We calculate the socially optimal level of illiquidity in an economy populated by households with taste shocks and present bias (Amador, Werning, and Angeletos 2006).The government chooses mandatory contributions to respective spending/savings accounts, each with a different preretirement withdrawal penalty.Penalties collected by the government are redistributed through the tax system.When naive households have heterogeneous present bias, the social optimum is well approximated by a three-account system: (i) a completely liquid account, (ii) a completely illiquid account, and (iii) an account with a ~10% early withdrawal penalty.In some ways this resembles the U.S. system, which includes completely liquid accounts, completely illiquid Social Security and 401(k)/IRA accounts with a 10% early withdrawal penalty.The social optimum is also well approximated by an even simpler two-account system-(i) a completely liquid account and (ii) a completely illiquid account-which is the most common retirement system in the world today.

We test a nudge in a field experiment on credit cards. The nudge shrouds the autopay enrollment option for cardholders to automatically pay exactly the credit card minimum payment each month. After six months, the nudge decreases the fraction of cardholders who only pay exactly the minimum by 23 percent. However, the nudge does not significantly reduce credit card debt. Nudged cardholders often choose autopay amounts that are only slightly higher than the minimum payment. The nudge reduces autopay enrollment, which increases missed payments. The nudge reduces manual payments by autopay enrollees. Cardholders frequently lacking liquid cash best explains our results. (JEL C93, D12, D91, G21, G51)

Does automatic enrollment into retirement saving increase household debt? We study the randomized roll-out of automatic enrollment pensions to ~160,000 employers in the United Kingdom with 2-29 employees. We find that the additional savings generated through automatic enrollment are partially offset by increases in unsecured debt. Over the first 41 months after enrollment, each additional month increases the average automatically enrolled employee’s pension savings by £33-£39, unsecured debt (such as personal loans and bank overdrafts) by £7, the likelihood of having a mortgage by 0.05 percentage points, and mortgage balances by £120. Automatic enrollment causes loan defaults to fall and credit scores to rise modestly.

We study the introduction of a choice architecture design intended to increase short-term savings among employees at five U.K. firms.Employees were offered the opportunity to opt into a payroll deduction program that auto-deposits funds from each paycheck into a short-term savings account from which withdrawals are possible at any time.We find that employees who opted into the program kept using it.Among employees whose accounts were created early enough to be observed over the first 12 months after their account activation and who did not separate from employment during this period, 96% still had a balance greater than £1 and 87% received an automatic payroll contribution in month 12.However, product take-up was very low: no more than 0.7% of eligible employees ever activated an account.Opt-in access to short-term savings programs does not elicit widespread participation.

We study the effects of monetary and fiscal policy in a heterogeneous-agent model where households have present-biased time preferences and naive beliefs. The model features a liquid asset and illiquid home equity, which households can use as collateral for borrowing. Because present bias substantially increases households' marginal propensity to consume (MPC), present bias increases the impact of fiscal policy. Present bias also amplifies the effect of monetary policy but, at the same time, slows down the speed of monetary transmission. Interest rate cuts incentivize households to conduct cash-out refinances, which become targeted liquidity-injections to high-MPC households. But present bias also introduces a motive for households to procrastinate refinancing their mortgages, which slows down the speed with which this monetary channel operates. Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at www.nber.org.

We estimate beta-delta time preferences and relative risk aversion (RRA) using a lifecycle model including stochastic income, liquid and illiquid assets, credit cards, dependents, Social Security, mortality, and bequests. Preference parameters are identified by cross-tabulating four lifecycle age intervals and four balance sheet moments: the share of households carrying (ie, revolving) credit card debt, average carried credit card debt, average net wealth among households carrying credit card debt, and average net wealth among households not carrying credit card debt. The 16 moments are approximately matched by (MSM) parameter estimates beta=0.53, delta=0.99, and RRA = 1.9.

