The number of data sources available for academic research on retirement economics and policy has increased rapidly in the past two decades. Data quality and comparability across studies have also improved considerably, with survey questionnaires progressively converging towards common ways of eliciting the same measurable concepts. Probability-based Internet panels have become a more accepted and recognized tool to obtain research data, allowing for fast, flexible, and cost-effective data collection compared to more traditional modes such as in-person and phone interviews. In an era of big data, academic research has also increasingly been able to access administrative records (e.g., Kostøl and Mogstad, 2014; Cesarini et al., 2016), private-sector financial records (e.g., Gelman et al., 2014), and administrative data married with surveys (Ameriks et al., 2020), to answer questions that could not be successfully tackled otherwise.

In an effort to promote comparative research on pensions, the Gateway to Global Aging Data is developing harmonized cross-national panel data on pension benefits and retirement incentives. Past research has varied in how it predicts pension benefits for individuals who have not yet claimed their benefits when administrative data on earnings histories is unavailable. We use the Gateway data to evaluate several alternative approaches to computing prospective pension benefits using common survey questions and validate them against matched administrative data. We find that in some settings naïve measures of pension benefit growth from continued work and delayed benefit claiming can perform as well as measures based on administrative data. We also find that prospective benefit levels are sensitive to the heterogeneity of lifecycle earnings dynamics, resulting in substantial measurement error even after accounting for work history.

A large number of Americans do not have bank accounts (the ‘unbanked’) or rely on costly alternative financial services (AFS) such as payday loans (the ‘underbanked’), with implications for wealth accumulation and retirement preparedness. Using primary data, we document large racial/ethnic differences in unbanked and in frequent AFS usage rates. We study the role of socio-economic status (SES), financial literacy, trust in financial institutions, networks, and time preferences in explaining these gaps. While these variables explain a large fraction of the white-minority gaps in unbanked status the same is not true for gaps in AFS use. A Blinder-Oaxaca decomposition confirms these patterns: gaps in unbanked status are mostly explained by differences in endowments across groups, for AFS gaps differences in returns to endowments have the largest explanatory power. Our findings suggest that, while related, unbanked and underbanked are distinct concepts with different underlying causes that may require different policy responses.

To ensure the long-term sustainability of the US Social Security System, several policy alternatives can theoretically be implemented. However, in practice, consumer responses can be challenging for policymakers to anticipate. We conducted a randomized survey on the nationally-representative Understanding America Study (UAS) panel where respondents were presented with a series of ‘policy scenarios’ in which the government enacts alternative reforms aimed to reduce the expected shortfall in the trust fund that pays Social Security retirement benefits. These scenarios included an increase in the Social Security payroll tax, an increase in the wage ceiling, and a reduction of benefits. We find that changes in respondents’ subjective expectations about their benefits and how this will affect their behaviors are directionally consistent with what would be expected when individuals are attentive to the available information and form rational expectations (e.g., monthly benefit expectations increase upon the announcement of an increase in the Social Security tax rate or in the wage ceiling, and they decrease less-than-proportionally with hypothetical benefit cuts). However, surprisingly, these changes are not sensitive to the magnitudes of policy change (e.g., the increase in expected benefits is about the same regardless of the size of Social Security payroll tax increases). Individuals with higher levels of education, cognitive ability, and financial literacy are more likely to adjust their expectations as predicted by theory.

This paper presents results from an online experiment that tested the effect of simple, concise information on respondents' knowledge of the rules of Social Security's Retirement Earnings Test (RET). We find that a straightforward, simple information treatment has a large effect on knowledge of the RET and on prospective choices regarding working and claiming benefits. We tested two additional information treatments that provided more comprehensive information and used illustrative examples. One of the additional treatments used only text, while the other one included a visual tool. Our results suggest that the additional treatments did not further improve knowledge; the simplest intervention yielded the same effect as the more comprehensive ones.

Married men's Social Security claiming behavior does not always take into account spouse and survivor benefits. Specifically, married men tend to choose claiming ages that do not maximize the household's and widow's expected present value of benefits. To understand what contributes to this pattern, we conduct an online survey experiment with a representative sample of Americans. We randomly assign respondents to one of four vignettes that present information about a character who is choosing when to claim his retirement benefits. The vignettes differ by whether the character is married or not, whether information about survivor benefits is presented, and whether the information includes an illustrative example. We next ask respondents to provide advice to the character about when to claim. We find that (1) respondents do value survivor benefits for spouses, and (2) information about survivor benefits tends to increase the suggested claiming ages only among the subgroup of respondents who are the least knowledgeable about these benefits.

We study whether retirement preparedness of overconfident individuals – those with high self-rated but low objective financial knowledge – and underconfident individuals – those with low self-rated but high objective financial knowledge – differs from that of individuals for whom self-assessed and actual knowledge align. We find that overconfident individuals fare no different than others with similarly low levels of objective financial knowledge in terms of retirement preparedness, while they are less interested in improving their knowledge. Underconfident individuals exhibit worse economic outcomes than others with similarly high financial knowledge but show interest in learning more about retirement. Our results suggest that accompanying financial literacy campaigns with initiatives that increase awareness of one's actual knowledge may be an effective lever to encourage the overconfident to increase their financial competence and to prove to the underconfident that they have sufficient skills to start planning their financial future.

Individuals are increasingly asked to take responsibility for preparing for retirement and available financial products to do so are growing in sophistication. A better understanding of how non-cognitive skills influence financial capability and retirement preparation could help effective policy design. This area of research has been hampered by the struggle to find reliable measures of these skills. I argue that questionnaires themselves can be seen as performance tasks, such that measures of survey effort could lead to meaningful measures of non-cognitive skills. I exploit the fact that I observe respondents taking multiple survey modules covering different topics in different moments of time to build survey effort measures in a nationally representative internet panel. I use survey effort measures along with self-reports to study the role of non-cognitive skills on retirement preparation and financial capability. My results show that non-cognitive skills can have a significant role, beyond the role of cognitive ability.

Automatic enrollment has substantially increased employee participation in defined contribution plans. Yet little is known about how retirement plan design features influence retirement wealth accumulation in a setting of labor market turnover. We find that employees separating from jobs with automatic enrollment plans are significantly more likely to take a cash distribution (and potentially pay a tax penalty) than those separating from jobs with voluntary enrollment plans, offsetting some of the benefits from automatic enrollment. Yet given the sizeable improvements in plan participation from automatic enrollment, wealth accumulation for automatically enrolled participants, net of cash-outs and penalties, remain higher than it would have been under voluntary enrollment.