Steven J. Humphrey

Choice under Uncertainty: Evidence from Ethiopia, India and Uganda

We review experimental evidence collected from risky choice experiments using poor subjects in Ethiopia, India and Uganda. Using these data we estimate that just over 50% of our sample behaves in accordance with expected utility theory and that the rest subjectively weight probability according to prospect theory. Our results show that inferences about risk aversion are robust to whichever model we adopt when we estimate each model separately. However, when we allow both models to explain portions of the data simultaneously, we infer risk aversion for subjects behaving according to expected utility theory and risk-seeking behaviour for subjects behaving according to prospect theory.

Regret Aversion or Event-Splitting Effects? More Evidence under Risk and Uncertainty

Recent experimental evidence has concluded that experimentally observed juxtaposition effects, as predicted by regret theory1, are largely attributable to “event-splitting effects” (ESEs) whereby the subjective decision weight attached to an outcome depends on the number of, as well as on the combined probability of, the disjoint events in which that outcome occurs. An experiment is reported that discriminates between juxtaposition effects and ESEs under conditions of both complete and incomplete information. The results confirm that juxtaposition effects are indeed largely due to ESEs and are robust over different informational conditions.

Are Event-Splitting Effects Actually Boundary Effects?

An event-splitting effect (ESE) occurs when an event, which yields a positive outcome in one lottery but zero under another, is separated into two sub-events and this increases the relative attractiveness of the former lottery. ESEs are attributed to either (1) a preference for more positive outcomes in lotteries, or (2) an aversion to more zero outcomes. An experiment is reported which replicates ESEs whilst controlling for explanation (2). This favours explanation (1), but a lack of generality in replication suggests previous ESEs to be partially due to (2). Implications are discussed for a related choice phenomena, decision theory, and real-world decision-making.

The common consequence effect: testing a unified explanation of recent mixed evidence

The common consequence effect is perhaps the most investigated violation of expected utility theory, and yet no single generalised expected utility model convincingly organises the experimental data. This paper reports a test of a modified version of expected cardinality-specific utility theory with boundary effects [Neilson, W.S., 1992, Economics Letters 39, 275-278] which can explain some recent mixed evidence without recourse to subjective probability weighting models. The results do not convincingly support the boundary effect hypothesis, and neither support the more usual probability weighting arguments, but do indicate instability in preferences largely consistent with event-splitting effects.

Non-Transitive Choice: Event-Splitting Effects or Framing Effects?

Recent studies have examined possible causes of the robust empirical failure of the transitivity axiom of expected utility theory by pitting regret aversion against alternative explanations such as event-splitting effects. These tests show that cycles replicate when the latter are controlled, but are sensitive to changes in problem representation. The control for event-splitting effects, however, does not rule out their contribution to cyclical choices in some circumstances. An experiment is reported which investigates this possibility. Cyclical choices are observed that cannot be due to event-splitting effects, but appear attributable to within-event and between-act evaluations of decision problems plus framing effects. Copyright 2001 by The London School of Economics and Political Science

More mixed results on boundary effects

Abstract Neilson (1992) develops a model involving preferences over numbers of probable outcomes in lotteries which explains recent evidence on the performance of expected utility theory. Here the model is modified to explain additional evidence concerning numbers of outcomes, and the implications of the modifications are discussed.

Do anchoring effects underlie event-splitting effects? An experimental test

Abstract Starmer and Sugden ( Journal of Risk and Uncertainty , 1993, 6, 235–254) and Humphrey ( Journal of Risk and Uncertainty , 1995, 11, 263–274) report experiments investigating the event-splitting phenomenon in experimental tests of theories of decision making under risk and uncertainty. This paper reports a test of whether the anchoring and insufficient adjustment of probability assessments could underlie these findings and concludes that it does not.

Probability Learning, Event-Splitting Effects and the Economic Theory of Choice

This paper reports an experiment which investigates a possible cognitive antecedent of event-splitting effects (ESEs) experimentally observed by Starmer and Sugden (1993) and Humphrey (1995) – the learning of absolute frequency of event category impacting on the learning of probability of event category – and reveals some evidence that it is responsible for observed ESEs. It is also suggested and empirically substantiated that stripped-down prospect theory will accurately predict ESEs in some decision making tasks, but will not perform well in others. This contention, it is argued, is indicative of fundamental descriptive shortcomings in the economic conception of choice under uncertainty and may entail implications beyond the direct concerns of this paper.

Consumption experience, choice experience and the endowment effect

We report experiments investigating how experience influences the endowment effect. Our experiments feature endowments which are bundles of unfamiliar consumption goods. We examine how a subject’s willingness to swap items from their endowment is influenced by prior experiences of tasting the goods in question and by prior experiences of choosing between them. We do not find a statistically significant endowment effect in our baseline treatment and, because of this, we are unable to test for an effect of consumption experience. We do find an endowment effect when the endowment is acquired in two instalments and, in this setting, we find some evidence that choice experience increases trading. In a follow-up experiment, we find evidence that the absence of an endowment effect in our baseline treatment is due to subjects being more willing to swap when they do not have to give up the last unit of their endowment.