Portugal: Selected Issues

This Selected Issues paper of Portugal highlights the discussions on the requirement of policies to overcome structural and cyclical impediments to growth, and secure fiscal consolidation. It analyzes the strength of the company balance sheets in supporting the rebound from recession, and also the links between corporate balance sheet strength and investment. It reviews the challenges in the Portuguese economy, the impact of European Union enlargement on Portuguese trade, the pension prospects, and the implications of various policy reform scenarios.

Abstract

This Selected Issues paper of Portugal highlights the discussions on the requirement of policies to overcome structural and cyclical impediments to growth, and secure fiscal consolidation. It analyzes the strength of the company balance sheets in supporting the rebound from recession, and also the links between corporate balance sheet strength and investment. It reviews the challenges in the Portuguese economy, the impact of European Union enlargement on Portuguese trade, the pension prospects, and the implications of various policy reform scenarios.

IV. Pension Reform Issues in Portugal23

A. Introduction

60. As in most other European Union economies, Portugal will face adverse demographic trends in the coming decades, which will, absent further reforms, result in significant aging-related spending pressures. While the Portuguese public pension system has undergone a number of reforms, including most recently in 2000 and 2002, and pension expenditures are slightly below the European Union (EU) average, the system remains fundamentally unsustainable in its present form. Moreover, with the general government debt burden already approaching 60 percent of GDP, and with social insurance contribution rates above the high EU average, there is little scope to finance these coming spending pressures. Thus, addressing them most likely will involve reducing the scope and scale of pension benefits themselves.

61. This paper provides a brief overview of the public pension system in Portugal, and of prospective pension-related spending pressures in an international context, and considers some policy options. Section B reviews the structure of the Portuguese pension system, and compares some of its basic parameters (e.g., retirement ages, replacement and accrual rates, indexation, taxation), as well as demographic projections, with those in other EU economies. Section C discusses Portuguese pension expenditure projections, and examines their sensitivity to various macroeconomic and other assumptions. Section D considers some policy reform options, and Section E provides conclusions.

B. Background, Previous Reforms, and Demographic Outlook

62. Portugal’s pension system comprises a mandatory publicly run, pay-as-you-go “first pillar,” complemented by a small, but growing, voluntary tax-advantaged third pillar. The first pillar comprises separate schemes for private sector employees and for civil servants. The private sector scheme, Seguranga Social (SS), was founded in 1935 for wage earners in industry and the service sectors, and is managed by the Instituto de Gestao Financeira da Seguranga Social. Coverage was extended over time to other sectors of the economy, with benefits made universal following the 1974 revolution. The SS scheme subsequently incorporated other occupation-specific schemes (e.g., for agricultural workers).24

63. The civil servants’ scheme, the Caixa Geral de Aposentagoes (CGA), was founded in 1929. While the system covers a much smaller number of participants, civil servants’ pension benefits are much higher, reflecting a more generous benefit formula and a greater maturity of the system where, unlike the private sector, earnings are not underreported. This is clear,

64. as Gouveia and Sarmento (2002) and Rodrigues (2002) discuss, by considering the following magnitudes. In 2000, the SS had 4.4 million active contributors compared to only 747,000 in the CGA (17 percent of the SS total). There were 2.5 million SS old age, survivor, and disability pensioners, compared to 428,000 CGA pensioners (also 17 percent of the SS total). However, SS pension expenditures equaled 6.1 percent of GDP in 2000, while CGA pensions equaled 3.6 percent of GDP (about 60 percent of the SS total). Thus, the average CGA pension was almost 3½ times as large as the average SS pension. It should be noted that the SS scheme includes many small pension payments either to individuals that had short or nonexistent contributory periods, or had very low wages. Almost one half of all SS pensioners received minimum pensions in 2000, and agricultural pensioners, accounting for an additional 18 percent of total SS pensioners, received similarly low transfers.

