Contributions published at Financial Economics (Helga Fehr-Duda)

By now there is a large literature on experimentally elicited time preferences with respect to monetary rewards. Recently, time discounting of real effort tasks has received a lot of attention. However, comparatively little work has been done on discounting of charitable giving, which is the focus of this master thesis. Altruistic preferences are usually measured by subjects’ responses in dictator games, whereby they have to distribute a given endowment between themselves and others. Discounting comes into play when the payments occur not immediately but at some future time, and the question arises whether present bias, excessive discount rates, and time inconsistency manifest themselves not only in decisions pertaining to the decision makers themselves but also in decisions concerning charitable payments for other people. It may well be the case that behavior depends crucially on the type of charity. For example, disaster aid should reach the recipients very quickly, whereas donations to charitable institutions without any specific earmark may be perceived as not so urgent. There are many different methods of measuring time preferences. The approach most closely aligned with the allocations decisions in the dictator games is the convex time budget method, introduced by Andreoni and Sprenger (2012a), by which subjects allocate a given budget to two different payment dates.

This paper documents a large association between individuals' time discounting in incentivized experiments and their positions in the real-life wealth distribution derived from Danish high-quality administrative data for a large sample of middle-aged individuals. The association is stable over time, exists through the wealth distribution and remains large after controlling for education, income profile, school grades, initial wealth, parental wealth, credit constraints, demographics, risk preferences, and additional behavioral parameters. Our results suggest that savings behavior is a driver of the observed association between patience and wealth inequality as predicted by standard savings theory.

This work concentrates on examining whether there is a gender gap in the attitudes of professional athletes, professional swimmers in particular, toward the competition. Participants are engaged into the novel experiment, where they solve a real task, first under a noncompetitive individual incentive scheme, and then in a competitive group incentive scheme. While there is no gender differences in performance, only 38 percent of female athletes engage into competitive incentive scheme for the next performance comparing to 48 percent of male athletes. This gender gap in competition entry is not explained by performance and factors such as overconfidence, overplacement and risk preferences. The competition entry gap is driven by the individual preferences for performing in a competition. The results is that the professional athletes are trained for competition and they take this attitude with them in other life domains.

This work concentrates on examining whether there is a gender gap in the attitudes of professional athletes, professional swimmers in particular, toward the competition. Participants are engaged into the novel experiment, where they solve a real task, first under a noncompetitive individual incentive scheme, and then in a competitive group incentive scheme. While there is no gender differences in performance, only 38 percent of female athletes engage into competitive incentive scheme for the next performance comparing to 48 percent of male athletes. This gender gap in competition entry is not explained by performance and factors such as overconfidence, overplacement and risk preferences. The competition entry gap is driven by the individual preferences for performing in a competition. The results is that the professional athletes are trained for competition and they take this attitude with them in other life domains.

The common models of intertemporal decision making – exponential, hyperbolic, and subadditive discounting – have been found to fail to account for several behavioral biases. These integrative approaches take both attributes of the options, the payoff and the delay, to calculate their discounted value and then choose the option with the highest value. The best integrative model is the subadditive model which also has a comparative part. This is the result of subadditive discounting being the most widespread bias in the study. How-ever, when the decision maker does not discount subadditively, the results deteriorate sharply. The newly introduced similarity approach tries to incorporate all biases. As a comparative model it compares the attributes of the options of choice directly instead of discounting the payoffs. It uses the tradeoff of getting a payoff earlier or to wait longer to get a larger payoff in order to rephrase the situation to an advantage in time (getting the earlier payoff) and an advantage in value (getting the larger payoff). Like this, it can in-corporate the biases. The results show, that it has not only the best fit but is also better able to capture the biases of the participants. The problem of the similarity approach is that it cannot incorporate the biases independently. As the results show that the biases are indeed independent, it consequently fails to model decision making for certain combina-tions of biases. Furthermore, it cannot explain exponential discounting and has difficulties to model more complex settings like sequences of payoffs. It is concluded that the simi-larity approach is still not sophisticated enough to describe intertemporal decision making but it is a step into the right direction. Research has shown, that some decision makers indeed integrate the attributes and some compare them. Thus, the integrative and com-parative models still need to be refined by future research.

