CAIRN-INT.INFO : International Edition


1This paper [1] analyses the pension system reform introduced in Portugal in 2007. It argues that the new measures implemented, although not explicitly, were deeply transformative as they involved a change in policy objectives and principles. Up to 2007, the contributory scheme was directed to meet two objectives: workers’ income maintenance and the provision of a minimum income to those who earned insufficient pension entitlements. The new model has abandoned the objective of workers’ income maintenance. Moreover, although the system still operates on a pay-as-you-go basis, the principles that rule it have changed. The change was not explicit and sudden, but the development of the new measures over time will result in a gradual but significant change in the structure of the system.

2The paper is organized as follows. Section 2 examines the pension system reform introduced in Portugal in 2007 and the influence of European policies and orientations on its development. After drawing a short synthesis of the European strategy for pensions and the main developments in pension reforms in the Member States, the section presents an outline of the major legislative changes introduced in Portugal.

3Section 3 provides an assessment of the reform. The argument used to justify the reform was the need to contain pressures on public finances and the adjustment was achieved exclusively through benefit retrenchment. The strategy of benefit retrenchment is assessed on the basis of its consequences on pensioners’ well-being and the relative relevance attributed to the means and ends of policy. The normative choices underlying the reform and the change in the objectives and principles of policy are examined.

4Finally, Section 4 provides arguments to question the official discourse which stated that the reform preserved the fundamental characteristics of the scheme and, thus, the adjustment made was merely “parametric”. The conclusion is that there are sound reasons to consider that a new perspective regarding pension policy principles and objectives has been adopted. The reform may be described as “structural”.

The Reform of the Portuguese Social Security Pension Scheme in 2007

Pension Reforms in European Union Member States

5The reform program introduced in Portugal in 2007 should not be analyzed in isolation since it was deeply influenced by political decisions of the European Union (EU). The impact of the European level on pension reforms has been significant. The influence of the EU was twofold: firstly, European economic integration imposed constraints on national welfare states and their financing; secondly, European policy recommendations played an important role in framing the discussion on pension reforms.

6Firstly, it is important to mention the pressure imposed for reforming pensions by European fiscal policies. Since the early nineties, the Maastricht “culture of discipline” (Fitoussi and Laurent, 2009) has materialized into the definition of strict rules for fiscal policy management that have favoured the introduction of pension reforms. Since then, there have been direct pressures for reforming pensions, as national governments have been recommended to cut spending on pensions and other social transfer programs in order to lower public deficits and debts, thereby contributing to the consolidation of public finances as a whole. Additionally, some critics have observed that restrictive fiscal policy has imposed indirect pressures for reforming pensions through its contribution to the slowdown of growth and the increase of unemployment, thereby making “the burden of the welfare state heavier” (Fitoussi, 2004, p. 4).

7Secondly, the European framing of pension reform has also influenced the Portuguese strategy, the general policy orientation as well as specific measures introduced.

8A common vision of the “appropriate” design of pension systems has been built and promoted by the EU institutions. Since 1999, the reform of public pension systems has been included in the political agenda. In December, the Economic Policy Committee (EPC) organized a working group asked to examine the impact of population ageing on pensions’ expenditures and on public finances. Following an ECOFIN Council demand, the EPC presented projections for expenditure on public schemes financed on a pay-as-you-go (PAYG) basis up to 2050 (EPC, 2001) and drawn political recommendations based on the analysis of the working group. Member States were asked to adopt measures in order to guarantee the financial sustainability of PAYG schemes, mostly by containing future spending, and to encourage the development of fully funded schemes and voluntary provision.

9The effects of ageing on pensions’ spending were highlighted in several European Councils. The Stockholm Council, in March 2001, recommended Member States to reduce public debts faster, to reform their pension systems in order to contain pressures on public finances, and to reform their labour market in order to attain higher employment rates for women and the elderly. As requested since that Council, the effect of ageing on pension spending has been assessed in the framework of the Stability and Growth Pact. Since then, expenditures on pensions have been under attentive supervision by the EU institutions charged with economic and financial affairs, as a consequence of the budgetary discipline imposed to Member States by the Pact.

10In June 2001, the Gothenburg Council endorsed three principles for reforming pension systems: adequacy, financial sustainability and adaptability. In December, at the Laeken Council these principles were decomposed into eleven common objectives and an open method of coordination (OMC) was launched. The open coordination process involves the definition of common objectives, the use of a set of common indicators to assess progress towards objectives, and reporting by Member States on the basis of those objectives to the European Commission. As the definition of common objectives involves normative choices, this process has gradually created a shared view, which has been translated into orientations for pension policy, contributing to shape pension reforms.