65. The Portuguese pension system has undergone a number of substantial reforms in recent times, most notably in 1993 and again in 2000 and 2002.25 In 1993 reforms affecting the SS included: lowering the accrual rate (the annual incremental increase in the ratio that is multiplied against the reference wage in determining an initial pension) from 2.2 percent to 2 percent; increasing from 10 to 15 the number of required contribution years to obtain a pension; increasing, from the best 5 out of the last 10 years to the best 10 out of the last 15 years, the base period for determining the reference wage, while indexing for inflation the historical wages for this calculation; increasing gradually from 62 to 65 the female retirement age (phased in by 1999); and allowing pensioners to earn labor income. The minimum and maximum replacement rates were set at 30 and 80 percent of the reference wage (after 15 and 40 years), respectively.

66. Some further reforms to the SS system were introduced in 2000 and 2002—addressing important equity concerns, but not the fundamental financial unsustainability of the system. The number of years used to calculate the reference salary was to be gradually extended from the best 10 of the last 15 years to the entire contribution history (to be phased in from 2017 to 2035), thereby reducing incentives to evade contributions early on, and improve the system’s finances. However, the fixed 2 percent accrual rate was replaced with a progressive schedule of 2–2.3 percent (for those with more than 20 years of contributions) to help those with minimum and relatively low pensions.26 In addition, contributory minimum pensions with more than 30 years of contributions were to be increased so as to converge with the statutory minimum wage (net of pension contributions) over the 2003–07 period, with lower targets for shorter contribution periods.27 It has been estimated that the net result of the first two changes did not improve the system’s long-run financial position, and may in fact have worsened it (Rodrigues, 2002). Clearly, increasing minimum pensions, while socially important, will also impose an additional fiscal burden.

67. The civil service pension system also underwent a fundamental reform in 1993, with a splitting of the system into two groups: those employed in the civil service prior to and after September 1993. The system did not change for those employed prior to this date, with a minimum of only 5 years required to become vested for a civil service pension (compared to 15 years in the SS system). The monthly pension was equivalent to 100 percent of the wage earned in the last month of employment (gross of tax and social contributions, plus the average of the last two months of other forms of compensation) for those with 36 years of contributions (and at least 60 years old, unless the employer did not object).28 Thus the annual accrual rate was implicitly 2.8 percent (equivalent to 1/36 per year), and a proportional reduction in pensions applied for those who retired at age 65 with less than 36 years service. From 2004 onwards, the monthly wage taken into account in calculating initial pensions for all pre-1993 civil servants that hereafter retire was reduced to wages (and nonwage benefits) net of CGA employee pension contributions, thereby reducing these pensions effectively by about 10 percent. However, the practice discussed in the footnote below of not adjusting pensions in line with civil servants wage increases until they fell to 90 percent of their initial value would not apply to these new pensioners. Additionally, employees with at least 36 years of civil service may still retire earlier than the statutory retirement age, but now with a proportional reduction in their pensions for each year of the difference with the statutory age. In contrast to these still relatively generous pension benefits, civil servants employed after September 1993, are subject to the much less munificent SS rules.

68. From an international perspective, Portugal’s SS pension system is not an outlier, although considered by the European Commission to provide medium to high relative benefit levels (Table 1). Its statutory standard retirement age, 65 years, is most frequently used in EU economies, although its early retirement age is somewhat lower. However, the actual average retirement age in Portugal is not unusually low. Life expectancy in Portugal is the lowest among EU countries, although the projected increase by 2050 is the highest. The statutory replacement rate is, at 80 percent, among the highest, and the average replacement rate is also near the top.29 This reflects in part Portugal having the second highest accrual rate in the EU. It also reflects the Portuguese “indexation” system. While formally the scheme is ad hoc, SS pensions have traditionally been adjusted at about 1 percentage point higher than consumer price inflation, while many other EU systems have shifted to formal price indexation.30 CGA pension increases have been linked to the increases in the civil servants wage scale.31 Finally, Portuguese public pensions have very favorable personal income tax treatment, as contributions are deductible in determining taxable income, while pensions face much lower rates of effective taxation than do wages.32 This also is in contrast to the general trend in the EU to subject pension income more generally to taxation.