null

Standard economic models view risk taking and time discounting as two independent dimensions of decision makers' behavior. However, mounting experimental evidence demonstrates the existence of robust and systematic interaction effects. There are striking parallels in patterns of risk taking and time discounting behavior, which suggests that there is a common underlying force driving these interactions. Here we show that decision makers' anticipation of something going wrong in the future conjointly with their proneness to probability weighting generates a unifying framework for explaining seven puzzling regularities: delay-dependent risk tolerance, aversion to sequential resolution of uncertainty, preferences for resolution timing, hyperbolic discounting, subadditive discounting, the differential discounting of risky and certain outcomes, and the order dependence of prospect valuation. Finally, we discuss the implications of our framework for understanding real-world behavior, such as the coexistence of underinsuring and overinsuring.

The missing link model developed by Epper and Fehr-Duda hypothesises that the inherent uncertainty resting within all future events shapes our discounting behaviour in distinctive ways. The first prediction is that the uncertainty makes our time preferences hyperbolic. Second, the model hypothesises that our level of patience and the degree to which we discount hyperbolically depend on the level of the uncertainty. The model presents two channels through which this influence is exerted on us: the level of objective future risk and the extent to which we are anxious about the outcome of the future. This paper tests the above predictions through incentivised experiments run in Athens and Zurich. Results suggest that the participants living in environments with more objective risk were more impatient and discounted more hyperbolically. However, they were also more anxious about the future. The data thus supports the prediction that future risk makes us discount more strongly and hyperbolically in general, but which of the two factors most prominently generated this result could not be determined. Recommendations for uncovering this are made.

null

null

null

null

In a recent experimental study of intertemporal risky decision making, Andreoni and Sprenger (2012) find that subjects exhibit a preference for intertemporal diversification, which is inconsistent with discounted expected utility theory. It was claimed that their results are also at odds with models involving probability weighting, such as rank-dependent utility and cumulative prospect theory. Here we demonstrate, however, that rank-dependent probability weighting explains intertemporal diversification if decision makers care about portfolio risk. Moreover, we provide a unified account of all of Andreoni and Sprenger's key findings.

Uncertainty pervades most aspects of life. From selecting a new technology to choosing a career, decision makers rarely know in advance the exact outcomes of their decisions. Whereas the consequences of decisions in standard decision theory are explicitly described (the decision from description (DFD) paradigm), the consequences of decisions in the recent decision from experience (DFE) paradigm are learned from experience. In DFD, decision makers typically overrespond to rare events. That is, rare events have more impact on decisions than their objective probabilities warrant (overweighting). In DFE, decision makers typically exhibit the opposite pattern, underresponding to rare events. That is, rare events may have less impact on decisions than their objective probabilities warrant (underweighting). In extreme cases, rare events are completely neglected, a pattern known as the “Black Swan effect.” This contrast between DFD and DFE is known as a description–experience gap. In this paper, we discuss several tentative interpretations arising from our interdisciplinary examination of this gap. First, while a source of underweighting of rare events in DFE may be sampling error, we observe that a robust description–experience gap remains when these factors are not at play. Second, the residual description–experience gap is not only about experience per se but also about the way in which information concerning the probability distribution over the outcomes is learned in DFE. Econometric error theories may reveal that different assumed error structures in DFD and DFE also contribute to the gap.

Probability weighting has been largely ignored by economics despite abundant evidence that riskattitudes are in fact probability dependent. Probability weighting, however, provides a unifyingaccount of many real-world phenomena that are diffcult to reconcile with expected utility theory,such as the equity premium puzzle, the long-shot bias in betting markets, households'underdiversification and their willingness to buy small-scale insurance at exorbitant prices. Recentfindings suggest that probability dependence is not just a feature of laboratory data but is indeedpresent in financial, insurance and betting markets. Thus, the neglect of probability weightingprevents economists from comprehending important phenomena. In this paper, we discuss thefoundations and economic consequences of probability weighting and offer a practitioner's guideto understanding and modeling probability-dependent risk preferences.