11The OMC has been influential for decisions on national pension policy. It is indeed possible to identify common trends in reform trajectories of several Member States. After the publication of the first wave of National Strategy Reports, in 2002, the Commission and the Council recommended governments of Member States to undertake comprehensive pension reforms in order to meet the common objectives (EC, 2003). The States were called on to improve incentives for older workers to remain in the labour market, to tighten the link between contributions and benefits, to increase the share of public and private funding in pension provision and to ensure adequate minimum retirement income in order to prevent poverty. After the publication of the second wave of National Strategy Reports, the analysis of progress made in reforming pension (EC, 2006) recognized that, since 2003, Member States have introduced significant reforms with the aim of attaining the three goals, i.e. adequacy, financial sustainability and adaptability to changes in the labour market and society. The comparative analysis of these reforms revealed the presence of common trends. Many countries have introduced rule changes like incentives for working longer coupled with disincentives for early retirement, a stronger link between contributions and benefits, and the inclusion of life expectancy in the pension formula. Some countries have increased the levels of guaranteed minimum pensions in order to prevent poverty in old age. The role of private provision, complementing or partially replacing public provision, has been promoted. Portugal was no exception to the rule, as will be detailed in this contribution.

12This pension policy agenda in Portugal was strongly influenced by the economic policy paradigm prevailing in the EU. The policy orientation adopted for the introduction of the single currency was defined in accordance with the prescriptions of “New Classical Economics” (Creel, 2011). The role attributed to fiscal and monetary policies in stabilizing economies is limited. Monetary policy is committed to the sole objective of price stability and national budgetary policies are subject to strict limits. Therefore, the capacity of governments to promote growth and employment is severely limited. The underlying assumption is that growth and employment may be obtained not through expansionary macroeconomic policies but through supply-side economic policies. In turn, labour market and social policies should adapt to this economic orientation and cope with the economic environment created by external trade liberalization. Labour market and social policies should contribute to promote sound public finances and the competitiveness of economies.

13Thus, pension policy – like other social policies – is conceived in an instrumental way, as it is placed at the service of fiscal consolidation and the good performance of the labour market. In order to avert increasing public deficits and debts, pension provision should not rely exclusively on public intervention: expenditure on public schemes should be limited and an increased role for private funded schemes and voluntary provision should be allowed. Besides, pension policy should support the good functioning of labour market by promoting employment and competitiveness. It should focus on providing appropriate work incentives and activation strategies. Hence, it should concentrate on containing contributions to avoid the increase in labour costs and the consequent decrease in labour demand, on providing benefits that do not disincentive labour supply, and on activating older workers by providing incentives to work and discouraging early retirement (EC, 2006).

The Main Legislative Changes Introduced in Portugal In 2007

14In Portugal the general social security scheme is mandatory for employees and self-employed workers in the private sector. This scheme is by far the largest in the country. There is also a special scheme for public servants but, since 2005, it has been gradually harmonized with the general scheme (Law 60/2005, 29 December). Other occupational plans also exist but cover a limited percentage of the total active population.

15The general social security scheme provides earnings related benefits, retirement pensions and other benefits, like unemployment, sickness, parental leave, occupational sickness, disability, and survivor benefits. The scheme is financed by receipts from social contributions. The contribution rate is 34.75% (11% as employee contribution and 23.75% employer contribution). Since 2011, on the total 34.75 % of the contribution rate 20.21% is allocated to retirement pensions.

16The general social security pension scheme has been reformed by Law 4/2007, 16 January and Decree-Law 187/2007, 10 May. The main argument used to justify the reform was the need to contain budgetary pressures in order to ensure the long term fiscal sustainability of social security. The argument has been invoked previously in the government’s program (PCM, 2005) and in the text with the strategic lines of the reform: “The perspectives that social security system faces today point out to an imbalance in a decade, if meanwhile nothing is done… [The] structural trend is clear and imposes a reformist strategy that faces decisively the consequences of population ageing” (MTSS, 2006b, p. 9). The strategic lines of the reform include measures intended to contain spending and promote active ageing. The idea of reinforcing the contributory nature of the system and the principle of active ageing have been accepted by the social partners (CES, 2006) and were stated in the text of the law (Decree-Law 187/2007). The measures adopted were considered appropriate in order to face demographic ageing and its impact on the economy and public finances. The reform was presented to public opinion as absolutely necessary to guarantee the sustainability of the social security system.

17Regarding the set of new rules introduced, three main measures should be highlighted. Firstly, a central element was the introduction of a “sustainability factor” in the pension formula. It is a demographic adjustment factor that reduces pensions as average life expectancy increases. In order to avoid the decrease in pensions, workers may choose to extend their working lives or to increase their voluntary contributions to a new complementary public scheme of individual accounts.