Using data from Vanguard 401(k) plans, we establish a new empirical fact: retirement plan contributions are remarkably stable after loans and hardship withdrawals. Relative to a control group, loan takers' contribution rates fall by just 0.8 percentage points in the two years following loan issuance, despite simultaneously making large loan repayments. Relatedly, most participants continue making elective contributions after hardship withdrawals. For plan sponsors considering the introduction of penalty-free emergency withdrawals newly permitted under SECURE 2.0, our results suggest that most participants would be able to access this liquidity feature while maintaining their long-term retirement savings through an 'automatic repayment' feature.

When solving discrete-time consumption models with present-biased time preferences, backwards induction generates equilibria that are non-robust in the sense that policy functions are often sensitive to parameter choices, including the modeler's choice of the time-step. The current paper identifies a range of "sweet-spot" time-steps that (i) contains the psychologically relevant present bias horizons, and, (ii) generates numerically indistinguishable (i.e., robust) policy functions. This sweet spot includes both a computationally feasible range of discrete-time cases and the limiting continuous-time case (Harris and Laibson, 2013). Accordingly, researchers modeling present bias in buffer stock models can choose either discrete-time cases calibrated to be in the sweet spot or the analytically tractable continuous-time case; these approaches yield essentially identical policy functions.Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at www.nber.org.

PURPOSE:To evaluate if nudges delivered by text message prior to an upcoming primary care visit can increase influenza vaccination rates. DESIGN:Randomized, controlled trial. SETTING:Two health systems in the Northeastern US between September 2020 and March 2021. SUBJECTS:74,811 adults. INTERVENTIONS:Patients in the 19 intervention arms received 1-2 text messages in the 3 days preceding their appointment that varied in their format, interactivity, and content. MEASURES:Influenza vaccination. ANALYSIS:Intention-to-treat. RESULTS:Participants had a mean (SD) age of 50.7 (16.2) years; 55.8% (41,771) were female, 70.6% (52,826) were White, and 19.0% (14,222) were Black. Among the interventions, 5 of 19 (26.3%) had a significantly greater vaccination rate than control. On average, the 19 interventions increased vaccination relative to control by 1.8 percentage points or 6.1% (P = .005). The top performing text message described the vaccine to the patient as "reserved for you" and led to a 3.1 percentage point increase (95% CI, 1.3 to 4.9; P < .001) in vaccination relative to control. Three of the top five performing messages described the vaccine as "reserved for you." None of the interventions performed worse than control. CONCLUSIONS:Text messages encouraging vaccination and delivered prior to an upcoming appointment significantly increased influenza vaccination rates and could be a scalable approach to increase vaccination more broadly.

Understanding moral acceptability and willingness to use is crucial for informing policy.

We study a retirement savings plan with a default contribution rate of 12 percent of income, which is much higher than previously studied defaults. Twenty-five percent of employees had not opted out of this default 12 months after hire; a literature review finds that the corresponding fraction in plans with lower defaults is approximately one-half. Because only contributions above 12 percent were matched by the employer, 12 percent was likely to be a suboptimal contribution rate for employees. Employees who remained at the 12 percent default contribution rate had average income that was approximately one-third lower than would be predicted from the relationship between salaries and contribution rates among employees who were not at 12 percent. Defaults may influence low-income employees more strongly in part because these employees face higher psychological barriers to active decision making.

ABSTRACTDoes automatic enrollment into a retirement plan increase financial distress due to increased borrowing outside the plan? We study a natural experiment created when the U.S. Army began automatically enrolling newly hired civilian employees into the Thrift Savings Plan. Four years after hire, automatic enrollmentincreases cumulative contributions to the plan by 4.1% of annual salary, but we find little evidence ofincreased financial distress. Automatic enrollment causes no significant change in credit scores, debt balances excluding auto debt and first mortgages, or adverse credit outcomes, with the possible exception of increasedfirst‐mortgage balances in foreclosure.