Table 1.

Portugal: Summary of Public Pension System Parameters

article image
Sources: OECD (2003); European Commission (2000, 2001, 2002).

The first figures before the slash is for males, the second is for females.

Data for Denmark, Germany, Spain, Greece, Austria, Netherlands, Portugal, Finland, and Sweden is for 2000; for Italy, Luxembourg, and the United Kingdom from 1999; for Belgium, France, and Ireland is from 1998.

For the first pillar system, before tax, in percent of average wage.

Figures after the slash indicate the outcome at the end of the transitional period.

W: indexed to wages; P: indexed to prices; M: mixed wages and prices; A: indexation is ad hoc.

F: fully taxed; P:preferentially taxed.

69. Social security contribution rates are already relatively high in Portugal (Table 2).33 While these include charges for other social security programs in addition to pensions (including sickness, maternity, family, unemployment and death benefits, occupational hazards and unemployment compensation, and subsidies for the handicapped), prospects for financing higher pension expenditures with higher contributions appear to be quite limited.

Table 2.

Portugal: Total Social Security Contribution Rates, 2002

article image
Source: U.S. Government “Social Security Programs Throughout the World: Europe” (2002).

Includes old age, disability, and survivors; sickness and maternity; work injury; unemployment; and family allowances. In some countries, the rate may not cover all of these programs. In some cases, only certain groups, such as wage earners, are represented. When contribution rates vary, either the average of the lowest rate in the range is used.

Contributions are subject to a ceiling for some benefits.

Portion of set amount for old age, disability, and survivors. Central and local

The government pays the total cost of family allowances.

Range according to earnings bracket. Higher rate is shown, which applies to highest earnings class.

Government pays the total cost of basic unemployment benefits.

70. Demographic projections suggest that population aging in Portugal would be broadly comparable to developments in the EU on average (Figure 1). The old-age dependency ratio was 23 percent in 2000, almost equal to the 24 percent EU average, well below the 27 percent rate in Italy and Sweden, while sizably above the 17 percent rate in Ireland. Based on Eurostat’s central population projection variant, the Portuguese dependency ratio is expected to increase in 2050 to 46 percent, compared to 49 percent on average in the EU.34 With the fertility rate now below the replacement rate, as in all EU countries, the Portuguese population is projected to peak at slightly above 10.9 million in 2040, thereafter declining to 10.8 million in 2050.

Figure 1.
Figure 1.

Portugal: Trends in Old-Age Dependency Ratios, 2000-50

Citation: IMF Staff Country Reports 2004, 081; 10.5089/9781451832181.002.A004

Sources: Eurostat, as cited in Economic Policy Committee, European Commission (2001).

C. Pension Expenditure Projections

71. Pension expenditures in Portugal were equivalent to 9.8 percent of GDP in 2000, somewhat below the EU average (Table 3). The authorities project that pension expenditures will steadily increase, peaking at 12.8 percent of GDP in 2035, before declining to 12.1 percent in 2050, with one-third of the increase accounted for by much smaller CGA pension system.35 The projected increase is also somewhat below the EU average. Nevertheless, this, along with projected aging-related increases in healthcare expenditures, would without offsetting actions result in a clearly unsustainable fiscal outlook (Figure 2).36

Table 3.

Portugal: Pension Expenditure Projections, 2000-2050

(In percent of GDP)

article image
Sources: European Commission, Economic Policy Committee (2001, 2003); and Portuguese authorities.

Updated projection included in European Commission (2003)/.

Data in 2000 column refer to 2005.

Expressed as a share of GNP.

As shown in the latest Portuguese Stability Program.

Based on projections contained in European Commission (2001)/.

Figure 2.
Figure 2.

Portugal: Long-Term Aging-Related Fiscal Projections, 2000-50

(In percent of GDP)

Citation: IMF Staff Country Reports 2004, 081; 10.5089/9781451832181.002.A004

Sources: Portuguese authorities; and Fund staff calculations.1/ Excludes long-term care.