18Secondly, the pension formula was changed. From 2007 on, pension levels are calculated on the basis of a reference remuneration that considers the earnings of the entire career. Since 1994, pension levels were calculated on the basis of the average earnings of the ten best years of the final fifteen. It is worth mentioning that a Law published in 2000 (Decree-Law 17/2000, 8 August) already stated that pensions should be calculated taking into account gradually the earnings of the entire career. The new method of calculating pensions was only adopted when a law introduced in 2002 (Decree-Law 35/2002, 19 February) came into effect. This law stated that the earnings of the entire career would be taken into account in the pension formula, but subject to a maximum of 40 years. However, the law also defined a transitional period, from 2002 to 2016, during which the most favourable method of calculation - the former, the latter or a weighted average of both - could be used to calculate pensions in order to guarantee beneficiaries the highest amounts. Thus, the new rule did not produce its full effects. The reform of 2007 has accelerated the transition to the new pension formula. The new rules state that pensions are now determined by a weighted average of the best ten years of earnings of the final fifteen and earnings of the whole career subject to a maximum of 40 years. The weight of each component depends on the year of retirement (before or after 2016) and the number of years of contributions before and after a specified year of reference (the reference year is 2006 for those who retire before 2016, or 2001, for those who retire after 2016). Table 1A in the Appendix summarizes the new calculation rules. Past earnings are indexed to prices, except the most recent ones (after 2002) that are indexed to prices and to the growth of earnings in the economy. The cut-off year will be periodically updated.

19Thirdly, a central element was the change in the rules for indexing the benefits after retirement (Law 53-B/2006, 29 December). Before the reform all benefits were adjusted to inflation, so that no retiree would lose purchasing power. After the reform, only pensions below a certain level were protected against inflation – the level is determined by the “Indexante de Apoios Sociais” (IAS), a level defined each year by the government, If economic growth is high enough (over 3% annual rate) all pensions are adjusted to inflation. If economic growth is between 2 and 3%, the highest pensions (more than 6 times the IAS) are not adjusted (their purchasing power decreases by the rate of inflation). If economic growth is lower than 2%, all those with pensions higher than 1.5 times the IAS lose purchasing power at the rate of inflation. This is a major rule change, as the previous objective of safeguarding the real value of all benefits was abandoned. In fact, the percentage of pensions higher that 1.5 IAS was not significant in 2007. It was around 11 percent of the total old-age pensions of the general social security scheme. However, this percentage will be much higher over time, since the rules governing the special scheme of public servants are gradually converging to these. Public servants present much longer insurance careers (on average 6 years longer) and average higher wages due to their much higher average education levels.

Table 1

Minimum Wage, IAS and Minimum Pension Level

Table 1
Years Minimum Wage (MW) (euros) Net MW (euros) IAS (euros) Minimum Pension Level (MPL) (career > 30 years) (euros) MPL/ Net MW (%) 2006 385,9 343,45 343,45 100,0 2007 403 358,67 397,86 354,10 98,7 2008 426 379,14 407,41 363,81 96,0 2009 450 400,50 419,22 374,36 93,5 2010 475 422,75 419,22 379,04 89,7 2011 485 431,65 419,22 379,04 78,2 2012 485 431,65 419,22 379,04 78,2

Minimum Wage, IAS and Minimum Pension Level

NOTE • monthly wages and pensions.

20Similarly, the process of gradual convergence of the minimum pension toward the minimum wage, which started almost ten years before and was completed in 2005, has been immediately reversed. In 2006 pensioners with insurance careers longer than thirty years were entitled to a pension equal to the net minimum wage (net of employees’ contributions). Those with shorter insurance careers were entitled to a certain percentage of that amount. However, this policy has been immediately reversed, as the government decided to disconnect the minimum pension levels from the minimum wage. The IAS now serves as the reference for minimum pension benefits. The minimum pension levels even for those with long careers started to diverge from the net minimum wage, as the following table shows.

21Additionally, the law also introduced new rules intended to promote “active ageing”, by encouraging people to work longer. Higher penalties for early retirement and incentives to postpone retirement were introduced.

Change of Path In Pension Policy

The Strategy of Benefit Retrenchment

22The reform of 2007 was justified with the argument that, if nothing was done, demographic ageing would cause deficits in the social security budget within a decade, as official projections revealed (MTSS, 2006a). The changes introduced were considered necessary to reduce the future increase in public pension expenditure and ensure its fiscal sustainability. The decision of the government was to base the adjustment exclusively on benefit retrenchment. This option may be criticized for its adverse effects on pensioners’ well-being. Also the methodology followed is debatable.

23The attempt to contain spending on pensions has serious consequences in terms of pensioners’ economic wellbeing. In Portugal, due to the late development of the social security pension scheme and past labour market conditions, at the time the reform was introduced the majority of the retired population had not yet reached a significant income security in old age (Murteira, 2008). At retirement, pensioners still suffered significant income losses, due to low levels of past earnings and short insurance periods, and pensions were just indexed to prices. Thus, pensioners’ standard of living was on average much lower than that of the active population and their risk of poverty was much higher. In this context, the strategy of lowering pensions may be criticized because it has placed the burden of the adjustment just on pensioners.

24In fact, the three main measures mentioned above reduce the level of benefits, affecting either the replacement rate or the subsequent income path.