We conduct a genome-wide association study (GWAS) of educational attainment (EA) in a sample of ~3 million individuals and identify 3,952 approximately uncorrelated genome-wide-significant single-nucleotide polymorphisms (SNPs). A genome-wide polygenic predictor, or polygenic index (PGI), explains 12-16% of EA variance and contributes to risk prediction for ten diseases. Direct effects (i.e., controlling for parental PGIs) explain roughly half the PGI's magnitude of association with EA and other phenotypes. The correlation between mate-pair PGIs is far too large to be consistent with phenotypic assortment alone, implying additional assortment on PGI-associated factors. In an additional GWAS of dominance deviations from the additive model, we identify no genome-wide-significant SNPs, and a separate X-chromosome additive GWAS identifies 57.

Resource allocations are jointly determined by the actions of social planners and households. In this paper, we highlight the distinction between planner optimization and household optimization. We show that planner optimization is a partial substitute for household optimization and that this is still true when households have private in- formation about their preferences. We illustrate a novel identification problem: some aggregate equilibrium properties that are implied by household optimization are also implied by planner optimization. A rational utilitarian planner enacts policies that cause the Euler equation to hold on average in a population of non-optimizing households, which we refer to as pseudo-rationality. Hence, widely studied allocative optimality conditions also arise in an allocatively inefficient economy with a rational planner, who uses tools such as default savings and Social Security to influence the consumption choices of myopic and passive households. Accordingly, aggregate classical optimization tests implemented with data from economies with paternalistic governments cannot reveal whether households would make optimal decisions on their own.

Standard consumption models assume a notional consumption flow that does not distinguish between nondurable and durable consumption. Such notional-consumption models generate notional marginal propensities to consume (MPC). By contrast, empirical work and policy discussions often highlight marginal propensities for expenditure (MPX), which incorporate spending on a durable stock. We compare the notional-consumption model to an isomorphic model with a durable stock, and map notional MPCs into MPXs. The mapping is especially simple for a one-period horizon: MPX = (1 - s + s/(r+d)) x MPC, with durable share s, real interest rate r, and durable depreciation rate d.Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at www.nber.org.

Present bias causes procrastination, which leads households to stick with auto-enrollment defaults. However, present bias also engenders overconsumption. Separation from each employer generates a rollover of 401(k) balances to an individual retirement account (IRA) that is more liquid than the 401(k) account. Households with sufficient present bias will partially or fully deplete these rollover IRAs before retirement. Present-biased agents may be whipsawed by auto-enrollment. They follow their employer's default while still employed and then spend some or all of the new savings shortly after they separate from each employer.

We study nudges that turn out to have precise null effects in reducing long‐run credit card debt. We test nudges across two field experiments covering 183,441 UK cardholders. Our first experiment studies nudges added to monthly credit card statements. Our second experiment studies letters and email nudges (separate from monthly statements) sent to cardholders who signed up to automatically pay the minimum required payment. In a follow‐up survey to our second experiment, we find that 96% of respondents underestimate the time it would take to fully repay a debt if the cardholder made only the minimum required payment. The nudges reduce this confusion, but underestimation remains overwhelmingly common.

Polygenic indexes (PGIs) are DNA-based predictors. Their value for research in many scientific disciplines is growing rapidly. As a resource for researchers, we used a consistent methodology to construct PGIs for 47 phenotypes in 11 datasets. To maximize the PGIs' prediction accuracies, we constructed them using genome-wide association studies-some not previously published-from multiple data sources, including 23andMe and UK Biobank. We present a theoretical framework to help interpret analyses involving PGIs. A key insight is that a PGI can be understood as an unbiased but noisy measure of a latent variable we call the 'additive SNP factor'. Regressions in which the true regressor is this factor but the PGI is used as its proxy therefore suffer from errors-in-variables bias. We derive an estimator that corrects for the bias, illustrate the correction, and make a Python tool for implementing it publicly available.