This paper focuses on the Mexican economy, its output fluctuations, and analyzes the country's external competitiveness in the context of medium-term external current account sustainability. The study also presents different measures of the fiscal deficit, including the traditional budget balance, and reviews measures on the fiscal stance and fiscal impulse. It examines the evolving relationship between inflation and the exchange rate, using fixed and time-varying coefficient models. The paper describes the methodology, provides the rationale for estimating time-varying parameter models, and presents the estimation results.

Abstract

This paper focuses on the Mexican economy, its output fluctuations, and analyzes the country's external competitiveness in the context of medium-term external current account sustainability. The study also presents different measures of the fiscal deficit, including the traditional budget balance, and reviews measures on the fiscal stance and fiscal impulse. It examines the evolving relationship between inflation and the exchange rate, using fixed and time-varying coefficient models. The paper describes the methodology, provides the rationale for estimating time-varying parameter models, and presents the estimation results.

III. Alternative Measures of the Fiscal Deficit and Fiscal Sustainability1

A. Introduction

1. The Mexican authorities recently began to publish a more comprehensive definition of the overall fiscal balance. The latter definition goes beyond the traditional (budget) measure by including some quasi-fiscal operations that had been excluded previously from the narrower definition.2 Such operations have been substantial in Mexico, being mainly tied to the carrying costs of government debt instruments issued during the banking sector crisis of the mid-1990s.

2. The broader measure of the fiscal balance yields a deficit that is some 3 percentage points of GDP higher than the official statistics. While the broader measure shows a higher deficit, it also shows that the fiscal consolidation that has taken place since the mid-1990s has been even larger than revealed by official statistics. In addition, the government has expressed its commitment to sound fiscal policy and presented to congress a tax reform package. The reform aims at reducing fiscal dependence on oil revenues by increasing tax revenues by 1.7 percentage points of GDP a year.3 The latter would be complemented by administrative measures that the authorities estimate could yield an additional 1.5 percentage points of GDP over the next five years. The estimates presented in this chapter indicate that the authorities’ plans, if approved unchanged, would lead to a substantial reduction in the debt/GDP ratio and would thus make an important contribution to fiscal sustainability, by reducing dependence on oil revenues, while allowing for a moderate increase in social spending.

3. This chapter presents different measures of the fiscal deficit in Mexico, including the traditional budget balance, the adjusted balance now being published by the authorities, and the augmented fiscal balance that has been used by Fund staff, and which is similar to the authorities’ adjusted balance. While the more comprehensive definitions provide a better indication of the fiscal position to assess long-term sustainability, their high interest content makes it difficult to assess the impact of fiscal policy on aggregate demand. Therefore, it is necessary to complement the more comprehensive definitions with other measures of the fiscal position, such as the primary balance and the operational balance. Measures of the fiscal stance and fiscal impulse are also reviewed in this chapter.

B. Alternative Measures of the Fiscal Balance

4. The first step in deciding on how to measure the fiscal balance is to define the appropriate coverage of the public sector. In Mexico, official statistics and Fund programs have included the operations of the federal government, nonfinancial public enterprises, and extra budgetary funds, but have excluded the operations of state and local governments and those of public financial entities. State and local governments have been excluded, largely because of long lags in the production of their fiscal statistics. This omission however, is not significant as sub-national governments have limited revenue and borrowing capacities and rely on revenue sharing and federal government transfers for their operations.4 Public financial entities on the other hand, have traditionally engaged in large quasi-fiscal operations. Until recently, the information on the most important of these operations was reported separately—especially those associated with the banking crisis of the mid-1990s—but were not included in the official fiscal statistics.5

5. The traditional fiscal balance in Mexico is reported on a cash basis, but it deviates from the 1986 Government Finance Statistics (GFS) Manual in four main aspects: 1) revenues from privatization and unrealized capital gains on international reserves and debt buy-back operations are recorded as income; 2) the inflation component of indexed bonds is not included; 3) spending on some public investment projects is recorded when the financial obligations are serviced, not when the investment is made;6 and, 4) the costs of quasi-fiscal operations are not included.7