72. The pension projections are sensitive to demographic and macroeconomic assumptions. Notably, the Portuguese authorities’ projections assume a relatively rapid convergence in real living standards, such that per capita incomes would reach 90 percent of the EU average level by 2040 (versus 71 percent on a purchasing power parity standard in 2003), with complete convergence by 2050.37 Given the demographic and labor market assumptions, this has implications for the growth in average labor productivity (see text table).38 Clearly, the implicit increases in labor productivity are quite ambitious, especially when compared with developments in Portugal and the EU on average in the last decade (Table 4). And pension projections are sensitive to this assumption. As an illustration, Figure 3 compares the authorities’ baseline projections with one in which average labor productivity were to grow by ½ percentage point less annually (averaging 2.1 percent compared to 2.6 percent in the SS system projection), with pension expenditures increasing by an additional 2 percentage points of GDP. Thus, the health of the pension system is likely to be sensitive to the economy’s performance, with a possibility that the baseline projection may be too optimistic.

Figure 3.
Figure 3.

Portugal: Public Pension Expenditures, 2000-50

(In percent of GDP)

Citation: IMF Staff Country Reports 2004, 081; 10.5089/9781451832181.002.A004

Sources: Fund staff simulations.

Portugal: Authorities’ Official Penison Scenario

(Annual Average Growth)

article image
Table 4.

Portugal: Average Labor Productivity Growth

article image
Sources: OECD, Economic Outlook database (No. 73 - June 2003); and Fund staff calculations.

Unweighted average.

73. Additionally, the authorities’ baseline projection assumes that female labor participation rates will increase significantly by 2010, converging to male participation rates (Table 5). Portuguese labor force participation rates, male, female and overall, are already above the EU average rates. In particular, female participation rates for those aged 25–54, and aged 55–64 are already more than 4 and 10 percentage points above their EU average counterparts, respectively (Figure 4). The projected increase would still leave the 25–54 cohort participation rate below those presently in Sweden, Finland, and Denmark, while the increase for the 55–64 cohort would leave it trailing only Sweden. Alternatively, were the increased female participation rates to be only one-half as large, projected pension expenditures would increase only marginally (at most by ¼ percentage point of GDP). Combining both the lower projected increase in female labor force participation rates and lower productivity growth would increase projected pension expenditures by 2 percentage points.

Figure 4.
Figure 4.

Portugal: Labor Force Participation Rates in Selected OECD Countries, 2002

Citation: IMF Staff Country Reports 2004, 081; 10.5089/9781451832181.002.A004

Source: OECD, Employment Outlook, 2003.
Table 5.

Portugal: Labor Force Participation Rates

(Percent of Population)

article image
Sources: Portuguese authorities; and Fund staff estimates.

74. There are a number of additional aspects that suggest that the authorities’ baseline projections may be quite optimistic. First, because indexation of CGA pensions to civil servants’ wage increases is ad hoc, rather than a legal entitlement, the baseline CGA projection assumes annual updates in line with inflation. If full wage growth indexation were assumed, however, CGA pension expenditures would be about 2½ percentage points of GDP higher in 2040, and some 1 ¾ percentage points of GDP higher in 2050. A second reason why the baseline projections may be too optimistic is because they do not include the implications of the CGA system’s recent incorporation of the postal system’s pension scheme, although estimates of this transaction are not presently available.39 The effects of the optimistic bias of the official figures for the civil servants pension system are partially offset by the recent changes in the rules followed by the calculation of initial pensions, which are not included in the projections (see paragraph 66 above). Finally, for the private sector pension system, the baseline projections do not model the impact of raising contributory minimum pensions to align them with targeted fractions of the minimum wage (net of pension contributions). While it is difficult to estimate precisely this effect, the fact that more than one half of all SS pensioners effectively received minimum pensions, and that minimum pensions are only about 50 percent of the minimum wage, it could be sizable, at least initially until contributors establish on average longer contributory histories, and the share of those receiving minimum pensions declines.

