, we have explicitly modeled the impact on the primary surplus. However, in the case of higher real interest rates, we have not considered the potentially negative effect this might have over real growth and the primary surplus. If such transmission mechanism were to be explicitly modeled, it would have magnified the potentiality destabilizing effect of monetary tightening on the debt to GDP ratio.

E. Conclusion and Policy Implications

32. Given that the baseline scenario cannot sustain high levels of stress both to real GDP growth or the level of real interest rateseven in a baseline scenario in which discretionary expenditure shows no real growth throughout the periodthe exercise performed in this section indicates that it is imperative that the fiscal effort be sustained during the decade and at a level that allows for a reasonable decline of debt to GDP. That is, fiscal dividends arising from structural fiscal reforms and further reform of the social security system, particularly for public sector workers, are critical to allay concerns regarding Brazil’s vulnerability to adverse macroeconomic shocks, be it domestic or external, and to avoid the need to resort to abrupt fiscal retrenchments in case such shocks materialize—since hasty fiscal adjustment programs are usually more disruptive and tend to be suboptimal in terms of welfare than more gradual ex-ante fiscal consolidation strategies. In addition, staying the course of fiscal discipline considerably improves the chances of attaining an investment-grade sovereign rating sooner, which would further assist the fiscal and external adjustment processes since it would lead to lower domestic interest rates and lower external financing costs.

33. The baseline scenario implied debt dynamics profile also shows that it is imperative to honor the spirit of the recently approved Fiscal Responsibility Law and avoid the creation of permanent spending items without the identification of the concomitant permanent sources of financing. Prudence and responsibility in the setting of the main drivers of automatic spending such as the level of the minimum wage and the real increase in social security benefits13 and public sector wages are in this regard critical since they have a permanent nature and can potentially crowd out other meritorious and needed social spending items. Reduction of the earmarking of federal revenues would also aide in the management of fiscal policy.

34. The high sensitivity of the debt path to GDP growth also points the importance of pushing forward with structural macroeconomic and microeconomic reforms that can improve the noninflationary growth potential of the economy in the long run and raise the national savings rate. In that regard, a comprehensive tax reform designed to stabilize the tax burden, while reducing microeconomic distortions and spreading the tax burden in an equitable way among different economic agents/sectors, would not only improve the competitiveness of the domestic production but also have a potential indirect positive fiscal impact by enlarging the taxable base.

35. With regard to the intertemporal tradeoffs, the stronger the fiscal effort earlier in the decade, the greater the present discounted value of the dividends to be extracted from fiscal adjustment in terms of further room to increase social spending in the future and the concomitant improvement in social indicators, since savings (interest bill) will be higher earlier in the decade when real interest rates exceed the steady-state neutral level but also because promises of fiscal adjustment in the future are discounted by markets leading to higher contemporaneous interest rate risk premiums than otherwise.

36. To preserve and institutionalize the needed strong fiscal stance after 2003, and to assist in the formation of market expectations with regard to the path of fiscal policy, it would be desirable that, at least the central government announces primary fiscal and debt targets for the subsequent three years, mirroring the current practice with regard to the inflation targets. Currently, the Budget Guidelines Law sets the target for the primary surplus for the following year and indicative targets for the subsequent two, while the Fiscal Responsibility Law requests that the President should propose to the senate the indebtedness ceilings for all levels of government. However, it would be desirable to go beyond these requisites and set in the law the requirement for a floor on the primary surplus and a specific provision envisaging a decline of net public debt, on average, for the next three years.

37. It is also critical that, at the subnational level the fiscal adjustment effort undertaken in recent years is not brought to a halt. Further fiscal consolidation will not only release additional resources for needed investment and social expenditure by these government levels but will also help reduce the current federal government debt default-risk exposure to other levels of government. Therefore, in the spirit of the recently approved Fiscal Responsibility Law, the central government should strive to enforce the debt restructuring agreements signed with state and local governments, as done in the past, and encourage subnational governments to continue down the road of fiscal responsibility. Concurrently, the federal government should vigorously pursue the remaining structural fiscal reform agenda, continuously assess the debt-stock impact of the incorporation of contingencies (explicit and implicit contingent liabilities), and be ready to proactively respond, with fiscal instruments, to any surprises that might deteriorate the government’s creditworthiness.

References

  • Blanchard, Olivier, J-C Chouraqui, R.P. Hagemann, and N. Sartor, 1990, “The Sustainability of Fiscal Policy: New Answers to an Old Question,” OECD Economic Studies, Vol. 15, pp. 7 –36.

