Definition
Behavioral retirement saving research documents how psychological factors — rather than forward-looking optimization — govern individual decisions about how much to save, in what vehicles, at what contribution rates, and when to access savings. The key finding is that defaults and automatic features dominate voluntary choices for the majority of workers, while behavioral biases (present bias, exponential-growth bias, procrastination) systematically suppress saving among those who make active choices.
Key Ideas
- Auto-enrollment as the dominant policy tool: Automatic enrollment in 401(k) plans increases participation and raises account balances by approximately 10 percentage points of first-year salary for workers with 5+ years of tenure (Beshears et al. 2015). Higher debt partially offsets the gain but net wealth still rises. The mechanism is inertia — workers accept defaults rather than actively opting in or out (see the Danish passive-saver evidence below).
- Default parameters matter in both directions: A 2015 TSP (federal employees' Thrift Savings Plan) default switch from a low-risk government securities fund to a lifecycle fund increased passive acceptance of the new fund but reduced average contribution rates, because more workers passively accepted the lower default contribution rate. Choosing the right default parameter — not just "auto-enrollment" generically — is critical.
- Present bias and exponential-growth bias reduce saving substantially: Present bias (dynamically inconsistent impatience) and exponential-growth bias (underestimating compound interest) are quantifiable from survey responses and are highly predictive of saving behavior. Goda et al. (2015, 2017) estimate that eliminating both biases could raise retirement wealth by up to 70%.
- Financial literacy is a second-order but positive factor: A one-standard-deviation increase in financial literacy (understanding of inflation, diversification, compound interest) is associated with an 18% increase in the likelihood of maximizing plan contributions, but operates on the intensive margin after enrollment is resolved.
- Framing and message delivery shift behavior at low cost: Employers using gender- and racially-similar messengers, or delivering descriptive ("what others do") rather than injunctive ("what one should do") norms, increase savings plan uptake. Targeted linking of saving increase opportunities to psychologically meaningful dates (birthdays) reduces procrastination costs.
- Illiquidity provides commitment value but must be balanced: Beshears et al. (2014a, 2017) model retirement saving as an optimal illiquidity problem. A two-account system (one liquid, one illiquid before retirement) approximates the social optimum, balancing commitment against liquidity needs for unexpected shocks.
- IRA rollover fees represent a persistent wealth leak: More than one-third of TSP participants roll balances into Individual Retirement Accounts (IRAs) with higher fees upon separation. Higher education slightly improves selection of lower-fee IRAs but does not eliminate the leak.
How It Works
The Passive Saver Evidence (Denmark)
Chetty, Friedman, Leth-Petersen, Nielsen, and Olsen (2014) provide the cleanest evidence using Danish administrative data covering the full retirement saving system. Key finding: 85% of individuals are passive savers who do not adjust savings in response to tax subsidies — meaning $1 of tax expenditure on retirement saving subsidies raises total net saving by only about 1 cent. Active savers (15%) offset any policy change by shifting assets across accounts without raising total saving.
Implication: voluntary incentive structures (IRAs, matching contributions) primarily reshuffle saving across accounts for the active minority rather than increasing aggregate household saving. Automatic contribution policies — which increase saving without requiring any action — are the only instruments proven effective for passive savers.
Why Defaults Work
Workers who receive a default allocation or contribution rate bear the psychological cost of deviating from it: they must make an active choice, understand alternatives, and override the inertia. For passive savers, this cost exceeds the perceived benefit of optimizing. Defaults thus function as implicit recommendations — workers interpret them as what a well-informed planner would choose — even when defaults are set suboptimally (as the TSP lifecycle switch demonstrated).
Present Bias and Procrastination
Workers with present bias are dynamically inconsistent — they prefer patience for their future selves but act impatiently now. Brown and Previtero (2014) identify procrastinators as workers who wait until the last day of open enrollment to choose their health plan; procrastinators are 2.4 pp less likely to participate in a supplemental retirement plan and take 44–85 days longer to sign up once offered. Offering a delay option (saving the increase for a future date) reduces near-term uptake, but linking the delay to a salient future date (a birthday) largely undoes the negative effect.
Why It Matters
- Disability Insurance (DI) / Supplemental Security Income (SSI) interaction: The passive-saver finding is directly relevant to the DI policy context: workers with the least education and lowest lifetime earnings — those most likely to eventually apply for DI — are disproportionately passive savers or non-savers. The absence of retirement saving makes DI the dominant post-disability income source, reinforcing the strong education-DI gradient documented in Education and DI Participation.
- Policy design: Tax incentives for retirement saving largely fail to reach their target population. Automatic contribution escalation and default-fund design are more effective per dollar of policy cost. The Danish evidence (1¢/$1 for tax subsidies) directly contradicts the assumption underlying the current U.S. system of tax-deferred retirement savings.
- Early retirement implications: Workers with depleted retirement savings at age 62 have fewer alternatives to early Old Age and Survivors Insurance (OASI) claiming — consistent with the Armour and Knapp (2021) finding that age-62 claimers have significantly lower retirement wealth. The behavioral saving research identifies the upstream mechanism: passive savers accumulate less over working years, arriving at 62 with fewer assets.
Open Questions
- What is the optimal default contribution rate for automatic enrollment? The TSP evidence shows that a lower passive-acceptance rate can be worse than the prior voluntary equilibrium.
- How much of the passive saver share (85% in Denmark) is portable to the U.S., where financial literacy, income levels, and plan structures differ?
- Does behavioral nudging in retirement saving affect the timing of DI applications, or do behavioral and DI pathways operate in separate populations?
Related
Sources