75. The increase in the authorities’ baseline pension expenditure projections are somewhat below that in the EU on average. While projections over such a long time period are subject to considerable uncertainty, there appear to be a number of considerations, reflecting macroeconomic and demographic aspects, suggesting that the baseline projections may be somewhat optimistic. Thus, the need for expenditure reforms may be even greater than suggested by the official projections.

D. Policy Reform Possibilities

76. It is clear from the discussion above that Portugal will face significant aging-related pension expenditure pressures, and possibly to a greater degree than suggested by the authorities’ projections. This section examines changes in various parameters, and in combinations of parameters, affecting pension benefits in order to gauge the magnitudes of the improvements that would result.40 Clearly, other comprehensive reforms could also be considered, including shifting to a defined contribution (or notional defined contribution) pillar, but are not examined here.

77. One possible reform to be considered is a phased increase in the retirement age. Figure 5 displays the revised pension expenditure path that would result from a steady increase in the retirement age to 69 years by 2040.41 This is broadly equal to the projected 5½ year increase in life expectancies over this period. It would reduce projected pension expenditures by almost 1 percentage point of GDP by 2050 under the authorities’ baseline macroeconomic scenario, and by similar amount under the low-growth macroeconomic scenario (with both lower productivity growth and lower increases in female participation, as shown in the bottom panel of Figure 5). Thus, it would appear that this reform alone would not be sufficient to make the system fundamentally sustainable.

Figure 5.
Figure 5.

Portugal: Single Parametric Reform Scenarios, 2000-50

(In percent of GDP)

Citation: IMF Staff Country Reports 2004, 081; 10.5089/9781451832181.002.A004

Source: Fund staff estimates.
Figure 6.
Figure 6.

Portugal: Combined Parametric Reforms Scenarios, 2000-50

(In percent of GDP)

Citation: IMF Staff Country Reports 2004, 081; 10.5089/9781451832181.002.A004

Source: Fund staff estimates.

78. A second possible reform is a reduction in the annual accrual rate. As an example, Figure 5 shows the effects of reducing the rate to 1.85 percent for both the SS and CGA systems beginning in 2010. This would also reduce the increase in the combined projected pension expenditure by roughly 1 percentage point of GDP under both sets of macroeconomic/demographic assumptions by 2050.

79. A third option is to reduce the rate of indexation. As noted, the CGA projections already assume that indexation covers only inflation, while the SS pensions increase each year by about 1 percent in real terms. Reducing the latter to cover only inflation beginning in 2005 (except for minimum pensions, which continue to get a 1 percentage point real increase per year) would also reduce expenditures by about 1 percentage point of GDP under the baseline macro/demographic assumptions, as well as under the less optimistic case by 2050 (Figure 5).

80. A fourth option considered is to increase the effective rate of taxation of pension incomes. For illustrative purposes, Figure 5 includes the case where standard (i.e., nonminimum) SS pensions and CGA pensions are subject to a flat 10 percent income tax rate. This increase is quite sizable, and, of all of the options considered has the largest impact, reducing net pension expenditures by ½ percentage points of GDP under both macroeconomic scenarios, which offsets more than half of the increase under the baseline macro/demographic scenario. Given the high share of minimum pensions in the present system, introducing a more complex taxation scheme (e.g., subjecting pensions to progressive taxation) may yield different, possibly smaller, net improvements to the system.

81. A fifth possibility considered here is to combine a number of these reform options. Introducing both a higher retirement age and lower indexation would reduce the projected increase in pension expenditures by about 1¾ percentage points of GDP under the baseline scenario, and by 1½ percentage points of GDP under the more pessimistic case in 2050. Combining these two reforms with a 10 percent income tax would reduce net expenditures by 2¾ percentage points under the baseline scenario, and by 2½ percentage points under the other case. This final combination would be sufficient to contain net expenditures at its current level under the authorities’ baseline scenario, while it would offset almost 60 percent of the increase in the more pessimistic scenario.

