Pension systems are currently under radical reconstruction, and in most European countries there is a shift from defined benefit (DB) to defined contribution (DC) schemes.
The main rationale for the change is to eliminate financial incentives to early withdrawal from the labour market. It used to be the case that when workers in DB schemes reached retirement age, they were (actuarially) indifferent between retiring and working longer. In DC schemes, on the other hand, the decision to stay in work has a double effect: as well as accumulating greater pension wealth, the shorter life expectancy in retirement results in higher pension benefits. Workers have a strong financial incentive therefore to postpone retirement.
The shift from DB to DC systems should also extend freedom when it comes to deciding when to retire. Policy-makers do not have to compel citizens to work longer, but rather create incentives and opportunities to postpone retirement. This is what has happened, for example, in Poland. Although the statutory retirement age has been decreased, the general paradigm of retirement has changed. It is now understood that the attainment of retirement age does not automatically signal retirement. It allows individuals to access their retirement pension, although it is advisable (and of course permitted) to work for longer. Individuals have to make their own decision about the timing of retirement in the light of their own needs, preferences and opportunities.
Bounded rationality of retirement decisions
Free choice in retirement decision-making is based on the key assumption of classical economics: human rationality. Individuals act so as to maximize expected utility. In deciding whether or not to retire (now or later), they review their financial needs, health status, work capacity, and preferences for leisure in order to identify the optimal timing of retirement decision. But do they?
There is a lot of evidence that people hardly ever make retirement decisions in this way. They usually opt to retire as soon as it is possible. A recent German study, for example, reported that less than 10% of respondents were planning to work after reaching statutory retirement age - despite financial incentives and the availability of flexible retirement. The percentages, furthermore, are lower among woman and people with low education attainment. Since both of those groups typically have lower pension entitlements, the opportunity to accumulate more pension wealth should be especially attractive to them. The evidence suggests - and it is available from other countries also - that people may retire before the optimal age (for them).
Even commonsense thinking might lead us to regard retirement decisions as puzzling. Over the last few decades life expectancy has increased significantly. People not only live longer, but also remain active and healthy for much longer, so they are more able to work in later life. The nature of work has changed in ways that should (?) make it easier for older workers to continue in employment. Despite these changes, the effective retirement age has decreased. We are able to work for longer (and there are clear benefits attached), but choose not to do so. We make suboptimal retirement decisions.
This tendency to make suboptimal decisions has been called "bounded rationality". Investigating its sources is a major part of economic research these days. Generally, they put into one of three categories. Firstly, there are factors that affect self-control and the capacity to postpone present consumption for greater future consumption. Some people are impatient, or myopic when faced with the temptation of ‘premature’ consumption. Secondly, there is imperfect knowledge. People may lack appropriate information about the financial consequences of their decisions (financial illiteracy) - though solid knowledge in finance does not protect individuals from irrational decisions. Thirdly, there are cognitive biases.
Cognitive biases and retirement decisions
Cognitive biases are systematic errors in thinking that form an integral part of our psychological make-up that influence judgment and decision-making under conditions of uncertainty: they lead people not to choose the optimal option in decision situations (providing highest expected value). Moreover, they display regularities that allow us to investigate them empirically.
Behavioural economics provides a new perspective on retirement decisions. More and more researchers are looking at retirement decisions in the light of possible cognitive biases. What non-financial factors (not incorporated in traditional economic models) are relevant to retirement decision-making? So, for example, classical life-cycle theory has been expanded to include non-financial determinants to develop the behavioural life-cycle (BLC) hypothesis.
Affective forecasting is a form of cognitive bias that seems particularly relevant to retirement decisions. When we think about how the future will turn out, our imaginings are often emotionally charged as this allows us to anticipate how we would feel if we chose one line of action rather than another. The problem is that what we imagine will happen is typically more extreme (better or worse) than what actually happens. We exaggerate how much better off we will be if things turn out well (and how much worse off we will be if they turn out badly). This leads to suboptimal decisions. A tendency to exaggerate the unpleasantness of work as well as the attractions of retirement makes people more willing to retire early.
Also relevant to retirement decisions is hyperbolic discounting. In standard discounting models, individuals discount future at a constant rate. When hyperbolic discounting comes into play, preferences for future consumption change as the future comes closer. People are able to wait for a long time if the reward depends on it, but if they are faced with a choice between a smaller-sooner reward and larger-later one, they are usually reluctant to wait. So when retirement is a long way away, workers usually plan to work far beyond retirement age, but as it gets closer, their preferences change. The option of early withdrawal starts to dominate over long-term financial wellbeing. The closer individuals are to retirement, the more they are ready to sacrifice longer-term financial gain to get benefits earlier.
The way of presenting information also has a huge impact on the process of decision making. This is called the framing effect. As people tend to compare options in relative rather than absolute measures, they use some available anchor as a reference point, to compare it with analyzed options and to assess them in terms of gains or losses. Moreover, natural loss aversion leads individuals to choose those options which allow them to avoid loss. The influence of framing on retirement choices was highlighted in a social experiment that presented individuals with different default retirement ages, so that retirement options were framed as gains or losses from these different reference points.
My own research focuses on a certainty effect. Preliminary results show a strong influence of pension system predictability on retirement behavior. In countries with relatively stable pension systems, people are more willing to postpone retirement until after the official retirement age, whereas in countries with unstable pension systems, people usually retire as soon as it is possible. The instability of the pension system creates uncertainty about future regulations; and in such case individuals choose what they see as the safer bet, which is early retirement with stated benefits, even if it provides lower expected value (is less beneficial overall) than the decision to postpone retiremen
To summarize, despite growing interest in the behavioural aspects of retirement decisions, the field still remains poorly recognized. Much has been theorized, but empirical evidence is still limited. There are many important issues to explore in order to pin down the different psychological factors that influence retirement decisions. There is a growing need (in science and policy) to investigate the sources of bounded rationality in this domain of life.
About the Author
Dr Lukasz Jurek was an Academic Visitor at the Oxford Institute of Population Ageing in 2019.
Lukasz Jurek is an economist and gerontologist. He is an Assistant Professor in the Department of Sociology and Social Policy at Wroclaw University of Economics, Poland.
Comments Welcome
We welcome your comments on this or any of the Institute's blog posts. Please feel free to email comments to be posted on your behalf to administrator@ageing.ox.ac.uk or use the Disqus facility linked below.
Opinions of the blogger is their own and not endorsed by the Institute
Comments Welcome: We welcome your comments on this or any of the Institute's blog posts. Please feel free to email comments to be posted on your behalf to administrator@ageing.ox.ac.uk or use the Disqus facility linked below.
