Journal Issue

Mexico: Selected Issues

International Monetary Fund
Published Date:
August 2011
  • ShareShare
Show Summary Details

III. Long-Term Fiscal Challenges in Mexico1

Mexico’s fiscal position is solid. Fiscal credibility, underpinned by prudent fiscal management and a strong fiscal framework, permitted a countercyclical fiscal response during the 2009 global crisis. The envisaged fiscal consolidation during 2010—12 will return the structural fiscal stance to pre-crisis levels. Mexico is expected to return to the budget balance specified under the fiscal rule in 2012 and maintain it thereafter, ensuring a stable public debt path. However, Mexico faces significant long-term fiscal challenges that are not easily appreciated in a standard medium-term fiscal analysis. Oil revenues may decline by 4 percent of GDP over the next two decades due to stagnation in production, while age-related spending could increase over 3 percent of GDP in response to demographic and other factors. The combination of age-related spending pressures and declining oil revenues implies that sustaining prudent levels of public debt would require a large (non-oil) revenue mobilization effort and an expenditure reform strategy. Addressing effectively such fiscal challenge would require early action since corrective measures may have long implementation lags.

A. Medium-Term Fiscal Outlook

1. To assess Mexico’s medium-term fiscal outlook this section lays out a fiscal forecast for the period 2012-16 anchored in the balance-budget rule and under the following macroeconomic assumptions:

Table 1.Mexico: Macroeconomic Assumptions 2012-16
Real GDP growth (percent)
GDP deflator growth (percent)
Nominal exchange rate (average)12.312.512.612.812.9
Interest rate (public debt; percent)
Oil price world markets (US dollars per barrel)1051021009999
Oil production (thousand barrels per day)2,5502,5502,5502,5502,550
Oil exports (thousand barrels per day)1,3471,3471,3471,3471,347
Oil derivatives consumption growth (percent)
Oil derivatives domestic price growth (percent)
Source: IMF Staff estimates.
Source: IMF Staff estimates.

2. The projected real GDP growth path is consistent with potential output growth estimates discussed in Chapter 1. Oil prices projections are obtained from the WEO forecast,2 oil production projections come from SHCP (2010), and oil exports are projected assuming that the ratio between oil exports and oil production observed in 2010 is maintained during 2012—16. Oil derivatives consumption growth is assumed to be equal to real GDP growth and the domestic price of oil derivatives is projected to increase 12 percent during 2012—13 to reach at least international prices. In addition, we assume that 5 percent of projected duties on hydrocarbons are saved and transferred to oil stabilization funds.3

3. Fiscal projections under these assumptions are as follows:

Table 2.Mexico: Financial Operations of the Public Sector 2012-16(in percent of GDP)
Oil revenue7.
Non-oil revenue14.614.414.114.214.1
of whick revenue sharing3.
Traditional Balance 1/-2.0-1.9-1.8-1.7-1.6
Traditional Balance for fiscal rule 2/
Public Sector Borrowing Requirements 3/
Source: IMF Staff and “Criterios 2011”.

This is the same path as in “Criterios 2011”, p.101.

This is the Traditional Balance excluding investment by PEMEX.

These include the Traditional Balance deficit plus some adjustments (i.e., PIDIREGAS, IPAB, Budgetary Adjustments, FARAC, Debtor support, and net lending by development banks) minus transfers to oil stabilization funds.

Source: IMF Staff and “Criterios 2011”.

This is the same path as in “Criterios 2011”, p.101.

This is the Traditional Balance excluding investment by PEMEX.

These include the Traditional Balance deficit plus some adjustments (i.e., PIDIREGAS, IPAB, Budgetary Adjustments, FARAC, Debtor support, and net lending by development banks) minus transfers to oil stabilization funds.