6. The adjusted fiscal balance now being published by the authorities in parallel to the budget measure seeks to correct these shortcomings. The adjusted measure excludes the revenue from privatization and unrealized capital gains and includes the inflation component of indexed bonds. It also includes the net costs of the deferred investment projects (PIDTREGAS), the costs of debtor-support programs (including toll-road concessions) and the financial requirements of IPAB and of development banks.8

7. The augmented fiscal balance used by the staff, incorporates the above adjustments and, in addition, it includes the nonrecurrent costs of the quasi-fiscal operations, namely the value of the debt issued for bank restructuring and debtor-support programs, net of asset recoveries. This measure better reflects the changes in the stock of domestic debt during the period.9

8. Given the relatively high interest burden (current and imputed) in Mexico (6 percent of GDP in 2000) and the relatively high rates of inflation seen recently, measures such as the operational deficit, which excludes the inflation component of domestic interest payments, and the primary balance, which measures the fiscal balance net of interest receipts and payments, are also needed to assess the fiscal stance. Finally, in the case of Mexico, it is important to look at the evolution of the non-oil fiscal balance, as public sector oil revenues make up over 25 percent of total revenues.

9. The different measures of the fiscal balance are presented in Table 1. The traditional fiscal deficit averaged 0.5 percent of GDP during the 1990s and has been close to 1 percent of GDP during the last four years. On the other hand, the adjusted balance averaged -3.5 percent of GDP during the last decade and over 5 percent of GDP over the last four years (Figure l).10 The traditional balance underestimates the magnitude of the fiscal deficit; thus, reliance on such a measure does not provide a clear indication of the need for further fiscal adjustment to improve the sustainability of public sector debt in Mexico over the medium term.

Figure 1.
Figure 1.

Mexico: Public Sector Borrowing Requirements Under Alternative Definitions

(In percent of GDP)

Citation: IMF Staff Country Reports 2001, 191; 10.5089/9781451825572.002.A003

Table 1.

Mexico: Alternative Measures of the Fiscal Deficit, 1990–2000

(In percent of GDP)

article image
Sources: Secretarial of Finance and Public Credit; and Fund staff estimates.

Authorities’ definition. Includes federal government and public enterprises.

Treats transfers to IPAB as interest expenditure.

Excludes oil balance: total public sector oil revenue minus PEMEX expenditure.

Adjusted to exclude privatization revenue, unrealized valuation gains and including net costs of PIDIREGAS. Includes inflation adjustment to indexed bonds, imputed interest on bank-restructuring and debtor-support programs and financial requirements of development banks.

Treats transfers to IPAB and financial requirements of development banks as interest expenditure.

Augmented to include the net nonrecurrent capital costs of bank-restructuring and debtor-support programs.

10. Given that the interest component of the adjusted measure is large and inflation was quite high during 1995-98, the path of the adjusted balance is influenced by developments in the actual and imputed interest bill. The adjusted primary surplus deteriorated from 3.5 percent of GDP in 1996 to 0.4 percent of GDP in 1998, in large part, due to increased use of PIDIREGAS projects. Since then, it has recovered to 1.5 percent of GDP, but it still remains substantially below the 1990-95 average (4.2 percent of GDP). Furthermore, the adjusted non-oil primary balance moved from a modest surplus in the early 1990s to a deficit of almost 3 percent of GDP in 2000. This trend reflects the fact that recent declines in the adjusted deficit are largely due to the decline in interest costs—following the reduction in inflation and in nominal interest rates—but that the underlying fiscal position needs to be strengthened further. The recent trend also highlights Mexico’s continued dependence on oil export revenues and underscores the need for undertaking a tax reform, as the authorities have proposed.