E. Conclusions

82. Despite earlier reforms, the Portuguese pension system remains fundamentally financially unsustainable. The increase in the authorities’ projected pension expenditures are broadly in line with those in other EU countries, but may, for macroeconomic and demographic reasons, be somewhat understated. Given already high employee and employer contribution rates and a sizable public debt, pension benefits themselves would most likely need to be reduced (in relation to GDP) to stabilize the system. A number of parametric reforms were considered in this chapter. However, each alone would most likely not be sufficient to address fully the anticipated spending pressures. This suggests that consideration be given to combine a number of different reform options.

References

  • Caldas, G. C., and P. G. Rodrigues, 2003, “Budgetary Costs of an Aging Population: The Case of Health Care in Portugal,” Portuguese Ministry of Finance DGEP Working Paper No. 31, Lisbon, January.

    • Search Google Scholar
    • Export Citation
  • European Commission, 2000, “Progress Report to the Ecofin Council on the Impact of Aging Populations on Public Pension Systems,” Economic Policy Committee, EPC/ECFIN/581/00-EN—Rev. 1, Brussels, November 6.

    • Search Google Scholar
    • Export Citation
  • European Commission, 2001, “Budgetary Challenges Posed by Aging Populations: The Impact on Public Spending on Pensions, Health and Long-Term Care for the Elderly and Possible Indicators of the Long-Term Sustainability of Public Finances,” Economic Policy Committee, EPC/ECFIN/655/01-EN—final, Brussels, October 24.

    • Search Google Scholar
    • Export Citation
  • European Commission, 2002, “Reform Challenges Facing Public Pension Systems: The Impact of Certain Parametric Reforms on Pension Expenditure,” Economic Policy Committee, EPC/ECFIN/237/02-EN—final, Brussels, July 5.

    • Search Google Scholar
    • Export Citation
  • Eurostat, 2004, “Regional GDP per capita in the EU and the Acceding Countries in 2001,” News Release 21/2004, February 18, 2004 (http://europa.eu.int/comm/eurostat/Public/datashop/print-product/EN?catalogue=Eurostat…product=2-18022004-EN-AP-…mode-download).

    • Search Google Scholar
    • Export Citation
  • Gouveia, M., and L. M. Sarmento, 2002, “Financing Pensions for Public Sector Workers in Portugal: Estimates of the Long-Run Impact on Public Finances,” in Economic Bulletin, Banco de Portugal, Lisbon, June, pp. 41-57.

    • Search Google Scholar
    • Export Citation
  • Government of Portugal, 2003, “Stability and Growth Program: Update for 2004-2007,” Lisbon, December (available at http://www.dgep.pt/pconvnational.html).

    • Search Google Scholar
    • Export Citation
  • OECD, 2003, “Monitoring the Future Social Implications of Today’s Pension Policies,” Working Party on Social Policy paper prepared for Meeting Held at Chateau del la Muette, Paris, November 17-18.

    • Search Google Scholar
    • Export Citation
  • Pereira, A., and P. G. Rodrigues, 2001, “Aging and Public Pensions in Portugal: A Snapshot Before the Reform,” manuscript, October. (Available on the EU website at http://europa.eu.int/comm/economy_finance/epc/documents/pt_en.pdf).

    • Search Google Scholar
    • Export Citation
  • Price WaterhouseCoopers, 2002, Individual Taxes 2002-2003: Worldwide Summaries, John Wiley and Sons, Hoboken, New Jersey.

  • Rodrigues, P. G., 2002, “Social Security in Portugal: An Update of Long-Term Projections,” Portuguese Ministry of Finance DGEP Working Paper no. 27, Lisbon, October.

    • Search Google Scholar
    • Export Citation
  • World Bank, 2000, “Pension Reform at Your Fingertips,” in Spectrum, Spring 2000, pp. 25-26 (available on the World Bank website at http://wwwl.worldbank.org/hdnetwork/pdf_notes/spring2000.pdf).

