Abstract Both insufficient saving and early withdrawal of retirement savings jeopardize financial well-being in old age, but the latter is rarely studied. We examined which individual differences relate to both regular saving and early withdrawal decisions. These decisions are economically similar (accumulating retirement assets) but psychologically different (incurring regular small losses vs. abstaining from a one-off large gain). We analyzed three representative surveys conducted in Estonia in 2020, 2021, and 2022, surrounding a policy reform that introduced an early withdrawal option to a previously mandatory pension saving scheme. Limited trust in financial institutions and non-Estonian nationality emerged as statistically and economically significant factors associated with the pair of decisions both not to save and to withdraw early. Fostering institutional trust represents a key strategic goal for pension systems. Our findings also call for caution with pension freedom reforms, which may put those who already save too little in double jeopardy.

Insufficient retirement savings threaten financial security worldwide as individuals increasingly bear responsibility for funding their retirement. We present the first population-level test of behaviourally informed messaging for boosting saving in retirement accounts. In collaboration with the Estonian Ministry of Finance, we conducted a preregistered randomized controlled trial among all eligible account holders (N = 127,974) testing whether email reminders about tax benefits boost voluntary pension contributions. Across nine behaviourally designed reminders versus a no-reminder control, reminders increased contribution likelihood by 10.49% and raised average contributions by 14.75% within one week before the deadline. Most framings increased participation but notably, a family-security message also increased individual contribution size (from €868 to €1,009). Exploratory causal forest analysis revealed the baseline message most strongly influenced younger, higher-income individuals with prior contributions, while the family-security message had the largest impact on high-income individuals. The intervention generated ~€1.2 million in additional retirement savings within one week. These results show that simple, low-cost behavioural reminders can meaningfully amplify tax incentives at scale, offering practical tools for policymakers.

We study how cognitive constraints relate to each distinct step of the planning and execution process for retirement, that is, individuals’ propensity to plan, savings goals set, and economic outcomes (wealth accumulation and portfolio choice). We find that different cognitive constraints play distinct roles: Higher advanced financial literacy (and quantitative reasoning ability) predicts a greater propensity to plan, while higher basic financial literacy and verbal cognition predict setting higher savings goals. Math-related abilities are not associated with savings goals in a systematic way. Furthermore, our evidence shows that the economic consequences of retirement planning depend on the earlier set savings goals. In comparison to non-planners, only planners with a higher savings goal (above the median) accumulate more wealth and are more likely to hold risky assets and private annuities. Our findings suggest that when crafting public policy to develop individuals’ retirement readiness, next to improving financial literacy, other targets could be to enhance cognitive skills and to support setting concrete savings goals by, for example, providing better access to planning relevant information and tools.

Subjective financial scarcity poses a significant concern that negatively impacts individuals' wellbeing. With attention tunneling to present financial worries, individuals might neglect their future financial situation, even if they objectively have enough funds to save. Such behavior can contribute to a deficient financial situation in retirement. To assess the impact of subjective financial scarcity on the intention to save for retirement, we conduct an online vignette survey experiment (n = 134). Using the two-limit tobit model, we find that subjective financial scarcity leads to lower retirement savings rate. We contribute to the literature by testing theoretical predictions of scarcity theory, providing experimental evidence for the myopic financial behavior orientation of retirement saving rates. We offer practical implications for policymakers, suggesting that interventions that promote saving for retirement should be designed with subjective financial scarcity and stress reduction messages in mind.

Researchers have long been trying to understand why individuals dislike annuities. Here, we investigate if the process individuals use to assess the financial value of annuities may lead them to inaccurately value annuities. In Study 1, participants were asked to assess the monthly payments associated with a specific annuity lump sum or the annuity lump sum associated with a specific monthly payment. They were then asked to describe how they arrived at their answers. We find that when making this assessment, 42% of participants report attempts at using math, with some even describing mathematical formulas. Most other participants reported guessing instead. Reporting attempts at math is more common among participants with higher financial literacy and numeracy. Reported attempts at math, financial literacy, and numeracy predict arriving at more realistic financial values for annuities, as well as incorporating assessments of life expectancy in the math. Based on this process knowledge, we then designed an experiment in Study 2 and tested the effect of presenting information about life expectancy, providing feedback about payouts or their combination. We find that we can thereby change the assessed financial value of annuities and increase participants' interest in annuities, especially among participants that reported attempts at using math. Understanding the processes individuals use to assess the value of annuities informs theory and practice.