C. Fiscal Impulse in the 1990s

11. One way to assess the initial impact of fiscal policy on aggregate demand is to estimate the fiscal stance and impulse (Table 2). The fiscal stance is defined as the difference between the cyclically neutral balance and the actual fiscal balance, and thus provides an estimate of the initial amount of expansionary or contractionary pressure placed by the budget on aggregate demand (Schinasi, et al. (1991)). To estimate the cyclically neutral balance, we assume a unitary elasticity of government revenues with respect to actual GDP and that government expenditures grow with potential GDP.11 The fiscal impulse is measured as the yearly change in the fiscal stance and captures the effect on the budget of changes in fiscal policy with respect to the cyclically neutral position.

Table 2.

Mexico: Summary Impulse Measures

(In percent of GDP)

article image
Source: Fund staff estimates.

Excludes privatization revenue BOM profits, unrealized capital gains on debt buy-back operations, and oil-export revenue.

Treats transfers to IPAB as interest expenditure.

In addition to revenue adjustments as in footnote 1, expenditures include the inflation component of indexed bonds, imputed interest on bank-restructuring and debtor-support liabilities, and the net cost of PIDIREGAS. It does not include the financial requirements of development banks nor the nonrecurrent costs from banking sector restructuring operations.

12. To better gauge the effect of fiscal policies on the budget, revenues coming from oil exports are excluded, since these resources do not affect domestic absorption, nor do changes in these revenues typically reflect changes in government policies. The assessment of the fiscal stance in the second half of the 1990s is quite different depending on the definition of the fiscal balance used.12 For example, while the traditional definition yields a negative impulse of 0.7 percentage point of GDP in 1995, the adjusted measure provides a positive impulse of 2 percentage points of GDP, reflecting in large part the costs of the banking crisis and the effect of higher inflation on indexed bonds. In 2000, the traditional definition yields a positive impulse while the adjusted measure results in a negative one. In this latter case the difference is attributed mainly to the effect of lower interest rates (from 21 percent in 1999 to 15 percent 2000) on the cost of the bank-restructuring and debtor-support liabilities.

13. In order to have a more direct measure of discretionary fiscal policy, the primary impulse (excluding interest expenditures) is examined.13 The primary impulse does not follow the same path as the total impulse. In 1995, while the adjusted total impulse was 2 percent of GDP, the adjusted primary impulse was negative 2.6 percent of GDP. On the other hand, the path followed by the primary impulse under the traditional and adjusted methodologies is very similar, providing a more consistent description of the role played by fiscal policy during the period.14 With the exception of 1999, fiscal policy provided a positive impulse every year since 1996.

14. As with the adjusted fiscal deficit, the adjusted total impulse is influenced by the large nominal interest component and thus the adjusted primary impulse provides a better indicator of the fiscal stimulus, since the private sector is unlikely to consider the inflation component of domestic interest payments as a transfer from the public sector. On the other hand, by excluding all the costs of the bank-restructuring and debtor-support operations, the adjusted primary impulse is likely to underestimate the stimulus provided by the public sector to the extent that some of these operations contain a transfer component which is likely to be treated as an increase in net worth by the beneficiaries of these programs.15

D. Public Debt and Fiscal Sustainability

15. The adjustments to the traditional balance also imply adjustments to the coverage of the public sector debt. Traditionally in Mexico, public sector debt refers to financial obligations of the federal government, but excludes the liabilities arising from the bank restructuring and debtor-support programs (including toll-road concessions) and those associated with the PIDIREGAS projects. The augmented public sector debt includes these liabilities (Table 3).

Table 3.

Mexico: Total Public Sector Debt

(In percent of GDP)

article image
Sources: Secretariat of Finance and Public Credit; and Fund staff estimates.