25The replacement rates have been reduced due to the first two measures. Firstly, the new pension formula makes benefits dependent on the earnings of the entire career. Taking into account the highest earnings of the final working years or the earnings of the entire career as a basis for pension calculation has different implications as regards the replacement rate. As in general working careers present rising earnings in later years, when the benefits relate to the earnings of the entire career, the replacement rate is lower. In Portugal, the real average earnings of dependent workers had an upward trend since the beginning of the eighties. Thus, when the most favourable method of calculation (based on final earnings, on the earnings of the entire career or on a weighted average of both) could be used to calculate the pension level, in 82.5% of the cases, pensions were defined according to the former (best 10 years of final 15). Only workers with “atypical” careers benefited from the rule change. It has been estimated that, “during the period between 2002 and 2004, some 238,060 pensions were granted, of which 41,713 (approximately 17.5%) were in accordance with lifetime contributions (…) with their recipients thus being guaranteed the most favourable pension amount.” (EC, 2005, p. 3). This rule change reduces the level of benefits for the majority of new pensioners. Pensions have now a closer link to workers’ average earnings but are more distant from their final earnings and from current average earnings in society at the time of retirement. Secondly, the effect of the sustainability factor is gradual but significant. According to the OECD, it might lower future pensions substantially: “benefits are expected to be 81% of their value under current rules as the result of the link to life expectancy for an individual spending a whole career with this adjustment.” (2007, p. 68). In general, future benefits are expected to be cut more than 30% below what people would have been entitled before the reforms (OECD, 2007).

26Beyond the cut at retirement, pensioners’ income paths will be affected by the new rules of adjustment, which will cause a gradual erosion of the purchasing power of pensions higher than 1.5 IAS. Both the decrease in replacement rates and the new indexation method produce a growing gap between pensioners’ income path and the average earnings path in society.

27These outcomes were never discussed explicitly. The living standards perspective was never considered an argument by the proponents of the reform. Overall, the reform focused on the means of policies (spending on pensions) and neglected its ends (the social objectives). In fact, the level of benefits has been taken as the adjustment variable and treated as a subsidiary subject in a policy aimed at achieving financial sustainability. This should have been a matter of major concern because, as was mentioned above, Portuguese pensioners presented lower living standards and a higher risk of poverty than the overall population.

28It is worth mentioning that the objectives to attain regarding income security in retirement – including earnings replacement and the adjustment of income over time – were not mentioned either in the reports from social security administration (MTSS, 2006b) or in the text of the law (Decree-Law 187/2007). Only the State Budget Report (MFAP, 2006) has mentioned the planned decrease in the level of benefits. It seems that pension policy has been viewed exclusively from a financial perspective. According to the official projections, the expected increase in public expenditure on pensions (old age, invalidity and survival) pointed out to the system imbalance in a decade. The application of the new rules was forecast to allow a future reduction both of spending and of the real growth of average pension (Figure 1-A, in the Appendix). These effects were considered positive from the state budget perspective. Implicitly, the level of pensions was taken as the adjustment variable.

Figure 1A

Real growth rate of average pension

Figure 1A

Real growth rate of average pension

29From the social policy perspective, this methodology is disputable. The purpose of pension schemes is to provide retirement income security. Thus, a coherent policy design requires a clear specification of this aim. Social objectives in the sphere of retirement income security should always be treated as fundamental principles of any reform. They should be clearly stated at the outset and, obviously, assumed as the departure point in the design of any reform. The budgetary balance of the system should not be seen as the “objective” but as the “constraint”.

30The process of reforming pensions in Portugal has been considered a success by its promoters within the government and its supporters on the political right. This appraisal, however, results from a biased point of view that consists in assessing policies solely by their effects on inter-temporal budgetary balance, and neglecting the social objectives of pension provision. Atkinson (1999) has criticized this way of analyzing the welfare state that considers its costs but ignores the functions it is intended to perform. An appropriate assessment of pension policy should not view it exclusively from a financial perspective.

A Change of Objectives

31The analysis of the main legislative amendments introduced with the reform of 2007, indicates that a new perspective regarding pension policy objectives has been adopted in Portugal.

32Firstly, the change in the calculation method expresses a different choice of objectives. Underlying any calculation formula there is an implicit view regarding policy objectives. When the objective is to safeguard the previous living standards, benefits should be related to the final earnings, instead of the earnings of the entire career. When pensions depend on the earnings of the entire career, they diverge from workers’ final earnings and from the average earnings current in society at the time of retirement and are made to reflect the amount contributed by the pensioner, moving closer to the rationale of a retirement plan based on private savings. Thus, the rule change involved a shift from the former to the latter view, indicating a major change in policy objectives. This fundamental shift is discussed in more details in sections 3.3 and 3.4 below.

33Secondly, the shift in the method of adjusting pensions also reflects a major change of objectives: the adjustment of all benefits in line with inflation was abandoned as well as the adjustment of the minimum pensions in line with the minimum wage. A growing divergence between the pensioners’ income path and the overall income path in the society is implicitly admitted.

34The decrease of replacement rates for the major part of pensioners caused by the new pension formula, coupled to the new way of adjusting benefits, clearly reveal that the objective of safeguarding living standards in retirement – implicit in the previous rules – has been abandoned As regards pension policy, state intervention is becoming focused on the provision of a modest income protection, disregarding the previous view which accepted that the fundamental purpose of pensions was to allow people to maintain, as much as possible, the living standards achieved during the working period.