4. Over the medium term, government revenues are expected to fall about 1.8 percent of GDP driven by lower oil revenues. The combination of broadly stable world oil prices with constant oil production/exports implies that oil revenues decline in percent of GDP. Non-oil revenues decrease slightly in 2012—13 as a result of the envisaged cut in income tax rates.4

5. Age-related spending, over the medium term, is expected to increase about 0.7 percent of GDP, mainly in response to demographic factors. Interest payments are also expected to increase in a context of a gradual increase in interest rates. Under the balance-budget rule, the combination of lower revenues and higher age-related spending would require a reduction in non age-related spending of about 3 percent of GDP during 2012—16.5 But since a significant fraction of non-age related spending are earmarked transfers to subnational governments, the spending compression would fall on other expenditures.6

6. Under the balanced-budget rule, the path for public debt would remain stable. This would be driven by an increasing primary balance that compensates for a declining growth-interest rate differential.7 In terms of sensitivity analysis, if GDP growth were 4 percent (as assumed in “Criterios 2011”) instead of 3.2 percent, then a slightly declining path for public debt would result. If the real effective exchange rate were to appreciate rather than remain constant, then a declining path for public debt would also result.

Mexico: Gross Public Debt

(in percent of GDP)

Sources: Staff calculations

B. Oil Revenue Outlook in the Long Term

7. This section extends oil revenue projections to 2030 and assesses its implications for the long-term sustainability of fiscal policy. Projecting over such a long horizon is subject to significant uncertainty, related inter alia to the path for the oil price, oil production, GDP growth, and the real exchange rate.

8. Production has fallen from peak levels but has stabilized in recent years.

  • In recent years oil production and reserves have fallen significantly. Proven reserves fell during 2004—07 and were broadly stabilized since then, with proven reserves 28 percent lower in 2010 than in 2003. Oil production reached a peak of 3,383 thousand barrels per day in 2004 and declined significantly in following years, but has stabilized more recently. The cumulative fall in production during 2004—10 has been 24 percent. The level of proven reserves in 2010 would last 10 years at the current production level.

  • Oil derivatives production has also been on a downward trend in recent years while imports have soared. The cumulative fall in oil derivatives production during 2004—10 has been 10 percent, explained by the fall in crude oil production.8 Oil derivatives imports in 2010 more than doubled those in 2004, led by imports of fuels.

  • Natural gas production has expanded significantly but remains a small share of oil revenues. The cumulative increase in natural gas production during 2004—10 has been more than 50 percent, but has stabilized since 2008. Imports of natural gas declined substantially during 2004—06 and recovered gradually thereafter. Natural gas accounted for 6 percent of total oil revenues in 2010.

9. The oil sector has been reformed in recent years, but further measures may be needed to realize the sector’s full potential. In 2008 there was an important reform of the state-owned oil sector (PEMEX). The main goals of the reform were to improve the governance of PEMEX and to attract private investment by allowing PEMEX to sign service contracts with private companies, which provide performance-based incentives.9 More recently some additional reforms were introduced giving PEMEX more financial flexibility to explore new fields in deep waters and in the field of Chincotepec. However, further reforms may be needed to foster the development of the sector, including fields in deep waters.10

Oil Sector

Sources: PEMEX, and staff calculations

10. Oil revenues could fall substantially in the long term. In our baseline scenario, in which the real price of oil is assumed to remain constant, revenues could fall by 2.5 percent of GDP during 2011—30.11 This result is explained by the fact that in a growing economy, a stagnant oil sector becomes less important as a source of revenue. In terms of sensitivity analysis, if oil prices (international and domestic) remain constant in nominal terms, then revenues would fall by around 4 percent of GDP. If crude oil production was gradually increased to its 2004 peak, then revenues would fall by 1.8 percent of GDP.

Oil Revenue

(percent of GDP)

Sources: Staff calculations

C. Pension Spending Outlook

11. Public pensions in Mexico do not appear relatively high compared with other emerging countries. In a sample of emerging countries, pensions in Mexico (2.6 percent of GDP) are well below the average (5.4 percent of GDP) in 2010. However, pensions have more than doubled in the last 15 years, increasing particularly rapidly since 2008.

Spending in Pensions

Sources: SHCP, IMF 2010a, and staff calculations.