    • Search Google Scholar
    • Export Citation
23

Prepared by Mark Lutz.

24

Bank employees, numbering some 46,000 (about 1 percent of total employment), remain an important exception to the general private sector scheme.

25

These reforms are discussed more fully in Pereira and Rodrigues (2001), Gouveia and Sarmento (2002), and Rodrigues (2002).

26

In addition to these reforms directly affecting pensions, reforms to the financing side were also adopted, including defining government financial obligations for nonpension social expenditures administered by the SS (e.g., sickpay, maternity benefits, unemployment compensation), and establishing a reserve fund for the pension system equivalent to two years of pension payments.

27

Noncontributory minimum pensions are to be increased to 50–60 percent of the minimum wage less the employees’ social insurance contribution (of 11 percent).

28

Thus, given that pensioners no longer made social security contributions, and little of their pensions were subject to income taxation, after-tax pensions were generally in excess of 100 percent of civil servants’ take home pay. However, the system encompasses an erosion of these pensions until they reach 90 percent of the reference wage (equivalent to the wage net of the 10 percent pension contribution) before indexation applies (see Gouveia and Sarmento, 2002).

29

See European Commission (2002) for a detailed discussion of the estimation of average replacement ratios.

30

As a rule, minimum pensions are often increased in line with minimum wages.

31

Although recall that the system allowed for some initial erosion in real terms to 90 percent of the initial pension before indexation applies (see Gouveia and Sarmento, 2002).

32

In 2002, the first €7,805.60 of pension income was tax exempt, unless the taxpayer’s income exceeded a certain limit (€72,433 in 2002, equivalent to the prime minister’s salary), in which case, the exemption was reduced by amount the income exceeded the limit. See PriceWaterhouseCoopers (2002). Pereira and Rodrigues (2001) estimate that only about 7.5 percent of all pension income was taxed in 1997.

33

The civil servants’ contribution rate is 10 percent, with the government providing additional financing for CGA pension payments, as needed.

34

It is useful to note that the “central” population variant concerns the size of the overall population, but contains the most dramatic increases in the old-age dependency ratios.

35

These projections are contained in the most recent Stability Program (Government of Portugal, 2003), and are discussed more fully in Rodrigues (2002), incorporating the CGA projections in Gouveia and Sarmento (2002). These projections were made using PROST (Pension Reform Option Simulation Toolkit), a brief description of which is contained in World Bank (2000).

36

These fiscal projections are based on pension expenditure and economic growth projections contained in the Stability Program, and Rodrigues (2002), and healthcare expenditure projections in Caldas and Rodrigues (2003). The general government nonaging related primary structural (i.e., cyclically-adjusted) balance is assumed to remain at its 2003 level, and the real interest rate is set at 3 percent.

37

PPP adjusted per capita data are from Eurostat (2004).

38

These data are consistent with Rodrigues (2002). The macroeconomic assumptions for the CGA system projections differ slightly but are broadly comparable.

39

The postal pension system was partially funded and its assets were also transferred to the CGA along with its future expenditure obligations. While this will worsen the general government’s position, it will not affect the net worth of the broader public sector (including government owned entities).

40

We thank the Portuguese authorities for providing the data necessary to run the various reform scenarios.

41

The retirement age was increased linearly to 66 in 2010, to 67 in 2020, 68 in 2030, and to 69 years in 2040.

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

    Portugal: Trends in Old-Age Dependency Ratios, 2000-50

  • View in gallery

    Portugal: Long-Term Aging-Related Fiscal Projections, 2000-50

    (In percent of GDP)

  • View in gallery

    Portugal: Public Pension Expenditures, 2000-50

    (In percent of GDP)

  • View in gallery

    Portugal: Labor Force Participation Rates in Selected OECD Countries, 2002

  • View in gallery

    Portugal: Single Parametric Reform Scenarios, 2000-50

    (In percent of GDP)

  • View in gallery

    Portugal: Combined Parametric Reforms Scenarios, 2000-50

    (In percent of GDP)