Impacto das mudanças na composição familiar brasileira sobre o resultado atuarial dos RPPS: um estudo aplicado aos benefícios de pensão por morte de um RPPS paraibano

Abstract

The “probability of being married” assumption directly influences actuarial calculations in pension schemes, especially with regard to survivor benefits. This study, therefore, analyzed the impact of the “probability of being married” assumption on programmed retirement benefits with reversion to a survivor's pension, as well as on the death benefit granted to the spouse of an active public servant, within a Public Pension Scheme (RPPS) in the state of Paraíba, Brazil. To this end, data from a Paraíba RPPS, comprising 2,001 active participants, was used. Smoothed and aggravated variations of 7%, 14%, and 21% were applied to the “probability of being married” assumption used in the RPPS’s actuarial calculations. The Present Value of Future Benefits (PVFB), the Normal Cost, and the Mathematical Reserve for Benefits to Be Granted were calculated for the baseline scenario as well as for the adjusted scenarios. The results show that, in the case of the programmed retirement benefit with survivor pension reversion, variations in the “probability of being married” assumption had minimal impact on actuarial indicators. The maximum difference in Normal Cost was 0.37%, and in the Mathematical Reserve, approximately R$ 3.1 million between the extreme scenarios analyzed. However, in the case of the death benefit for the spouse of an active public servant, greater sensitivity was observed, with proportional variations of 7% in the Normal Cost for each 7% increase or decrease in the assumption. Even so, when considering both benefits jointly, it was found that a 7% change (either increase or decrease) in the “probability of being married” assumption impacted the total nominal Normal Cost by only 1.6%, due to the greater actuarial weight of the retirement benefit within the overall benefit plan.