16. At end-2000, gross (augmented) public debt was equivalent to 47.5 percent of GDP, and, while low by OECD standards, it is relatively high when compared with that of the larger countries in Latin America.16 Slightly more than one-third of the total adjusted debt is denominated in foreign currency (including liabilities associated with PIDIREGAS projects); another third is domestic debt issued under bank-restructuring and debtor-support programs, and the remainder is domestic market-issued debt (see Table 3). The federal government’s external debt/GDP ratio declined from 36 percent in 1995 to 15 percent in 2000. To a large extent, this decline has been due to an explicit effort by the authorities to substitute domestic obligations for external ones, but also, it has been the by-product of strong economic performance and of the currency appreciation in recent years.17 18

17. Given that the budget did not incorporate the issuance of bonds for bank restructuring and other debtor-support programs, the impact of the debt issued for this purpose was not fully incorporated into the traditional fiscal accounts. In addition, since a large portion of the initial bank-restructuring debt was issued in the form of zero-coupon bonds, the servicing of this debt has not placed pressure on the cash operations of the government. Nonetheless, the interest accruing on this debt (as recorded in the adjusted deficit measure) increases the stock of the debt and will eventually need to be financed explicitly. Recently, the authorities have made efforts to improve the profile of IPAB debt via liability management operations and, in 2000, began to issue market instruments—carrying market interest rates and coupons. Also, the authorities have announced that they intend to pay down the real interest accruing on this stock of debt, such that the stock will gradually decline as a share of GDP.

18. A standard debt-sustainability exercise indicates that under current policies (i.e., an adjusted primary surplus of around 1.5 percent of GDP) and with real interest rates on public debt of 6 percent (the average real interest rate on domestic debt instruments was 6.8 percent in 2000) a stable debt/GDP ratio would be obtained with annual GDP growth of 4 percent.18 Table 4 provides a summary grid of the debt/GDP ratio and fiscal balance resulting from current policies, but under different combinations of assumptions on growth and average real interest rates on public debt. With average real interest rates of 4-6 percent, the gross debt/GDP ratio would decline to 34-44 percent, if growth were above 4 percent per annum. With lower growth assumptions and/or higher real interest rates, the current adjusted primary balance would be insufficient to stabilize the debt/GDP ratio.

Table 4.

Mexico: Fiscal Balance and Debt/GDP Ratio, 2006, Sensitivity to GDP Growth and Real Interest Rates

(In percent of GDP, unless otherwise indicated)

article image
Source: Fund staff estimates.

19. Only under sustained strong growth and relatively low real interest rate conditions, would the current levels of the adjusted primary balance be sufficient to produce important reductions in the debt/GDP ratio, over the medium term. In addition, if one takes account of the relatively high fiscal dependence on oil revenues,19 the current level of the adjusted primary balance seems insufficient to ensure a continued reduction in the debt/GDP ratio. Therefore, an overarching objective of fiscal policy in Mexico in the near-term should be to improve the adjusted primary balance while reducing the dependence on oil revenues.

20. Approval of the tax reform proposal—as initially presented by the authorities—would achieve this objective, provided that all the revenue gains materialize and if most of these proceeds are saved.20 The authorities estimate that the reform could have a net yield of 1.2 percent of GDP and that, in addition, through administrative measures, the authorities could collect an additional 0.3 percentage point a year over the next five years. Under these circumstances, the increase in the primary balance would lead to a much faster decline in the debt/GDP ratio than presented in Table 4, for any combination of interest and growth rates. If, on the other hand, the tax reform is not passed, the fiscal deficit could remain above 4 percent of GDP through 2006 while the gross (augmented) debt/GDP ratio could reach 50 percent by the end of the period (Figure 2 and Appendix Figure).1 2

Figure 2.
Figure 2.