PurposeHow did investors in the Swedish Premium Pension System (PPS) react to the stock market shock ignited by the COVID-19 pandemic?Design/methodology/approachThe authors use fund-level data from the Swedish Pensions Agency on investment choices in the PPS. For each fund, the authors use monthly information on the number of investors and holdings' market value up to November 2020. The authors also use information on the total number of portfolio changes per day. For analyzing whether PPS investors reacted to the pandemic with claiming their pension, the authors use monthly data on the number of investors of a certain age group who initiate their public pension payment.FindingsTrades more than doubled, and shifted capital from equity funds to low risk interest funds. In economic terms, however, trading stayed at low levels–less than two percent of investors traded in March 2020 and there was no effect on pension withdrawals. The increased trading during the market tumult was disproportionately concentrated among investors in the top of the pension capital distribution.Research limitations/implicationsWith fund-level data, the authors cannot investigate what in particular made retirement investors stay calm in the midst of a severe market decline. Either, those investors have a long-term investment horizon as they save for their pension or particular features of the system's choice architecture induce inertia and discourage from trading. The sub-group analyses are more consistent with the explanation that PPS-induced inertia is responsible for the relatively small increase in trading activity, but future research could exploit individual level data to explore this in more detail.Practical implicationsThe often-criticized PPS choice architecture provided positive side effects in times of a severe market shock by shielding retail investors from committing trading mistakes when trying to outsmart the market.Originality/valueThe study complements previous evidence on the effects of COVID-19 on investor activity. The small response of PPS investors to COVID-19 is in line with earlier US findings on 401(k) accounts during the 2007 financial crisis (Tang et al., 2012) and industry reports about the COVID-19 period (see, e.g. Mitchell, 2020). The authors find no effects at all on public pension withdrawals in Sweden, while evidence from US 401(k) plans indicates a small share of workers taking COVID-related early withdrawals.

We study how age identity (measured by the difference between chronological age and perceived old age), influences financial planning among the older population (60+) in China. Using data from three waves of the China Longitudinal Aging Social Survey, we show that individuals who feel younger have a significantly higher probability of making financial plans. That such an effect exists in sub-samples divided by age and retirement status implies the relevance of financial planning even for individuals with advanced ages. It is consistent with the hypothesis that old individuals who feel younger have higher perceived cognitive abilities and hence higher motivation to make financial plans. However, an unfavorable perception of social aging culture moderates such a positive effect. Age identity can further impact the downstream economic be-haviors of saving and investing, either directly or indirectly, through financial planning. Finally, a younger age identity also increases an individual's willingness to internalize the responsibility of eldercare. Our findings imply that it is important to consider individuals' age identity when crafting and implementing old-age policies.

PurposeThe authors develop and validate a conceptual model, the retirement engagement model (REM), to understand the relationships between behavioral engagement (retirement information search), cognitive factors and engagement (e.g. beliefs and financial knowledge), emotional engagement (e.g. anxiety), and socio-demographic factors. Approach: The authors derive the REM through a three-step procedure: (1) an extensive literature review, (2) interactive feedback sessions with experts to confirm the model's academic and managerial relevance, and (3) an empirical test of the REM with field data (N = 583). The authors use a partial least squares (PLS) structural equation model and examine heterogeneity through a finite mixture model.Design/methodology/approachAround the globe, people are insufficiently engaged with retirement planning. The customer engagement literature offers rich insights into antecedents, outcomes, and barriers to engagement. However, customer engagement literature lacks insights into cognitive, emotional and behavioral factors that drive engagement in retirement planning, a utilitarian service context, which is important for financial well-being.FindingsBeliefs such as perceived susceptibility, severity, benefits, barriers, and self-efficacy, together with trust and retirement anxiety, explain people's search for pension information. These factors can be used to define three clear, actionable segments of consumers.Originality/valueThe findings advance the customer engagement and transformative service research literature by generating insights on engagement with retirement planning, a utilitarian rather than hedonic service context that is especially relevant for financial well-being. The findings inform managerial practice and emphasize the relevance of including cognitive and emotional engagement factors that trigger behavioral engagement. The REM can help to improve pension communication. For example, the results indicate that marketers should stress the benefits of, rather than the barriers to, acquiring information.