    • Search Google Scholar
    • Export Citation
  • Bogdanski, Joel, A. Antonio Tombini, and Sérgio Werlang, 2000, “Implementing Inflation Targeting in Brazil,” paper presented at High Level Seminar on Inflation Targeting, International Monetary Fund, Washington DC, March 20–20, (http://www.imf.org/external/pubs/ft/seminar/2000/targets/index.htm).

    • Search Google Scholar
    • Export Citation
  • Bohn, Henning, 1991, “Budget Balance through Revenue or Spending Adjustments? Some Historical Evidence for the United States,” Journal of Monetary Economics; Vol. 27 No. 3 (June), pp. 33359.

    • Search Google Scholar
    • Export Citation
  • Carvalho, J. Carlos, 2000, “Análise da Tendência de Longo Prazo das Finanças Públicas Brasileiras,” IPEA. Brasília.

  • Chalk, Nigel and Richard Hemming, 2000, “Assessing Fiscal Sustainability in Theory and Practice,” IMF Working Paper, WP/00/81.

  • Cochrane, John H., 1998b, “Long-term Debt and Optimal Policy in the Fiscal Theory of the Price Level,” NBER Working Paper No. W6771.

    • Search Google Scholar
    • Export Citation
  • Fraga, A., 2000, “Monetary Policy During the Transition to a Floating Exchange Rate,” Finance and Development, Vol. 37, No. 1 (March).

    • Search Google Scholar
    • Export Citation
  • Giambiagi, Fabio, 2000a, “Urna Proposta de Política Fiscal: O Orçamento Base Zero,” manuscript.

  • Giambiagi, Fabio, 2000b, “A Agenda Fiscal de Médio Prazo: Além do FMI,” manuscript.

  • Giambiagi, Fabio, 2000c, “The Brazilian Crisis of 1998–1999: Origins and Consequences,” BNDES, Textos para Discussão No. 77.

  • Giambiagi, Fabio, 2000d, “A Política Fiscal Depois de 2002: Algumas Simulações,” manuscript.

  • Giambiagi, Fabio, and Ana C. Além, 1999, Finanças Públicas—Teoría e Prática no Brasil. Editora Campus.

  • Guedes, Paulo, and José Carlos Carvalho, 1999, “The Point of No Return: A Política Fiscal na Beira do Abismo,” Insight Intelegência, No. 6, Feb/Apr.

    • Search Google Scholar
    • Export Citation
  • Hakkio, Craig, and Mark Rush, 1991, “Is the Budget Deficit ‘Too Large?’” Economic Inquiry; Vol. 29, No. 3, July 1991, pp. 42945.

    • Search Google Scholar
    • Export Citation
  • Hamilton, James D., and Marjorie A. Flavin, 1986, “On the Limitations of Government Borrowing: A Framework for Empirical Testing,” American Economic Review; Vol. 76, No. 4, (September), pp. 80819.

    • Search Google Scholar
    • Export Citation
  • Neri, Marcelo, and Fabio Giambiagi, 2000, “Previdência Social e Salário Mínimo: O Que Se Pode Fazer, Rspeitando a Testrição Orçamentária.

    • Search Google Scholar
    • Export Citation
  • Ramos, Alberto, and Evan Tanner, 2000, “Fiscal Sustainability and Monetary versus Fiscal Dominance: Evidence from Brazil, 1991–2000,” IMF Working Paper (forthcoming) (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Rocha, Fabiana, 1997, “Long-Run Limits on the Brazilian Government Debt,” Revista Brasileira de Economia, Vol. 51, No. 4 (October/December); pp. 447470.

    • Search Google Scholar
    • Export Citation
  • Rossi, José, 1997, “A Solvência da dívida: Testes para o Brasil” Working Paper No. 493, IPEA, Rio de Janiero (July).

  • Sargent, Thomas J., and Neil Wallace, 1981, “Some Unpleasant Monetarist Arithmetic” Federal Reserve Bank of Minneapolis Quarterly Review Fall 1981 (Reprinted in Rational Expectations and Inflation, New York: Harper Collins College Publishers, 1993).

    • Search Google Scholar
    • Export Citation
  • Sargent, Thomas J., and Neil Wallace, 1998, “Econometric Implications of the Government Budget Constraint,” Journal of Econometrics,”; Vol. 83, No. 1–2 (March–April), pages 919.

    • Search Google Scholar
    • Export Citation
  • Velloso, Raul, 1997, “Urna Proposta para Acelerar o Ajuste Fiscal,” in Brasil: Desafios de um País em Transformação; José Olimpio Editora, 1997.