Back to search

VAM-Velferd, arbeid og migrasjon

Coping with flexibility. Behavioral adaptations to the new pension system and their consequences in a lifetime perspective

Alternative title: Håndtering av fleksibilitet. Adferdsmessige tilpasninger til det nye pensjonssystemet og deres konsekvenser.

Awarded: NOK 4.0 mill.

The aim of the "Coping..."-project has been to analyze how older workers in Norway adapt to the flexibility in the reformed (post-2011) pension system. In particular, we look into what may look like a paradox: that the proportion who start early take-up increases, while the proportion who actually withdraw from employment early decreases. In other words, a new practice is developing where it is common to combine pension take-up and paid employment for some years after age 62. This option is particularly popular among men. The aim of this project has been, first, to better understand how older workers reason around these issues, and second, to investigate how they administer the extra assets they gain access to in this period. The project has been a cooperation between the Institute for social research, Statistics Norway, and research institutes NOVA and Fafo. It has been organised in three modules. Module A and C have aimed to generate original research, while the aim of module B has been to update a data set on individuals' access to workplace-based pensions (AFP and occupational schemes). Module A is based on analyses of register data. In the first analyses, we have looked into the high proportion who chose to take up pension as early as possible, while still working full time. This popular option will, in most cases, be associated with a smaller or larger loss of lifetime income. We start out by investigating whether early pension claiming can be explained by observable factors like life-expectancy based on socioeconomic differences, wealth accumulation and aspects of the family situation. Life-expectancy is estimated within the micro-simulation model, MOSART. MOSART is further used to compare the value of lifetime incomes given alternative choices with respect to the timing of pension claiming. By comparing the actual lifetime income of early claimers with the optional value that they could have optioned by postponing the claiming of pensions, we can derive a measure of their time preferences, i.e. the degree to which they are prone to give preference to their immediate over their long-term economic welfare. Finally, we use machine learning to disentangle which of these factors (observed characteristics versus unobserved preferences) that seem to take precedence. The results show that individuals have a higher probability of starting pension claiming early if they are married, have grown-up children, high debt, high housing wealth, and a short remaining life-expectancy. The correlation between the probability to start pension claiming early and the optional value of postponing retirement is particularly strong in the few cases where the early pension claiming is profitable also in a lifetime perspective. Nevertheless, there remains a substantial unexplained variation in pension claiming behaviour that can be attributed to (unobserved) differences in time preferences and hence in the propensity to give priority to the present at the expense of the future. It is well established in existing research that many individuals are prone to reject offers to buy annuity on favourable terms even if this would appear to be objectively rational given neutral time preferences. This phenomenon is known as the 'annuity paradox'. We conclude that the propensity to start taking out old age pensions early among those who continue working after age 62 is driven by the same mechanisms that drive the more general annuity paradox. In module B, we have updated and improved a set of microdata on access to workplace-based pensions in the private sector. This dataset is not finalized and ready for use. Within module C we are producing three articles all based on qualitative interviews that are focused on different aspects of the reasons individuals give for their behavioural adaptations to the new pension system. 28 interviews with private-sector employees aged 55-66 were conducted in October and November 2019. The first article addresses the same thematic as the above mentioned article from Modul A by investigating employees' reasons for drawing / not drawing old age pension while still in employment, and what those who combine wages and pension do with the extra liquidity they receive. The second article examines variations in planned retirement timing, as well as links between timing and aspirations for life in retirement. Article number three explores how older workers relate to the public rhetoric about longer careers, and how they balance the expectation of extended working lives against other values and their own preferences.

Hovedmålet i prosjektet har vært å få en bedre forståelse av hvordan eldre arbeidstakere tilpasser seg fleksibiliteten i det nye pensjonssystemet. Gjennom analyser av registerdata og kvalitative intervjuer har vi skapt ny innsikt, samtidig som vi har bygget opp et nytt datasett som vil gjøre originale analyser mulige i nær framtid. Resultatene er løpende formidlet til aktørene i feltet, blant annet gjennom seminarer hos Senter for seniorpolitikk og i Arbeids- og sosialdepartementet. Prosjektet har dermed bidratt til en mer informert, og mer nyansert, diskusjon om det nye pensjonssystemet. Formidlings- og publiseringsaktiviteten vil fortsette ut over prosjektperioden.

Module A contains three studies using register data and data from the micro-simulation model MOSART: a study of the revealed tendency, since the reform took force in 2011, for early take up of pension benefits in combination with continued labor-force participation, a study of generational changes in savings behavior, and finally a study of the accrual of pension rights and projected pension benefits among different categories of immigrants. Module B will consist of updating and analyzing a unique database combining longitudinal simulations of individual employment careers with data on the accrual of occupational pension rights and participation in the AFP-scheme. The data will be used to perform an up-to-date analysis of the interplay between the three sources of pension rights among Norwegian employees and its distributive implications. Finally, Module C is devoted to a study of subjective retirement planning in terms of expectations, preferences and behavioral dispositions among individuals approaching the early retirement age at 62 and individuals in the age group 62-67 who have made different choices concerning pension claiming and labor market withdrawal. This module offers a subjective and qualitative take on some of the same issues that are addressed in the two first sub-studies of module A and module B.

Publications from Cristin

No publications found

No publications found

Funding scheme:

VAM-Velferd, arbeid og migrasjon