Mexico: Cross (Augmonted) Debt Under Alternative reform Scenarios

(In percent of GDP)

Citation: IMF Staff Country Reports 2001, 191; 10.5089/9781451825572.002.A003

E. Conclusion

21. The traditional budget balance used in Mexico underestimated the magnitude of the fiscal deficit during the last decade. The recent publication by the authorities of a more comprehensive measure of the fiscal balance is welcome and represents a major improvement in fiscal transparency. The adjusted fiscal balance shows that there has been an important improvement in the fiscal position since the mid-1990s, but that further improvement is still needed to reduce the debt/GDP ratio over the medium term. In particular, improving the non-oil primary fiscal balance, as intended by the authorities through the proposed tax reform, could lead to a considerable fiscal consolidation, while helping to reduce fiscal vulnerability to oil price volatility.

APPENDIX

uapp01fig01

Public Sector Borrowing Requirements Under Alternative Reform Scenarios

(In percent of GDP)

Citation: IMF Staff Country Reports 2001, 191; 10.5089/9781451825572.002.A003

Mexico: Adustment to Budget Balance for Fiacal Impulse Measures

article image
Sources: Secretariat of Finance and Public Credit; and Fund staff es

Excludes interest payments and transfers to IPAB which are treated as interest expenditure.

Excludes financial requirements of development banks and non-recurrent costs of bank restructuring programs.

Excludes interest payments, inflation adjustment of indexed bonds and transfers to IPAB which are treated as interest expenditure.

List of References

  • Schinasi, Gary J, and Mark S. Lutz, 1991, “Fiscal Impulse,” IMF Working Paper 91/91 (Washington: International Monetary Fund).

  • Spaveta, Luigi, 1986, “The Growth of Public Debt: Sustainability, Fiscal Rules, and Monetary Rules,” IMF Working Paper 86/8, (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
1

This chapter was prepared by Juan Pablo Córdoba.

2

Information on the most important of these operations was published separately.

3

The authorities estimate a net yield from the reform of 1.2 percentage point of GDP as the federal government has to transfer approximately 23.5 percent of tax revenues to the states. An additional 0.15 percentage point will be spent on programs to compensate the poor for the elimination of exemptions and the zero-rating on domestic sales of basic goods.

4

States and municipalities collect less than 5 percent of total tax revenues.

5

The budget incorporates transfers to the Bank Savings Protection Institute (IPAB)—that administers the bank restructuring and debtor-support programs—to cover a portion of the real interest cost of these liabilities. Fund documents report an adjusted deficit measure, which includes an estimate of these costs.

6

This refers to investment projects that allow for deferred recording in the fiscal accounts (PIDIREGAS).

7

There are four types of quasi-fiscal operations that have been identified: 1) those related to the bank restructuring operations which are the largest component; 2) those related to debtor-support programs; 3) the liabilities assumed by the government from the failure of toll-road concessions; and 4) net financial requirements of development banks. At end-2000, total liabilities from these sources amounted to 17 percent of GDP.

8

The adjusted balance presented in this chapter does not coincide with that published by the authorities because the latter includes cash reserves made by the public pension fund (IMSS) in charge of administering the old pay-as-you-go system and the public institution (FOVISSTE) in charge of financing housing for public employees. Also, before 2000, there are differences in the estimation of the imputed cost of bank-restructuring and debtor-support operations. Discussions are underway with the authorities to reconcile the historical series and to agree on the appropriate treatment of these reserves.

9

The authorities’ adjusted balance includes asset recoveries since 2000, but it does not include the capital costs of bank restructuring and debtor-support programs before that.

10

The augmented balance averaged almost 5 percent of GDP during the 1990s and close to 7.5 percent of GDP since 1996, although it declined from 7 percent of GDP in 1999 to 4 percent in 2000.

11

The estimates take 1993 as the base year and assume that potential GDP grows at an annual rate of 4.2 percent in real terms. See Chapter 1 for estimates of potential GDP for Mexico.

12

In both measures of the fiscal impulse, the definition of revenues is adjusted to exclude privatization revenue as well as unrealized capital gains (from central bank profits and debt buy-back operations). The difference between the two definitions is due to the inclusion in the adjusted measure of the inflation component of indexed bonds, the imputed interest on bank-restructuring and debtor-support liabilities, and the net cost of PIDIREGAS projects. See Appendix Table for a detailed presentation of the adjustments made to estimate the fiscal impulse.