How did investors in the Swedish Premium Pension System (PPS) react to the stock markets shock ignited in 2020 by the COVID-19 pandemic? The share of investors that traded more than doubled, and trades shifted capital from equity funds to low risk interest funds. In economic terms, however, trading activity stayed at very low levels—less than two percent of investors traded in March 2020 and there was no effect on pension withdrawals. Given the vast evidence on retail investors’ strongly increasing trading volume in crisis times, the reaction of PPS investors looks surprisingly smart, i.e., avoiding the many mistakes that investors incur when they try to outsmart the market. Potentially, the often-criticized choice architecture of the PPS that induces strong inertia provided positive side effects in times of a severe market shock.

People spend very little time planning for retirement, which could have negative effects on their financial well-being. To address this troubling lack of engagement, the authors posit that the use of goal framing, a marketing practice that involves making strategic adjustments to wording of marketing communications, in technology-facilitated communication (e.g., email) is effective for stimulating consumers’ behavioral engagement with pension information that is relevant for their long-term financial well-being. Field, online, and laboratory studies consistently show that a prevention-oriented assurance frame in technology-facilitated communication is twice as effective as a promotion-oriented investment frame for increasing participants’ engagement behavior. The findings have important implications for marketers and policy makers who seek to increase consumers’ retirement engagement behavior and financial well-being.

How do individuals approach financial decisions? Do they apply sophisticated formulas, 'back of the envelope'-math, do they guess, or something completely different? We show that the verbal protocol method – i.e., having individuals explain their approaches – generates new insights to answer these questions. Our findings on approaches discovered are reliable within sample and have predictive power out of sample. Among other things, we find that when assessing the value of an annuity, 40% of individuals use some math, and some even use formulas similar to actuaries; the other 60% seem to guess. Valuation approaches used predict valuation outcomes and valuation precision, explain puzzling findings reported in earlier literature, and predict a behavioral reaction to an anchoring intervention designed to match one of the approaches discovered. The verbal protocol method opens new pathways to understand financial decision-making as well as to design effective policy interventions.

Building on recent findings in psychology, we study the impact of subjective age identity (feeling younger or older than one's chronological age) on economic behaviors. Using data from the Health and Retirement Study we find: Individuals with a younger age identity have higher work engagement, and their savings profile, as a function of the subjective age gap, is hump-shaped. The effects are economically significant, for example, increasing the subjective age gap by one standard deviation increases an individual's likelihood to be employed in a subsequent HRS wave by 1.1% (about 21% of the conditional mean). The relationships found are consistent with an interplay of two subjective age channels: Ability (self-perceived abilities to perform certain economic behaviors) and Preference (choosing (avoiding) “young” (“old”) behaviors). Our results have implications for policy and financial advice that traditionally target individuals based on chronological age. That is, for example, allowing more flexibility with respect to retirement decisions as well as aligning financial products and services with subjective age identities.