12. Fiscal risks from pensions have been addressed by a series of important pension reforms. The pension system covering private sector employees (IMSS) was changed from a defined benefit system to a defined contribution system in 1997 and the pension system covering most public sector employees (ISSSTE) had a similar reform in 2007. IMSS’ own employees’ regime was reformed in 2004 and the regime of the public electricity company (CFE) was reformed in 2008.12

13. However, important fiscal risks and challenges from pensions remain. There are significant pension regimes that have not been reformed (i.e., PEMEX and subnational governments).13 Moreover, there is a high transition cost in shifting from a defined benefit to a defined contribution scheme There is also the risk that, for a fraction of workers, accumulated contributions in a defined contribution system may result in inadequate pensions, with the attendant fiscal risk.14 Finally, there is a relatively narrow fraction of the population participating in pensions systems,15 and pressures may emerge to expand noncontributory pensions which would create a substantial fiscal liability.

14. Pension spending is projected to increase to 4.6 percent of GDP by 2030.16 The projected pension spending increase of 2 percent of GDP during 2010—30 is relatively high compared to a sample of emerging countries (IMF (2010a)).17 The projected increase in pensions may be a conservative estimate given the current narrow coverage of the pension system. It should be underscored that pension spending is projected to peak in 2035 and then go down as the effects of the pension reforms kick in.18

Pension Spending Projections

Sources: SHCP, IMF 2010a, and staff calculations

D. Health Spending Outlook

15. Public spending in health in Mexico is below other emerging countries. Public spending in health has grown less than the spending in pensions, while it has been more volatile. Health indicators in Mexican have shown significant improvements over the past decades, but remain behind most OECD countries.19

Spending in Health

Sources: SHCP, IMF 2010a, and staff calculations

16. Mexico’s health care system has lagged in efficiency among OECD countries. Schwellnus (2009) compared health outcomes (e.g., life expectancy) with health inputs (e.g., health spending) and found that Mexico is among the least efficient health care systems in OECD countries and average among emerging countries. One key explanation behind these findings is that Mexico has a highly segmented public health system in which different health providers (supply side) have no access to each others’ services. In addition, there is lack of competition because patient choice is limited.

17. Mexico is close to achieving universal insurance of basic health services. The most important recent reform in the sector was the Health System of Social Protection (SPSS) that started in 2004. It introduced a system of family insurance, Seguro Popular (SP), to facilitate access to affordable health insurance for those without social security. Family contributions to SP are based on a sliding fee scale and are waived for families meeting the low income criteria. The relatively narrow health package provided by SP entails fiscal risks as pressures might emerge to expand the package in line with standard social security packages.

18. Mexico has yet to complete an epidemiological transition. Communicable and infectious diseases (e.g., influenza, pneumonia) are decreasing while chronic diseases (e.g., cancer, diabetes) are rising. This transition comes hand in hand with economic development and would imply higher health care costs in the future.

19. The main determinants of public health spending are demographics and excess growing costs (EGC). IMF (2010b) documents an average EGC of 1 percent for a sample of 27 advanced countries during 1980-2007.20 To project public health spending in Mexico for 2011—30 we follow IMF (2010b) and assume that the EGC is 1 percent each year and that spending increases proportionally with the share of the population aged 60 or more.21

Mexico: Public Health Spending Projections, 2017-30

(in percent of GDP)

Sources: Staff calculations

20. Public health spending would increase about 1.1 percent of GDP during 2010—30. The projected health spending increase is average among emerging countries. The increase in public health spending is a permanent factor, unlike the projected increase in public spending in pensions which would eventually fade away after 2035 (when the transition costs of pension reforms go down).

Projected Increase in Public Spending in Health, 2010-30

(Percent of GDP)

Sources: IMF 2010a, and staff calculations.

Appendix 1

Pension spending can be broken down in the following way:

  • PE / GDP = (Pensioners * Average Pension)/GDP

  • • PE / GDP = Pensioners * (Average Pension / Average Wage) * (Average Wage / GDP)

  • PE / GDP = (Pensioners / Pop65+) * (Average Pension / Average Wage) * [(Average Wage * Workers) / GDP] * (Pop65+/Workers)

  • PE / GDP = (Pensioners / Pop65+) * (Average Pension/Average Wage)* (Wages / GDP) * (Pop65+ / Pop15-64) * (Pop15-64 / Workers)

(Pensioners / Pop65+) measures the coverage of the pension system.

