Old-Age Pension Systems: Characterization and Comparability

The main objective of a contributive scheme is to replace the income of those who leave the market for reasons of age and is financed by the participants (current or former participants). On the other hand, non-contributive systems usually take the form of public programmes for the alleviation of poverty, invalidity or sickness in old age and their financing should come from general taxation not linked to the labour market.

According to Valdés-Prieto (2002), the “degree of funding is obtained dividing the market value of the pension fund by the expected discount value of accrued liabilities on the same date”.

In Portugal, old-age individuals with at least 15 years of earnings are entitled to an earnings-related pension (contributory system). The calculation of pensions is based on the number of contribution years and the average wage of the employee (however, there is no actuarial analysis to calculate pension benefits, with the exception of the so-called sustainability factor where the pension benefit has been adjusted in accordance with the evolution of average life expectancy of the Portuguese population since 2007).

An example of an NDC system is the Swedish model where part of the contributions of employees/firms are transferred to an account of the beneficiary (abstract or notional account), which is capitalized over time through a composite capitalization rate (incorporating, among other variables, the growth rate of the economy). The accumulated value (Notional) is transformed at retirement date into an annuity according to actuarial rules prevailing at the time—Defined Contribution systems. The payment of the annuity is made through the contributions of employees/employer in business—a PAYGO system.

The Swedish system of notional accounts has a buffer fund that is financed by the surplus between contributions and pension disbursements and charged in the opposite situation. This Fund consists of financial assets that are invested in capital markets and therefore are credited or debited depending on their gains or losses, respectively (The Portuguese system also has a buffer fund).

In his model, Samuelson (1958) shows that, if the population grows at a rate of m, the system is able to pay m interest (‘biological’ interest rate) and pensions can increase accordingly, thus supporting the necessary condition, ceteris paribus, for a PAYGO model to be sustainable. According to Banyár (2016) the solution to the longevity problem can be solved by “the regular indexation of the pensionable age” (This has already been introduced in several countries and was also recommended by the European Commission in its Pension White Paper).

An equivalent conclusion could be reached with a PAYGO system with defined benefits if we assume actuarial fairness. In fact, the impact of a reduction in the discount rate on the value of contributions during the working period is “equivalent” to the effect of a future deterioration in the relationship between the number of employees and the number of pensioners, ceteris paribus. On the other hand, a future deterioration in the relationship between the number of employees and the number of pensioners, ceteris paribus, will create an implied debt in the system that in the future will correspond to real debt. This system’s implied debt can be compensated for by an increase in current contributions which can be used to reduce the current public debt, and consequently, reduce future public debt.

Individual A it is a real pensioner of the Portuguese social security system who retired on 26 August 2007. The total amount of remuneration received by the worker during his/her 41-year working life (adjusted for inflation) corresponded to €1,003,662.02 (€1,002,244.85 during the last 40 years).

P₁ ≤ 12 × IAS

In Portugal the overall contributory rate is 34.75 percentage points (p.p.) (23.75 p.p. for employees and 11 p.p. for employers) and the old-age risk component is 20.1 p.p. (excluding administrative costs), according to Law No. 110/2009, 16 September. The old-age risk component, excluding administrative costs, was 13.34 p.p. between 1993 and 1998 (Decree Law No. 326/93, 25 September), and 15.57 p.p. between 1999 and 2009 (Decree Law No. 200/99, 8 June. It should be noted that before 2000 there was no breakdown of the overall rate. In this context, we will assume in our example that the contributions to the system (old-age risk component) correspond to 13.34 p.p. of the wage between 1967 and 1998 and 15.57 p.p. between 1999 and 2007.

It is assumed that the discount rate is equal to the monetary adjustment factor published annually by the Portuguese Government.

The capitalized value of contributions at retirement age is €142,433. The retirement age of the pensioner was 65 years of age.

Assuming a global contribution rate of 34.75 p.p., the capitalized value of the contribution at retirement age is €348,773.

If we assume that the annual discount rate is 2% (4%), the annual pension benefits in a funded actuarial system with a defined contribution will be €21,325 (€26,263).

According to Banyár (2016), assuming that the PAYGO system is still valid, there are several possibilities of reform which, however, have several limitations:

According to Brimblecombe and McClanahan (2019), “recent reforms have increased the number of years required for entitlement to a full and/or minimum social security pension”, and it is the case that “Meeting these requirements is particularly challenging for women workers who spend on average significantly less time in the workforce than their male counterparts”.

According to Brimblecombe and McClanahan (2019) “the introduction of defined contribution systems (notional or funded) for first pillar (social security) provision has penalized women beneficiaries in a number of ways. Most significantly, the lower salary and greater incidence of part time working amongst women translates directly into lower pension entitlements at retirement”.

In this context, Brimblecombe and McClanahan (2019) argue that “policymakers should therefore assess systems on the distribution of outcomes rather than average outcomes”.