PurposePeople around the world are not sufficiently capable or willing to engage in retirement planning. New technological tools have been proposed as a promising solution to foster involvement and consequently encourage retirement planning. This paper aims to test whether an interactive online pension planner can improve participants’ behaviour, behavioural intentions, attitude, knowledge and perceived ease of use, usefulness and enjoyment.Design/methodology/approachIn collaboration with a company specialised in technologically advanced pension planners, three different versions of an online pension planner were created. The control condition only allowed participants to check their pension situation and the composition of future retirement income. In the medium interactivity level, participants could choose to modify certain variables affecting their pension income, on top of the features from the control. The highly interactive planner additionally included an interactive budget tool and showed whether the accumulated pension income was sufficient to cover the desired spending. Data were collected with the help of an online panel (N = 285).FindingsThis paper finds a positive effect of interactivity on behaviour within the planner, that is, the number of clicked options, as well as on participants’ intention to check their personal pension situation in the upcoming three to six months. Moreover, this paper finds gender differences: male participants prefer a high level of interactivity, while women prefer a medium level.Research limitations/implicationsAn interesting modification to the current research design would be to use personal, self-relevant data in the online pension planner. Moreover, conducting the study in a computer laboratory could increase concentration on the task, and hence involvement. Next to gender, there might be other factors that possibly influence the results. It would be interesting to investigate other measures of behaviour such as the time spent on the pension planner. Further research should also study the effects of other features that shape user’s perception of interactivity, which include human-to-human interactivity.Practical implicationsThe results show that technological services, such as advanced online pension planners, can positively affect engagement with retirement planning. Thus, pension providers are encouraged to use interactive online pension planners. The results with respect to gender suggest tailoring pension planners to match specific preferences of recipients. New service technologies provide novel opportunities to cater to individual differences by, for example, integrating less interactive features for women than for men in a pension planner. Moreover, cognitive involvement should be stimulated by integrating relevant, interesting and valuable information.Social implicationsLack of engagement with retirement planning is an important challenge to Western societies. People who do not sufficiently search for information about their expected pension benefits may encounter significant pension gaps resulting in detrimental welfare effects at retirement. This problem is enhanced by the fact that increasingly, the risks and responsibility for retirement planning are being shifted towards pension plan participants themselves. Thus, finding ways to increase engagement with retirement planning by making use of advances in service technologies brings benefits to society.Originality/valueFirst, this paper contributes to the customer engagement literature by studying how new technological interfaces improve user experiences, knowledge and engagement within the low involvement context of retirement planning. Second, this paper advances service research by zooming in on customer heterogeneity in using the technology-based online pension planner and studying the moderating effect of involvement and gender more closely. Third, this paper contributes to the financial services literature by studying how new service technologies can help to increase attitudes, knowledge and engagement with retirement planning.

Many individuals avoid information relevant for retirement planning. This behavior is worrying as pension systems shift risks and responsibilities to individuals. Individuals who avoid pension information fail to discover whether they save too little for retirement, negatively affecting their long-run financial well-being. We generate knowledge about the factors that stimulate or hinder the search for pension information. Using an interdisciplinary lens, we develop a unifying model - the Retirement Belief Model (RBM) - and empirically validate it with field data from the Netherlands and United Kingdom. We find that the RBM core beliefs (susceptibility, severity, benefits, barriers, and self-efficacy) as well as trust and emotions significantly explain search for pension information. Our findings help both pension providers and policy makers in improving pension communication by stressing, for example, the benefits of information acquisition and designing segment specific approaches.

Descriptive information of peers’ behavior is used to nudge individuals to behave according to a norm. Yet, since such descriptive information is often a quantitative metric, it is possible that the behavioral change is due to anchoring. We disentangle peer and anchoring effects, and test boundary conditions in two experimental studies with a retirement saving contribution rate scenario. In study 1, we find a strong similarity between a peer and anchoring effect. In follow-up study 2, no anchoring effect is found when more extreme values (peer norms or anchors more distant from the control group behavior) are used, whereas the peer effect remains present. Furthermore, we find evidence that the informational component – as opposed to the normative component – of peer information plays a stronger role for peer effects.

We analyze preferences and beliefs of members in a DC pension scheme from Sweden, one of the first countries that launched choice-based funded individual pension accounts. Based on a survey among 2,646 members, we study the effect of choice overload, risk tolerance, and subjective knowledge on choice behavior and financial well-being. We find that more risk-averse and less knowledgeable members tend to invest in the default fund – a fund that is, however, one of the riskiest options on the choice menu. On top of this mismatch between members’ risk preferences and their investment choices, we find those members are more likely to feel negative about their future financial well-being. We also find a positive correlation between financial well-being and choice appreciation, whereas the act of choosing a fund has only minor impact.

Prior research shows that investors with smaller belief updates trade less actively, which positively affects their return performance. We examine the effect of different default frames of presenting past return information on investors’ belief updating. In particular, we analyze whether presenting longer information horizons as a default is associated with smaller belief updates. In lab and online experiments, we expose subjects to different past return information defaults and measure updates in their beliefs. Different from previous research, our subjects can easily opt out of the default to obtain additional information. We find that presenting long-term return information is not effective in reducing belief updates on average. Whereas belief updates are reduced for subjects who remain in their default, for those who opt out, we observe the opposite.