35The strategy of lowering pensions occurs at the same time the government has declared that the prevention of poverty among the elderly was a political priority. This strategy embodies a turning point in the system’s design thereafter focused on relieving poverty. An idea of social assistance, rather than a conception of social security, seems to have inspired this move. As Barry observed, “in a well-ordered welfare state almost all the job of relieving poverty will be done by policies whose objective and rationale are quite different. (…) if the welfare state is to be identified with one objective, it is that of income maintenance rather than the relief of poverty” (Barry, 1990, p. 73).

36This policy shift is clearly expressed by the measures addressed to pensioners with low incomes, such as a new mechanism of income protection for poor pensioners, a benefit that requires means testing (“complemento solidário para idosos”), created to alleviate old age poverty, which was introduced by the government in the beginning of legislature. In the foreword of the law that enacted it (Decree-Law 232/2005, 29 December), it is declared that the mechanism intends to be an element of a new policy of minimum benefits for the elderly. The previous policy, based on a general increase in minimum benefits, was considered unsustainable and had to be replaced by the strategy of concentrating the benefits on those in need. On this ground, a year later, the objective of convergence of the minimum pension level of the contributory scheme to the minimum wage was abandoned. Pensioners with full, or long, working careers were no longer entitled to a minimum pension at the level of the net minimum wage.

37The fact that this new mechanism was intended to replace the general increase in the minimum pensions of the contributory scheme is debatable. Means tested benefits are linked to the existence of a situation of need, while pensions of the contributory scheme are intended to replace earnings. Up till then, minimum pensions of the contributory scheme were not regarded as an instrument for avoiding poverty in old age, as a “national solidarity mechanism” for the working poor. Instead, the right to a minimum pension was conceived as a “work-related right”: the benefits were intended to replace earnings and, thus, were linked to the wage. According to the previous understanding, the net minimum wage was the appropriate reference for the definition of minimum pension levels. A minimum pension equal to the net minimum wage was granted to those who presented full or long working careers (the sufficient condition for entitlement). As the minimum wage was low – near the poverty line – it was assumed that the minimum pension should not be set below its net level.

On the Link between Contributions and Benefits

38Despite the negative consequences of the reform on benefit retrenchment, in the text of the law (Decree-Law, 187/2007) the transition to the new calculation rules is considered a requisite of justice, as it reinforces the contributory nature of the scheme.

39Actually, PAYG public pension schemes may have different designs as regards the link between individuals’ lifetime contributions and expected pension benefits. Different policy choices relate to different views on the desirable presence within the scheme of mechanisms of interindividual redistribution. One view advocates a strict link between individuals’ lifetime contributions and expected benefits. Thus, PAYG schemes should work like an intertemporal mechanism of individual income transfer between the working years and retirement. In such a scheme interindividual redistribution is absent ex ante, though it always occurs ex post due to differences in individual longevity. Actuarial neutrality is the implicit norm of justice. An opposite view emphasizes that PAYG public pension schemes do not need to keep a strict link between contributions and benefits. This link may be deliberately weakened and redistribution allowed for achieving welfare objectives. The schemes may include mechanisms such as “minimum pension”, higher replacement rates for lower incomes, etc. When the objective is to guarantee income maintenance, pensions should be tied to the previous wages, rather than to individual contributions.

40The logic underlying the Portuguese reform approaches the first stance: by setting a closer link between lifetime contributions and benefits, redistribution within the scheme is reduced. This move increases uncertainty (as individuals are not protected from the risk to having irregular careers) and tends to increase inequalities ex post as irregular careers tend to be systematically related to social or demographic characteristics like gender, income level, etc. Some authors argue that it is possible to offset this effect if the schemes allow people to acquire pension credits for periods of career breaks, such as childcare or old-age care, unemployment, sickness, etc. allowing the accrual of their pension rights. Yet, these offsetting mechanisms have merely partial effects. Cutting back the level of state guarantees, putting emphasis only on minimum pensions, and placing increasing responsibility and risk on individuals induces them to search for alternative sources of retirement income. In general, the distributive consequences are adverse since women and low-wage earners tend to lose: not only they usually have large non-contributory periods, broken employment histories, but also lower levels of earnings and savings.

41Additionally, as was mentioned above, the large majority of pensioners lose due to the decrease of the replacement rate caused by the new calculation rule based on the earnings of the entire career. Some authors argue that the reduction of replacement rates caused by this rule change is not problematic, since it will affect more high-wage earners than low-wage earners, as in general the former have faster growing earnings profiles in later working years. Therefore, the change in the pension formula should produce a higher drop of replacement rates for high-wage earners and, thus, reduce earnings inequality. However, this view is debatable. First, though inequality might have been reduced, incomes of most pensioners decrease. Second, even if low-wage earners lose less than high wage earners in absolute terms, a small earnings reduction when the income is low may be much more damaging than a large one when the income is high. Small earnings reductions for low-wage earners may cause major adverse effects in terms of welfare. Moreover, pension schemes might include mechanisms designed to offset the effect of faster growing earnings profiles in later working years for high-wage earners. The Portuguese scheme, for example, sets higher replacement rates for low-wage earners as it defines annual accrual rates in accordance to the level of reference earnings.

