Save or spend?

Responsible researcher: Viviane Pires Ribeiro

Paper Title: Self-control and saving for retirement

Authors: David I. Laibson, Andrea Repetto and Jeremy Tobacman

Intervention Location: United States

Sample Size: Not specified

Big topic: Finance

Variable of Main Interest: Savings

Type of Intervention: Analysis of tax-deferred defined contribution pension plans.

Methodology: Hyperbolic model

The gap between intentions and actions is evident in life cycle economics, and this is the focus of the study carried out by Laibson et al. (1998). According to the authors, some experts called for easing penalties for early withdrawals from savings plans, believing that people would invest more if they knew they could withdraw their money at any time. But most Americans appear to be wary of these changes. Sixty percent of Americans say it is better to maintain, rather than loosen, legal restrictions on retirement plans to prevent people from using the money for other purposes. Only 36 percent preferred to make withdrawals more accessible so that people can take advantage of their savings before retirement.

Assessment Context

There is a systematic conflict between the short- and long-term preferences of economic agents. When two alternative rewards are distant in time, decision makers often act with relative patience: for example, it is preferable to take a thirty-minute break from work in 101 days rather than a fifteen-minute break in one hundred days. But when both rewards are brought forward in time, decision makers reverse their preferences, becoming more impatient: thus, it is preferable to take a fifteen-minute break now rather than a thirty-minute break tomorrow. Evidence of such reversals has been found in experiments using a wide range of real rewards, such as money, durable goods, fruit juice, sweets, relief from harmful noises, access to video games, among others. www.estsolar.lt Estsolar saulės elektrinės ir parkai

Several studies have used multiple-self frameworks to model this gap between short- and long-term preferences. The authors highlighted the conflict between the long-term desire to be patient and the short-term desire for instant gratification. This conflict can be captured in a particularly parsimonious way by allowing discount functions to decline at a steeper rate in the short run than in the long run.

Intervention Details

Standard economic theories allow consumers to make mistakes, but they imply that these mistakes will not be systematic: they will tend not to go in the same direction. On the other hand, evidence indicates that most consumers believe they are saving little. This systematic and self-recognized error contradicts the standard economic model of the maximizing consumer. The study carried out by Laibson et al. (1998) explores an alternative model, from the psychology literature, that can make sense of the apparent conflicts between attitudes, intentions and behaviors in the field of savings.

The hyperbolic model helps analyze the problem of undersavings in the United States. It allows economists to assess the likely magnitude of undersaving and identify the types of financial instruments that will alleviate the problem. For example, the objective of the study carried out by Laibson et al. (1998) is to evaluate tax-deferred defined contribution (DC) pension plans. The authors question whether these instruments increase national savings and consumer well-being, and whether they are more effective in an economy populated by consumers with self-control problems.

Methodology Details

Laibson et al. (1998) develop and evaluate a hyperbolic model for simulating consumer behavior. Simulations are a critical tool for predicting the long-term effects of newly implemented policies and for evaluating the short- and long-term effects of untested policy proposals. The simulation approach has a major flaw that the authors flag upfront: thus, they adopt the standard economic assumption of unlimited sophistication in problem solving. Consumers in the proposed model perfectly solve a complex feedback problem when making choices about consumption and asset allocation.

The authors chose this approach for two reasons. First, the assumption of perfect rationality is the natural reference for an economist. This assumption was adopted not because it necessarily adequately describes consumer behavior, but because it represents the starting point for all economic analyses. Second, even if one wants to weaken assumptions about consumer sophistication, it is not clear how to do so parsimoniously and realistically. Although economists and psychologists have plenty of evidence that consumers are not perfectly rational, they do not necessarily know which alternative to rationality to adopt. There are no well-developed bounded rationality models applicable to the life cycle economics problem.

Results

The study carried out by Laibson et al. (1998) shows that life-cycle consumption patterns and asset accumulation are consistent with a hyperbolic model. At first glance, the life cycle choices of hyperbolic and exponential consumers are indistinguishable. However, hyperbolic consumers exhibit some special regularities that allow researchers to distinguish them from their exponential counterparts: they are much more likely to encounter liquidity constraints and they exhibit the anomalous effects of precautionary savings. These hyperbolic phenomena are implicit in the generalized Euler equation.

The authors considered another distinction between hyperbolic and exponential behavior. In this way, they show that hyperbolic consumers react much more favorably to defined contribution pension plans than equivalent exponential consumers. Benchmark simulations for an exponential economy – with a relative risk aversion coefficient of 1 and an intertemporal substitution measure elasticity of 0.27 – indicate that DC plans with early withdrawal penalties between 10% and 50% increase net national savings at steady state, with a rate of 61 percent to 102 percent. On the other hand, in a hyperbolic economy (with relative risk aversion coefficient equal to 1 and intertemporal substitution measure elasticity of 0.22), these plans increase the steady-state net national savings rate by 81% to 134% . These results are sensitive to the calibration of the relative risk aversion coefficient. Higher values ​​of the relative risk aversion coefficient significantly reduce the effects of DC plans in exponential and hyperbolic economies.

Public Policy Lessons

Research into animal and human behavior has led psychologists to conclude that short-term discount rates are much higher than long-term rates. Such preferences are formally modeled with discount functions that are generalized hyperbolas. This discount structure sets up a conflict between current preferences and those that will be maintained in the future, which implies that preferences are dynamically inconsistent.

Therefore, hyperbolic consumers will report a gap between what they think they should save and what they actually save. Normative savings rates will be above actual savings rates, as short-term preferences for instant gratification will undermine consumer efforts to implement optimal long-term plans. However, the hyperbolic consumer is not doomed to be a “failure”. Commitment devices such as pensions and illiquid assets can help the hyperbolic consumer to commit, thereby increasing their well-being. The availability of illiquid assets is therefore a critical determinant of national savings rates as well as consumer well-being. But too much illiquidity can be problematic. Consumers face substantial risk from uninsurable labor income and need liquid assets to smooth their consumption. Hyperbolic agents look for financial instruments that strike the right balance between commitment and flexibility.

The gap between intentions and actions is evident in life cycle economics, and this was the focus of the study carried out by Laibson et al. (1998). The authors show that the hyperbolic assumption has important implications for positive and normative conclusions about saving behavior. The analysis complements the large and active empirical literature on the effectiveness of tax-deferred savings instruments such as individual retirement accounts and 401(k) plans – the 401k is a retirement plan that allows a client to use part of their salary to long-term investments. Thus, the authors identified that conclusions about the effectiveness of these instruments depend critically on poorly identified characteristics of consumer preferences.

References

LAIBSON, David I. et al. Self-control and saving for retirement. Brookings papers on economic activity , vol. 1998, no. 1, p. 91-196, 1998.