Survey Evidence On Consumption And Portfolio Choice

Wealth Accumulation and the Propensity to Plan, with John Ameriks and John Leahy

As described in the psychological research page, John Ameriks, John Leahy, and I believe that it is becoming increasingly important to look beyond the principle of revealed preference in order to understand behavior. Wealth accumulation and portfolio choice are cases in point. Economists have developed elaborate theoretical structures in which the "discount rate" is the key to individual wealth accumulation, and "risk aversion" is the key to portfolio choice. If this is intended to be more than a tautology, then presumably there must be some method for measuring these "preference parameters" and confirming their behavioral importance. A direct survey approach to measurement was attempted by Barsky, Juster, Kimball, and Shapiro, with far from overwhelming results: inferred preference parameters had very little power in explaining actual wealth accumulation and portfolio choice. Other indirect approaches have yielded equally little in the way of positive evidence. In a nutshell, there is little evidence that differences in discount rates can explain differences in wealth accumulation.

A central goal of our research is to dig deeper into the underlying individual differences that give rise to different levels of wealth accumulation for economically and demographically similar individuals. We have now constructed two surveys devoted in large part to this exploration, and also to demonstrating other points of advantage of survey methodology. Economists really do need to get out more: we spend so much time living with our models that we have a hard time opening up to the truly astonishing variety of behaviors that surrounds us. We need to listen far more closely to what people have to say about their own behaviors. Surveys aren't the only way to do this, but at least they represent a step in the right direction.

Our first findings show that planning and budgeting skills may be of great value in the process of wealth accumulation. Psychologists (such as Peter Gollwitzer of NYU) have long argued that various types of preparatory activities can aid in goal achievement for tasks such as taking one's medicine on time. What is new is the finding that these same skills are of value in the context of the crucial life-long goal of building wealth.

Wealth Accumulation and the Propensity to Plan, with John Ameriks and John Leahy, forthcoming Quarterly Journal of Economics

Measuring Self Control, with John Ameriks, John Leahy, and Tom Tyler

While a finding such as this is fascinating, it is not entirely satisfying. Why do planners behave differently than non-planners? How do planners differ from non-planners in their savings behavior? In essence, the issue is one of modeling. A richer vision of how planning operates will be required if we are to incorporate it into the standard life cycle model of wealth accumulation. Our current research is aimed at enriching our understanding in just this respect.

Our primary hypothesis is that planning and budgeting skills are self control aids. Many of us find ourselves spending more money than we would like at some time in our lives. Typically, we find it relatively difficult to rein in our spending: being good at planning and budgeting may come in particularly handy at such times. Planners overcome self control problems, while non-planners do not. In essence, we believe that this vision fits with that of psychologists, in which planning is seen as helping people overcome an inclination to avoid acts that might otherwise be unpleasant.

Armed with this hypothesis, we have conducted another survey in which questions on self control took center stage. The findings, while preliminary, are exciting. We have found evidence in favor of the hypothesis linking planning and self control. In addition, we have identified a new class of self control problems in which individuals find it difficult to spend rather than difficult to save. Finally, we established a very strong connection between self control and conscientiousness, one of the big five factors beloved of personality theorists. The connection between self control and planning is essentially a corollary of that between self control and conscientiousness: indeed many psychologists see planning as one essential ingredient of the somewhat richer conscientiousness factor. The following paper contains these results, albeit in preliminary form.

Measuring Self Control, with John Ameriks, John Leahy, and Tom Tyler

The Absent-Minded Consumer, with John Ameriks and John Leahy

In our current work, we are exploring further aspects of the self control-planning-wealth link. In particular, we have turned our attention to survey findings on the use of specific budgetary and planning mechanisms to keep under control the temptation to spend. Note that temptation is a concept that is very hard to gauge from actual behaviors: of necessity, realistic measures cannot rely on revealed preference. This is another crucial point in favor of additional use of this methodology.

In addition to issues of self control, we have focused on issues of control. We are uncovering a great deal of evidence concerning the degree of individual ignorance concerning the level of spending: in a word, it is massive. Given this, it is not correct to see individuals as having complete control over their level of spending. We have developed a model in this spirit, using the imperfect memory formulation of Piccione and Rubinstein. We identity a form of precautionary spending that tries to compensate for present ignorance: this is the natural counterpart to the precautionary saving associated with ignorance about the future. Here is our paper on this subject:

The Absent-Minded Consumer, with John Ameriks and John Leahy

Retirement Consumption: Insights from a Survey, with John Ameriks and John Leahy

There is one other subject on which we have already gained new insights from our survey. There is evidence that consumption falls a great deal at retirement. Various papers have attempted to infer from purely behavioral evidence whether or not this fall was anticipated in the pre-retirement years. We took the radical step of asking people! Their answers came as something of a surprise. Indeed, it appears, many households anticipate well ahead of time the significant fall in consumption at retirement. There is little evidence that they are "shocked" to discover their new economic circumstances.

Retirement Consumption: Insights from a Survey, with John Ameriks and John Leahy

We hope to use our surveys to investigate several other topics, such as the liquidity of residential real estate, and the psychological background to financial behaviors. In all cases, the advantages of survey techniques over standard purely choice-based techniques of empirical work will be manifest.