42Above all, the fact that the new rules cause a growing gap between pensioners’ income path and the average earnings path in society justifies, in my view, a further discussion of this policy option on an ethical basis.

New Principles Ruling PAYG

43A more subtle but profound change has occurred. Although PAYG has not been replaced by a funding scheme, its nature has been deeply changed. Other underlying principles are now ruling it: the previous scheme guaranteed work related rights while the new approaches the logic of retirement-savings.

44The initial PAYG schemes organized a form of solidarity through work, as Friot (2010) has emphasized. These schemes guaranteed work-related rights and pensions were linked to wages, since they are financed through social contributions. The amount of social contributions paid by current workers, which is a share of total wages, is immediately redistributed in the form of social benefits (pensions, unemployment benefits, sickness benefits, etc.). The initial schemes provided benefits proportional to the length of the working career, somewhat dependent on past earnings but not strictly contributory. When the purpose of policy is income maintenance, pension levels are close to the final wages. In this case, the pension may be conceived as a “continued salary”, to use the terms of Friot (2010). This conception requires high replacement rates and a similar progression of pensions and wages over time.

45Recently, in several countries, PAYG schemes have adopted a different logic as they have reinforced the link between benefit received and past contribution made to the scheme. A strictly contributory scheme is the most radical expression of the new logic. A strict link between lifetime contributions and benefits removes all redistribution within the scheme, as it becomes similar in its workings to a mechanism of individual income transfer between the working years and retirement. Contributions are seen as individual “savings” and benefits as the “deferred value of contributions”. In this case, pensions are conceived as individual “deferred income” (Friot, 2010; Castel, 2009).

46As regards the Portuguese scheme, the objective implicit in the previous pension formula consisted in maintaining the living standards in retirement. Pensions levels depended on the final earnings, in order to guarantee, as much as possible, the previous living standards and the indexation method could, at least, guarantee the maintenance of the real value of benefits. For low- wage earners with full working careers, the minimum pension level was equal to the net minimum wage. Actually, before the reform, the average pension levels were still considerably lower than the average earnings. But that was the consequence of the immature status of the pension scheme. With the scheme’s maturation, the average replacement rate would grow and pensions would approach the wage levels.

47The new scheme, although still operating on a PAYG basis, approaches the logic of retirement-savings. According to the perspective stated in the law, the scheme should become more contributory and, thus, pensions should become more linked to workers’ previous contributions (rather than the final earnings). This transition is considered a requisite of justice. Actuarial neutrality is the implicit norm of justice.

48In Portugal, the conception of pension as a work-related right fades away, as the link between pensions and wages weakens and pensions become more closely linked to contributions at the individual level. Several measures illustrate this move. A key element is the reduction of replacement rates associated with the introduction of the new pension formula and the sustainability factor that weakens the link between pensions and wages. Workers are allowed to receive in retirement a sum of benefits less related to the previous wages and more related to their previous contributions, as pensions became linked to the earnings of the entire career, reflecting more accurately contributions over the whole working career. Another key element that also contributed to the divergence of pensions from wages is the change in rules indexing benefits. In Portugal, pensions were never raised in line with wages. But the new method does not even guarantee that all benefits are indexed to prices, contributing to increase the gap between pension levels and wage levels over the retirement period. Finally, another important feature of the same process regards the disconnection of minimum pension levels from the minimum wage. The right to a minimum pension from the contributory scheme has been questioned as a work-related right and became justified on the basis of national solidarity. Only minimum pensions of very low level are guaranteed which might be supplemented by means-tested benefits.

49At the same time, the scheme has approached the actuarial mode of benefit calculation, as some features similar to those usually applied in retirement savings plans have been introduced. A fundamental one is the definition of a closer link between lifetime contributions and benefits. Additionally, the inclusion of a sustainability factor in the pension formula, making benefit levels dependent on remaining life expectancy at retirement age, seems to have been inspired by the actuarial technique of converting a certain pension accumulation into an annuity, as the value of the annuity depends on further life expectancy. Moreover, the principle of actuarial neutrality was also invoked in the text of the law to justify strengthening both incentives to postpone retirement and disincentives for early retirement.

50The scheme is still working on a PAYG basis, but its nature has been deeply changed as it has abandoned redistributive principles intended to guarantee income maintenance to become more ruled by actuarial principles. The logic of work related rights is replaced by the logic of retirement-savings, as the wage reference is disappearing and pensions become more linked to the contributions. As the objective of income maintenance has been abandoned, retirement pensions are no longer conceived as “continued salary” but as “deferred income”.