(Average Pension/Average Wage) measures the generosity of the pension system.

(Pop65+ / Pop15-64) measures the old dependency ratio of the population.

(Pop15-64 / Workers) measures labor force participation.


Prepared by Pablo Lopez-Murphy (FAD)

The price of oil exports is obtained by assuming that the ratio between WEO prices and the price of Mexican oil exports in 2010 is maintained during 2012-16.

Transfers to oil stabilization funds are, by Law, a fraction of duties on hydrocarbons and excess tax revenues. Excess tax revenues are tax revenues higher than those projected in the budget. Medium-term projections assume no excess tax revenues.

Income tax rates were temporarily increased from 28 to 30 percent in 2010 as part of a fiscal package reverting the 2009 fiscal stimulus. The rates will be reduced to 29 percent in 2013 and to 28 percent in 2014.

Around 0.5 percent of the fall would result from lower investment by PEMEX given the expected increase in private investment as a result of recent changes to the fiscal regime of the oil sector.

Non age-related spending includes, spending in education and infrastructure. Schwellnus (2009) documents that education spending (and quality) in Mexico is low compared to other OECD countries.

Debt dynamics would be driven by: Dt - Dt-1 = PSBRt + et Dt-1* where Dt denotes public debt at the end of period t, et is the rate of depreciation of the exchange rate during period t, and D* is foreign currency debt (measured in domestic currency).

The decline in the production of oil derivatives is also explained by the fact that PEMEX has been reconfiguring some of the refineries to produce cleaner gasoline.

Payments should only be made in cash, and should never be a percentage of production, sales, or profits.

Joint ventures (i.e., production sharing deals) with other oil companies, which are common in the global industry, are constitutionally prohibited.

We assume that the macroeconomic parameters (i.e., growth, inflation, exchange rate, real oil price, oil production, oil exports) remain stationary. We also assume that the consumption of oil derivatives grows in line with the economy.

The reform of CFE’s pension system had a parametric component for current workers, and new workers participate in a defined contribution scheme based on individual accounts.

World Bank (2007) documents that pension schemes sponsored by subnational governments are predominantly defined benefit schemes with short contribution periods and relatively young retirement ages.

OECD (2011) estimates that projected replacements rates (i.e., pensions as a fraction of pre-retirement income) in Mexico are around 30 percent while the OECD average is 57 percent.

Levy (2009) documents that less than half of the labor force is covered by social security.

The appendix lays out the framework used to project pension spending. We assume that the dependency ratio increases from 9.9 percent in 2010 to 17.5 percent in 2030 in line with demographic projections, and that the employment ratio increases from 63.4 percent in 2010 to 66.8 percent in 2030 in line with recent trends. We also assume that coverage and generosity remain unchanged. The increase in pension spending might generate some small increase in income tax revenues but not much because no income tax is paid on pensions that are below 9 times the minimum wage.

IMSS (2010) shows cash flow projections in which pension spending grows from $144,688 million in 2010 to $487,333 million (at constant prices) in 2030. This cumulative growth during 2010-30 is in line with our projections.

Sales et al. (1999) estimate that the transition cost will peak in 2035. People who started working in 1997 will normally retire in 2037.

In 2007, life expectancy at birth in Mexico was 75 years, higher than in the Slovak Republic (74.3), Hungary (73.3), and Turkey (73.2).

Public health spending-to-GDP ratio in advanced countries has increased on average by almost 2 percent of GDP since 1980. IMF (2010b) presents evidence showing that spending increases do not appear to be correlated with initial levels of income per capita or rates of economic growth suggesting that non-income factors (e.g., aging of the population, medical technology) are the key drivers explaining cross-country and time series variations.

The assumption of 1 percent EGC means that the health spending-to-GDP ratio increases 1 percent each year (e.g., from 3 percent of GDP in one year to 3.03 percent of GDP the following year). This assumption may be conservative for a country that is yet to complete epidemiological transition.

Other Resources Citing This Publication