In preparation of the 2018 Pension Adequacy Report by the European Commission and the Social Protection Committee, teams from Belgium, Sweden and Italy use their microsimulation models to simulate the possible developments of pension adequacy while taking into account the set of economic and demographic projections developed by the AWG. The results of this exercise complement the AWG simulations of pension expenditures in a context of demographic. The results described in detail in this report were summarised in section 5.1.2. of the 2018 Pension Adequacy Report.
Pension cost takes up a large part of public expenditures in the EU member states (European Commission, 2015b, Graph 3, page 5), so pensions and pension reforms are a point of focus of the Ageing Working Group (AWG). The Ageing Report wants to map the prospective development of public (pension) expenditures under an “unchanged policy scenario” and identify the main drivers of these expenditures, including demography, labour market developments, and (the reform of) eligibility conditions and benefit formulas in public pensions and other age-related social security systems.
However, the current and prospective adequacy of social security benefits including pensions is also a dimension of social protection that requires and gets attention on the EU level. The Social Protection Committee (SPC) monitors social conditions in the EU. The SPC mandates the SPC Working Group on Ageing Issues (SPC WG-AGE) to prepare a “2018 Pension Adequacy Report” to examine current and projected pension adequacy trends.
Besides using AWG projections on the benefit ratios and extensive analysis of Theoretical Replacement Rates (TRR), the SPC WG-AGE already in its 2012 PAR recognised it could not assess the “extent to which pension systems in Member States will contribute to the goal of reducing the number of people exposed to poverty or social exclusion” (EC, 2012, 137), to add that “if all Member States were able to apply dynamic micro-simulation models to this task, likely scenarios which could offer guidance to policy makers could be constructed.” (op. cit., 138). In preparation of the 2018 PAR, microsimulation teams from Belgium, Sweden and Italy use their dynamic microsimulation models to simulate the possible developments of pension adequacy while taking into account the set of economic and demographic projections developed by the AWG, as well as the joint assumption of unchanged policy besides already legislated pension reform. As such, the results of this exercise allow to complement the AWG simulations of pension expenditures in a context of demographic ageing by projections on pension adequacy. The results described in detail in this report were summarised in the section 5.1.2. of the 2018 Pension
Contrary to the other countries, the cost of pensions shows a continuous increase in Belgium. This is because demographic ageing is stronger than the cost-reducing effects of, among other things, the increasing employment rate and the increase of the average exit age. The latter developments however result in a higher pension benefit after retirement, thus bringing down the poverty risk among pensioners. In Italy and relative to GDP, pension spending would remain stable at first, then increase until about 2040, after which a strong decrease would set in again. The increase before 2040 would be the result of low productivity growth and the transition of large cohorts into retirement. The subsequent decrease, however, would be the result of the gradual replacement of the stock of pensioners that had earned a benefit under the old Defined Benefit regime by those that have a benefit under the Notionally Defined Contribution regime (NDC) that was implemented in the 1995 and 2011 reforms. The less generous (particularly to intermittent careers) NDC regime would drive the benefit ratio down and may push the poverty risk up.
Gross pension expenditure in Sweden is expected to remain roughly constant relative to GDP, among other things driven by comparably high net immigration. However, the actuarial correction for longevity produced by the adoption of the NDC system in 1996 would drive the benefit ratio down. This decrease would be reinforced by the AWG assumption of a constant labour market exit age and would result in a considerable increase of the poverty risk among the elderly. While Italy has also adopted an NDC scheme, the contemporary implementation of an automatic adjustment of retirement ages to increases in life expectancy is set to counterbalance most of the effect of the actuarial correction. In any event, the performance of the labour market will be crucial in assuring longer, stable careers to older workers, essential to adequate levels of pension in an NDC scheme.
Although the three models used in this project differ in scope and size and have not been developed with a focus on international comparison, this project demonstrates how dynamic microsimulation can be used to bridge the gap between the assessment of pension sustainability and adequacy in comparative perspective. As such, it shows the potential of these simulation techniques in Europe.