A Structural Reform

51The literature on pension reforms usually distinguishes two kinds of changes, “parametric” and “structural”. According to Hinrichs and Lynch, parametric reforms “constitute incremental adjustments to elements of the basic equation linking contributions and benefits”, while structural reforms “are systemic changes that move systems „off path’” (Hinrichs and Lynch, 2010, p.362). Changes like those introduced in Portugal in 2007 are often considered parametric. In the analysis of Hinrichs and Lynch, measures such as tightening the contribution/benefit link, the incorporation of demographic parameters in the pension formula and changes in the indexation method are described as examples of parametric adjustments., It is true that these rule changes were presented as path dependent, and intended to preserve or even reinforce the existing scheme and its principles. However, as was demonstrated in this paper, their introduction gradually leads to a major overhaul of the system and its basic principles. Major shifts in policy orientation may be caused by the “cumulative impact over time of gradual changes” (Palier, 2005, p.128). Indeed, gradual changes may lead to institutional discontinuity, as the debate on institutional change has emphasized (Streeck and Thelen, 2005).

52In the Portuguese case, according to official discourse, the reform preserved the fundamental characteristics of the scheme; the adjustment was intended to be merely parametric. However, this perspective is debatable. Indeed, some parameters of the existing scheme have changed but in a way that changed some of its fundamental characteristics. These changes, while gradual, involve a significant benefit retrenchment and reveal a new perspective as regards the objectives and principles of policy. Firstly, these changes result in a significant benefit retrenchment in the mid and long-term. Over time, the decrease in benefits will be felt more acutely, due to the growing influence of the sustainability factor. Also, the new indexation rules will cause a growing divergence between pensioners’ earnings path in retirement and the average earnings path in society. Such benefit retrenchment represents a significant decrease in the main source of income in old age for the majority of retirees. Secondly, although not explicitly, the new rules for calculating and adjusting the benefits involve a shift in the main objective of the system from income maintenance to the guarantee of a low income level. Thirdly, although the scheme still works on a PAYG basis, its configuration has been deeply changed: the logic of work related rights is replaced by the logic of retirement-savings, while the wage reference is disappearing and pensions become more linked to the contributions. Redistributive principles intended to guarantee income maintenance are abandoned while the system adopts actuarial principles. At the same time, the conception of pensions shifted from “continued salary” to “deferred income”.

53Overall, the system has moved „off path’. This transformation did not occur explicitly and was not immediately perceived, as the effects of the new rules are gradual. However, the measures as a whole result in deep changes in the fundamental nature of the system. By means of parametric changes, a major shift in pension policy orientation has occurred. The reform should be considered structural.


54The reform of 2007 was presented to public opinion as merely parametric, aimed at reducing the level of benefits to improve the financial sustainability of the scheme while preserving its overall logic. This paper has argued that this reform did not have a limited impact in the structure of pension system. Indeed it involved a change in the principles and objectives of the scheme. PAYG was not replaced by a funded scheme, but the principles that ruled it and the objectives have changed. As the change was not explicit, it is more difficult to identify.

Table 1A

Benefit Formulae

Table 1A
Beneficiaries that acceded to the scheme before 31/12/2001 (transitional rules apply) Beneficiaries that acceded to the scheme after 1/1/2002 (new rules apply) Retirement up to 31/12/2016 Retirement after 1/1/2017 earnings period: <20 years earnings period: ≥20 years P – Statutory pension P – Statutory pension P1 – Pension calculated on the basis of the best 10 years of earnings of the final fifteen P1 – Pension calculated on the basis of the best 10 years of earnings of the final fifteen aar- anual accrual rate aar- anual accrual rate P2 – Pension calculated on the basis of all earnings P2 – Pension calculated on the basis of all earnings If aar×N<30%, P=RR×30% aar decreases as RR increases C1- number of years of contributions up to 31/12/2006 C3- number of years of contributions up to 31/12/2001 Nmax=40 aarmax=2,3% (RR≤1.1 IAS) C2- number of years of contributions after 1/1/2007 C4- number of years of contributions after 1/1/2002 If aar×N<30%, P=RR×30% Nmax=40

Benefit Formulae


  • [*]
    Assistant Professor at the Faculdade de Economia da Universidade de Coimbra (Portugal).
  • [1]
    I am grateful for the helpful comments and suggestions provided by Michel Grignon and anonymous reviewers that have contributed to improve significantly the content of this paper. I assume all the responsibility for remaining errors.

The purpose of this paper is to analyze the pension reform introduced in Portugal in 2007. First, the paper makes a brief reference to the influence of European policies and orientations on the Portuguese pension policy agenda and examines the main legislative changes introduced. Next, the paper provides an assessment of the reform. The reform strategy followed is criticized on the basis of its adverse effects on pensioners’ living standards and also for having focused exclusively on the budgetary perspective, neglecting social objectives. The normative choices underlying the reform and the change it has introduced in the objectives and principles of policy are examined. Finally, the paper concludes that the reform was deeply transformative and that it could be described as “structural”.