Empirical research has shown that people tend to overweight small probabilities and underweight large probabilities when valuing risky prospects, but little is known about factors influencing the shape of the probability weighting curve. Based on a laboratory experiment with monetary incentives, we demonstrate that pre-existing good mood is significantly associated with women’s probability weights: Women in a better than normal mood tend to weight probabilities relatively more optimistically. Many men, however, seem to be immunized against effects of incidental mood by applying a mechanical decision criterion such as maximization of expected value.

A large body of experimental research has demonstrated that, on average, people violate the axioms of expected utility theory as well as of discounted utility theory. In particular, aggregate behavior is best characterized by probability distortions and hyperbolic discounting. But is it the same people who are prone to these behaviors? Based on an experiment with salient monetary incentives we demonstrate that there is a strong and significant relationship between greater departures from linear probability weighting and the degree of decreasing discount rates at the level of individual behavior.We argue that this relationship can be rationalized by the uncertainty inherent in any future event, linking discounting behavior directly to risk preferences. Consequently, decreasing discount rates may be generated by people’s proneness to probability distortions.

Markets for energy-using durables seem to be characterized by an “energy efficiency gap”, the phenomenon that consumers could save on total costs if they bought high-efficiency products but often decide not to do so. This fact has been attributed to consumers undervaluing future cost savings relative to the upfront purchase price, possibly due to their high discount rates. Factors influencing discounting behavior include the following: 1. Individuals’ rates of pure time preference, i.e. the rate at which future utility is ex- changeable for present utility 2. Uncertainty concerning product life, service reliability and future energy costs 3. Individuals’ limited access to capital markets resulting in liquidity constraints 4. Difficulties obtaining and processing information regarding future running costs The objective of the current research was to develop a descriptive model of discounting be- havior which disentangles the effects of pure time preferences from other potentially decisive factors in the context of financial decision making and to test the model with novel experi- mental data. The data was collected from a representative sample of the German speaking Swiss population. We elicited discount rates, measures of risk aversion and other individual data relevant for decisions on energy-using durables. There were two different treatment groups: One group of the participants responded to hypothetical questions and received a flat participation fee, the other group was paid depending on their decisions in an incentive compatible manner. The following results emerged: 1. We found a substantial incentive effect: People who responded to hypothetical ques- tions exhibited an average discount rate of 47.5% p.a., compared to 36.4% p.a. of the incentivized group. Model estimates show that there is no significant difference between groups’ preference parameters other than their levels of pure time preference. The incentive effect can be traced back to the behaviors of comparatively small subgroups of people in the hypothetical treatment who reacted strongly to two factors: skepticism about the certainty of future payments and liquidity constraints, both of which drove up discount rates considerably. We did not find any evidence whatsoever that participants in the hypothetical treatment did not do their best to respond conscientiously and honestly. We conjecture that the incentive effect is due to people’s inherent difficulty of foreseeing how they will behave when they actually face real consequences of their choices. For this reason we recommend that the design of policy measures should be principally based on the analysis of real decisions. 2. Liquidity constraints are an important factor affecting people’s behavior in the incentivized treatment group. We estimate that discount rates are 40% higher for liquidity-constrained individuals than for unconstrained ones who exhibit an average rate of time preference of more than 30% p.a. The magnitudes of these rates suggest that there are considerable obstacles to investment in energy-efficient durables even for people with unlimited access to capital markets. 3. Participants reported to be concerned about uncertain future energy costs and to have difficulties with assessing monthly energy costs. Both factors may play a crucial role in people’s undervaluation of future energy cost savings. Product- and consumer-specific information on present values of expected running costs may help consumers in their decision making. 4. If our estimates of time preference rates are indeed manifestations of people’s innate preferences, information and education will most likely not alter people’s behavior. In this case, policy could influence relative prices of high-efficiency and low-efficiency durables directly via feebates, a combination of fees and rebates. This idea can be transferred to other aspects of the purchase decision, such as service, warranty and leasing contracts. In order to be able to quantify the relative magnitudes of fees and rebates, the extent of undervaluation of future costs will have to be assessed on a market-by-market basis.