Frank Caliendo

Hunting the unobservables for optimal social security: a general equilibrium approach

We study the optimal size of a pay-as-you-go social security program for an economy composed of both permanent-income and hand-to-mouth consumers. While previous work on this topic is framed within a two-period partial equilibrium setup, we study this issue in a life-cycle general equilibrium model. Because this type of welfare analysis depends critically on unobservable preference parameters, we methodically consider all parameterizations of the unobservables that are both feasible and reasonable—all parameterizations that can mimic key features of macro data (feasible) while still being consistent with micro evidence and convention (reasonable). The baseline model predicts that the optimal tax rate is between 6 percent and 15 percent of wage income.

Time Inconsistency and Retirement Choice

Hyperbolic discounting with naivete is widely believed to provide a better explanation than exponential discounting of why people borrow so much and why they wait so long to save for retirement. We reach a different set of conclusions. We show that if financial planning is enriched to include the choice of when to retire, then naive hyperbolic discounters may borrow far less and start saving for retirement significantly earlier than exponential discounters.

Discount Functions and Self-Control Problems

In this paper we propose a new strategy for comparing the behavior of a hyperbolic discounter who possesses self-control problems to an exponential discounter who does not. Our strategy controls for inherent differences in overall levels of impatience across discount functions, which thereby allows us to isolate the pure effect of self-control problems. We argue that self-control problems, in their pure form as we identify them, can have a very surprising effect on intertemporal choices. Our method reveals that differences in decision making that could be attributed to self-control problems are at least partly (and perhaps entirely) just an artifact of an uncontrolled comparison.

Does it pay to be SMarT

The Save More Tomorrow plan has proven effective at raising employee saving rates and appears to be popular among participants and the media. An important question has remained on the minds of economists despite this success: just how close does the prescriptive SMarT plan come to approximating the normative life-cycle/permanent-income consumption rule? That is, does it pay (in a lifetime utility sense) to participate in a SMarT plan? We attempt to provide some rigorous answers to this question by employing a quantitative-theoretic model to perform dynamic welfare analysis, and our results tend to support the SMarT plan as a decent first approximation to the life-cycle/permanent-income rule. We also consider the problem of an altruistic employer seeking to maximize the lifetime utility of employees by appropriately choosing the default SMarT parameters in the face of employee heterogeneity in saving rates and uncertainty about whether they will actually stick with the plan. The employers problem is augmented to include employee heterogeneity concerning risk taking, and we also consider the possibility that SMarT saving may be met with increased borrowing.

Time Inconsistency and Retirement Planning

We quantify the welfare gains from better retirement planning using a model in which retirement planning is time inconsistent. A modest increase in a household’s planning horizon by just a few years generates large aggregate and individual welfare gains.

New Findings on Strategic IRA Investing

This article explores the issue of strategic IRA investing, but does so from an unusual premise. Comparing the traditional IRA with the Roth IRA, the authors confirm the well-known conclusion that the Roth alternative is preferable if the income tax rate during the distribution period is greater than the tax rate during the accumulation period and vice versa. However, they take the analysis a step further, demonstrating that this conventional wisdom only holds, however, if the individual is investing only in an IRA. If an individual is making unsheltered investments and IRA contributions, the Roth IRA is preferable to the regular IRA when the two tax rates are equal. They conclude with a review of the implications of converting from a traditional to a Roth IRA.

Commitment and Welfare

The standard approach to welfare analysis under dynamically inconsistent preferences is to assume that the welfare of an individual is maximized if he can commit to his initial goal. We study a potential rationale for such welfare analysis. In some prominent, well-studied examples with intertemporal tradeoffs (like the choice between investing in a project now or later, doing an unpleasant task now or procrastinating it until later, and eating a cake), we find that the commitment allocation can multiself Pareto dominate the non-cooperative equilibrium allocation if the number of time-dated selves exceeds a low threshold. While a common concern with behavioral welfare analysis is that the later selves of an individual are unfairly hurt as a result of committing the individual to his initial goal, our findings indicate that this concern may be more relevant for settings where the number of sequential choices is very low.

Overconfidence in financial markets and consumption over the life cycle

Overconfidence is a widely documented phenomenon. Empirical evidence reveal two types of overconfidence in financial markets: investors both overestimate the average rate of return to their assets and underestimate uncertainty associated with the return. This paper explores implications of overconfidence in financial markets for consumption over the life cycle. The authors obtain a closed-form solution to the time-inconsistent problem facing an overconfident investor/consumer who has a CRRA utility function. They use this solution to show that overestimation of the mean return gives rise to a hump in consumption during the work life if and only if the elasticity of intertemporal substitution in consumption is less than unit. They find that underestimation of uncertainty has little effect on the long-run average behavior of consumption over the work life. Their calibrated model produces a hump-shaped work-life consumption profile with both the age and the amplitude of peak consumption consistent with empirical observations.

Strategies for Maximizing Estate Wealth

The conventional wisdom that using tax-sheltered accounts such as Keogh plans, IRAs, 401(k) programs, etc. is an excellent way to save for retirement but a poor way to accumulate an estate is shown to be incorrect at least under a variety of reasonable circumstances. It is shown that net estate wealth (i.e., after income and estate taxes) generally will be greater by saving in such plans compared to saving in conventional (i.e., taxable) accounts. The misconception arises because the net wealth left after taxes on two estates of equal size at death will be less when a significant part of the assets are in tax-sheltered plans, but this is not a fair comparison because if all other factors (e.g., income and consumption during the accumulation period, pre-tax rates of return, etc.) are held constant, the estate associated with the tax-sheltered investment will unequivocally be larger at the time of death. It is also shown that saving in tax-sheltered plans during the early years of the lifecycle, followed by saving in a taxable account in the later years, will generate a maximum after-tax transfer to heirs.

Sources of Error in Estimating the Personal Consumption Offset inRetirement

The economic loss appraisal in a wrongful death action typically includes a year-by-year calculation of the personal consumption expenditures of the deceased, which is then subtracted from earning capacity in order to determine a net loss of support to the survivors. 1 Sometimes, perhaps even often, this is simply done over the period of the worklife expectancy with consumption in the retirement years either ignored altogether or implicitly included by applying the offset factor to the component added to earnings for contributions to a retirement program during the worklife. For lack of a better term, the latter will be referred to as the “worklife only” or WLO appraisal approach. Generally, part of the fringe benefit payment includes the employer’s contribution to Social Security and to a private retirement program. To the extent that such contributions for retirement income are made to programs that are actuarially fair, 2 then applying the offset factor against such contributions does at least partially account for income and consumption in the retirement period. That is, the contribution is the present value of the future retirement benefit, and, by applying the offset factor against that contribution, at least part of the foregone future consumption of the deceased also is accounted for. However, such adjustments are likely to be imperfect at best. Unless the deceased died on the first day of his worklife, the approach will not account for contributions made in previous years that will generate retirement income and consumption from that income. Further, the consumption offset factor during the earlier working years tends to be lower than in the retirement years due to