Skip to ContentSkip to Navigation
About us Latest news News News articles

Lower income groups less well prepared for pension reductions

21 January 2013

People on lower incomes are less well prepared for their old-age pension than people in higher income groups. While the more wealthy are saving more because they expect a lower pension, people who have less to spend are not making such savings. This has been empirically demonstrated for the first time by Peter van Santen who will be awarded a PhD by the Faculty of Economics and Business of the University of Groningen on 28 January.

Peter van Santen’s PhD research confirms the hypothesis that lower income groups set no extra money aside for the period after their retirement due to both ignorance and because pension is something relatively less important to them. The failure to make these savings makes lower income groups more vulnerable to changes in the pension system. ‘The better-off save more to be able to maintain their pattern of spending at as constant a level as possible after retirement’, says Van Santen. Households on lower incomes are taking a greater risk of becoming worse off after reaching the age of 65.


Given that the end of the economic crisis is by no means in sight, and the funding ratios of pension funds have not been at the required level for years (to such an extent that pension payments are being curtailed or frozen), employees need to take steps to maintain the level of their future income. Uncertainty about the seriousness of the future situation plus the knowledge that income after retirement will in any event be less, leads to precautionary saving.

‘Saving is always based on a future expectation’, explains Van Santen. ‘To be able to establish empirically how great this savings response is in relation to income we asked households to indicate what their expected pension income will be as a percentage of their present income. This showed that, on average, people expect to have roughly 75% of their present income, with a standard deviation of 15%. So only a minority expect their income to remain unchanged.’


Since 2007 households have started to lower their pension expectations and they have become more uncertain. It also appears that in response to this the higher income groups in particular have started to put more money aside, also because of an increased life expectancy. Peter van Santen says that this is not surprising. ‘One explanation may be that poor households cannot afford to save. But in the Netherlands there is also the consideration that lower income groups don’t need to prepare themselves as much because the state old-age pension is not income-related. Nevertheless, the lack of a savings response by this group makes them more vulnerable to changes in the pension system.’


For employed people it is increasingly a question of taking matters into your own hands, because the certainty of a previously defined pension entitlement is disappearing. According to Van Santen, the Netherlands is moving towards a system in which a pension fund operates more like an upgraded investment fund, as has long been the case in the United States, for example. He concludes in his study that the better-off also take out more annuities, such as annuity insurance policies and single-premium policies, than poorer households. ‘Which is not surprising, because economists have already established that these are the most rational ways of saving for the future.’

Measurement errors

According to Peter van Santen, surveys conducted among the less well educated and less well-off households generally provide unreliable results about savings. Often it turns out that people in the lower socio-economic classes are not aware that they have bought some form of annuity or think that they have one while this is not so. ‘We think that there is a wide margin of error: we estimate that 32% of holders incorrectly enter that they do not have any annuity insurance or a single-premium policy. The reverse is true for 12%: they state that they do have an annuity or single-premium insurance while we believe that this is not the case. These wide deviations are subject to statistical uncertainty, but in absolute terms, this would increase the percentage of holders from 32% to 56%.’ Van Santen has corrected the results of his study to take this deviation into account.

Southern Europe

Van Santen has calculated that, on average, 48 to 60 euro cents extra are saved for every euro that the expected pension assets decline. Within Europe this compensation varies from 15 euro cents in Southern Europe to more than 90 euro cents in the better off northern countries. The adverse impact of a lower pension will therefore be partly offset but in Southern Europe, in particular, the savings will be insufficient to guarantee the same living standards. ‘I expect that pension reforms in Greece, Italy and Spain will have the most disastrous effects, owing to the failure to make the necessary savings.’

Curriculum Vitae

Peter van Santen (Dordrecht, 1985) studied Economics & Law at Utrecht University. He did his PhD research at the Institute for Economics, Econometrics and Finance at the University of Groningen. The title of his thesis is ‘Precautionary saving, wealth accumulation and pensions. An empirical microeconomic perspective’. His supervisor is Prof. R. Alessie and his co-supervisor is Dr A. Kalwij. Peter van Santen has been working as a researcher at the central bank of Sweden in Stockholm since September 2012.

Last modified:13 March 2020 02.18 a.m.
printView this page in: Nederlands

More news