13

The imputed carrying costs of bank-restructuring and debtor-support programs are treated as interest expenditure.

14

In 1998, the traditional and adjusted primary impulse measures differ mainly due to the PIDIREGAS projects whose net cost increases to 0.9 percentage point of GDP (from 0.2 percentage point in 1997). This change is not captured by the traditional primary impulse. Therefore, while the budget was showing a relatively small positive impulse in 1998, fiscal policy provided a larger stimulus incorporating the effect of PIDIREGAS projects.

15

However, it is very difficult to estimate the grant component of these programs or its likely effect on consumer behavior. Similarly, the operations of development banks may also affect aggregate demand, but these are not captured in the measures presented above.

16

This is equivalent to a net (augmented) debt/GDP of 41.7 percent of GDP as of end-2000. The authorities report net (augmented) debt/GDP at 40.2 percent; the difference is due to the authorities’ use of fourth quarter (annualized) GDP to deflate the series as opposed to the average annual GDP used by the staff.

17

While external debt has declined, domestic market-issued debt has increased since 1995 (from 9 to 13 percent of GDP). Although this move reduces the dependence on external financing and reduces the exchange rate risk of government debt, the strategy is not without risks, given the reduced domestic market for fixed-rate, medium- and long-term securities. In 2000, the authorities began issuing longer term fixed-rate instruments and they announced the issuance of a 10-year fixed rate bond in mid-2001 (see Appendix Box 2 in accompanying Staff Report).

18

Of the 21 percentage points of GDP decline in the external debt/GDP ratio between 1995 and 2000, 7 percentage points may be directly attributed to the real appreciation of the peso. That is, assuming a constant real exchange rate from end-1995, the debt/GDP ratio would have declined to 21.6 percent of GDP.

19

The methodology follows Spaveta (1986).

20

With an average oil price in 2000 of US$24.5 per barrel, oil revenues were 6 percent of GDP. Every US$1 change in oil prices represents roughly 0.1 percentage point of GDP in oil export revenue.

21

The main elements of the reform are: (1) elimination of most VAT exemptions and zero-rating on domestic sales—including for books, magazines, processed and unprocessed foods, equipment for agriculture and fishing, and land transportation; (2) limit the application of a reduced 10-percent VAT rate in border regions; (3) elimination of special income tax regimes for different sectors and regions; (4) reduction of the top marginal personal income tax rate from 40 to 32 percent; (5) elimination of the 5 percent withholding of distributed profits; and (6) unification of the corporate income tax rate with the top marginal rate for individuals.

22

Three simulations were conducted: the baseline scenario that uses the assumptions in the medium-term macroeconomic framework; a middle scenario, where only a partial version of the tax reform is passed (net yield of the reform is 0.6 percent of GDP and administrative measures only yield 0.1 percentage point of GDP a year) and this leads to slightly lower growth and higher interest rates; and a no-reform scenario where real interest rates would revert to 7 percent and GDP growth would remain around 4 percent a year.

23

The simulations presented in this chapter do not consider the possible negative effects of increases in government outlays due to the pension reform or to other contingent liabilities (for example, from conditional PIDIREGAS). According to the 1997 reform, the government guarantees that transition workers will receive a private pension at least as high as the one they would have received from the pay-as-you-go system, which could lead to considerable expenditure pressures.

Mexico: Selected Issues
Author: International Monetary Fund
  • View in gallery

    Mexico: Public Sector Borrowing Requirements Under Alternative Definitions

    (In percent of GDP)

  • View in gallery

    Mexico: Cross (Augmonted) Debt Under Alternative reform Scenarios

    (In percent of GDP)

  • View in gallery

    Public Sector Borrowing Requirements Under Alternative Reform Scenarios

    (In percent of GDP)