La réforme des retraites adoptée par le Portugal en 2007 : une réforme structurelle

Cet article analyse la réforme des retraites mise en œuvre au Portugal en 2007. Il évoque d’abord brièvement l’influence qu’ont exercée les politiques et les orientations de l’Union européenne sur les mesures prises par le Portugal dans le domaine des retraites et examine les principaux changements introduits dans la législation. Il évalue ensuite la réforme et critique la stratégie adoptée en raison de son incidence négative sur le niveau de vie des retraités et parce qu’elle est exclusivement axée sur les aspects budgétaires, au détriment des objectifs sociaux. L’article examine les choix normatifs sur lesquels reposent la réforme et le changement qu’elle induit au niveau des objectifs et de la finalité de la politique des retraites, avant de conclure que cette réforme, porteuse de transformations profondes, peut être qualifiée de « structurelle ».


  • Atkinson A. B. (1999)The Economic Consequences of Rolling Back the Welfare State, Cambridge Mass.,The MIT Press.
  • Barry B. (1990), «The Welfare State versus the Relief of Poverty» in A. Ware, R.E. Goodin (ed.s) Needs and Welfare, London, Sage.
  • Castel N. (2009)La Retraite des syndicats. Revenu différé contre salaire continué, Paris, La Dispute/SNÉDIT.
  • Conselho Económico e Social (CES) (2006)Acordo sobre a Reforma da Segurança Social, CES – Comissão Permanente de Concertação Social, Lisboa, 10 de Outubro.
  • Creel, J. (2011) “Les politiques macroéconomiques : des mutations fortes”, in La pensée économique contemporaine, Cahiers français n°. 363. Paris, La documentation française, pp. 57-61.
  • Economic Policy Committee (EPC) (2001)Budgetary Challenges Posed by Ageing Populations:the impact on public spending on pensions, health and health care for the elderly and possible indicators of the long term sustainability of public finances, 24 October, EPC/ECFIN/655 – EN final, Brussels, European Commission.
  • European Commission (EC) (2003)Joint Report from the European Commission and the Council on Adequate and Sustainable Pensions, Brussels, European Commission.
  • European Commission (EC) (2005)National Strategy Report: adequate and sustainable pension systems, Report submitted in July 2005 to the European Commission, Brussels, European Commission.
  • European Commission (EC) (2006), Adequate and Sustainable Pensions: Synthesis Report,Luxembourg, Office for Official Publications of the European Communities.
  • Fitoussi J.-P. (2004), «Reform Proposals for the Stability and Growth Pact», European Parliament, Briefing paper for the Committee for Economic and Monetary Affairs, n° 2, April.
  • Fitoussi J.-P., Laurent E. (2009), «Macroeconomic and social policies in the EU 15: the last two decades», Documents de travail de l’ OFCE, n°2009-20, July.
  • Friot B. (2010)L’enjeu des retraites, Paris, La Dispute/SNÉDIT.
  • Hinrichs K., Lynch J. F. (2010), “Old-age Pensions”, in Castles F. G. et al. (ed.s) The Oxford Handbook of the Welfare State, Oxford, Oxford University Press.
  • Ministério das Finanças e da Administração Pública (MFAP) (2006)Orçamento de Estado para 2007: Relatório, Outubro, Lisboa, MFAP.
  • Ministério do Trabalho e da Solidariedade Social (MTSS) (2006a)Relatório Técnico sobre a Sustentabilidade da Segurança Social, Maio, Lisboa, MTSS.
  • Ministério do Trabalho e da Solidariedade Social (MTSS) (2006b)Linhas Estratégicas da Reforma da Segurança Social, 3 de Maio, Lisboa, MTSS.
  • Murteira M. C. (2008), «Pension Policies and Income Security in Retirement: A Critical Assessment of Recent Reforms in Portugal», Papeles de Trabajo del Instituto de Estudios Fiscales, n° 12/08.
  • Organisation for Economic Co-operation and Development (OECD) (2007)Pensions at a Glance: Public Policies Across OECD Countries, Paris, OECD.
  • Palier B. (2005), “Ambiguous Agreement, Cumulative Change: French Social Policy in the 1990s”, in Streeck W., Thelen K. (ed.s), Beyond Continuity. Institutional Change in Advanced Political Economies, Oxford, Oxford University Press.
  • Presidência do Conselho de Ministros (PCM) (2005)Programa do XVII Governo Constitucional, Lisboa.
  • Streeck W., Thelen K. (2005), “Introduction: Institutional Change in Advanced Political Economies”, in Streeck W., Thelen K. (ed.s), Beyond Continuity. Institutional Change in Advanced Political Economies, Oxford, Oxford University Press.
Maria Clara Murteira [*]
  • [*]
    Assistant Professor at the Faculdade de Economia da Universidade de Coimbra (Portugal).
This is the latest publication of the author on cairn.
Uploaded on on 25/07/2015
Distribution électronique pour La Doc. française © La Doc. française. Tous droits réservés pour tous pays. Il est interdit, sauf accord préalable et écrit de l’éditeur, de reproduire (notamment par photocopie) partiellement ou totalement le présent article, de le stocker dans une banque de données ou de le communiquer au public sous quelque forme et de quelque manière que ce soit.
Loading... Please wait