This article examines insurance choices under different contract types through a laboratory experiment. We find that proposing a single insurance policy for multiple risks, known as bundled insurance, does not necessarily reduce the demand for coverage relative to the risk carried by the individual, as the effect depends on the level of risk aversion, in line with Expected Utility Theory predictions. In contrast, offering separate contracts increases the number of insured individuals but exposes insurers to greater adverse selection. We also test a new type of insurance called semi-bundled insurance, which lies between separate and bundled insurance and requires the insured to choose a minimum number of risks to cover. Relative to the shift from separate to bundled insurance, semi-bundled insurance produces a smaller improvement in the management of adverse selection and a smaller reduction in the number of insured individuals. It does not significantly affect relative coverage, making it a potential compromise between separate and bundled insurance. These findings provide promising perspectives for addressing individuals’ willingness to choose hazards to be covered while maintaining a minimum diversification of risk, which is essential for the sustainability of insurance.
In this paper, we consider resale in efficient auctions. The potential gains from trade arise from a delay in resale which reduces the bidders’ values. We consider two information states during resale: (a) complete information where all the values and bids are revealed and (b) incomplete information where bids and values are not revealed. Under complete information, we establish revenue equivalence between the first- and second-price auction for a family of trade rules where the market power is distributed between the reseller and buyer. We also show that, if all the market power lies with the reseller (resp., buyer), it is optimal (resp., not) to reveal information.
Behavioral models can accommodate some anomalous social choices that result from aggregation of individual preferences. However, the use of behavioral methods of revealed preference in social choice may produce new conceptual difficulties that do not exist under standard economic models. One such difficulty is the problem of unanimity in revealed preference: individual choices may reveal that every individual strictly prefers x to y and yet y is revealed to be strictly socially preferred to x. Under plurality rule, the problem of unanimity in revealed preference holds for some behavioral methods of revealed preference, but not for others.
Decision-making processes often involve ambiguity rather than risk, and multiple ambiguities rather than a single one. In this article, we consider how different sources of ambiguity, including two-source ambiguity, affect decision-making in relation to risk. We also examine the value of information that eliminates or reduces ambiguity. Finally, we analyse the effect of ambiguity preferences on the results. To this end, we propose an experiment in forest management in the context of climate change, a typical decision-making situation involving multiple ambiguities. In light of our empirical analyses, several key findings emerge. Firstly, on average, the private forest owners in our sample exhibit both risk aversion and ambiguity aversion. Secondly, the impact of ambiguity on the optimal cutting age of forest stands appears to be influenced by storm frequency. Thirdly, the estimates generally highlight the negative effect of ambiguity aversion on the optimal cutting age. Fourthly, the value of information that resolves ambiguity from a single source is consistently positive. However, this value is greater when the information replaces a situation of ambiguity with one of risk, rather than when it merely reduces ambiguity by replacing two sources with one. Finally, the results reveal no statistically significant effect of ambiguity aversion on the value of information.
In this paper, we report the results of two experiments that attempt to elicit social image effects. The first experiment ( N = 1,252 ) provides little evidence that individuals behave in more ‘prosocial’ ways when their choices are disclosed to other participants. If anything, imposing observability appears to make the participants slightly less prosocial, although this effect is not statistically significant. The second experiment ( N = 750 ) generates similar results and further suggests that our results are not dependent on the omission (or inclusion) of ranking information. We discuss why our experiments fail to generate the results that we had expected and why our results differ from those in the published literature.
While the effects of overconfidence on economic outcomes are well-documented, the implications of confidence sensitivity, the extent to which confidence judgments are informative about true abilities, have been largely overlooked. Yet in many contexts, a lack of sensitivity may be just as detrimental as overconfidence, as it prevents individuals from accurately identifying their strengths and weaknesses, a skill that is particularly crucial in educational and career choices. Moreover, recent research in cognitive psychology suggests that individuals can form informative beliefs about their own confidence sensitivity, but whether such “meta-confidence” influences their behavior remains an open question. To shed light on these issues, we design a laboratory experiment studying individuals’ compensation scheme choices using a canonical economic framework (Niederle Vesterlund, 2007), in which we measure confidence sensitivity and meta-confidence and allow participants to acquire information about their own ability prior to making their choice. Our results show that higher confidence sensitivity significantly increases participants’ earnings by making them more likely to choose payoff-maximizing compensation schemes. In addition, using insights from a stylized model of confidence formation in which individuals hold beliefs about the informativeness of their own confidence, we derive testable implications for how meta-confidence shapes individuals’ reliance on their confidence judgments and their feedback acquisition strategies. We test these implications causally by implementing a manipulation in our experiment designed to reduce participants’ meta-confidence. We find that lower meta-confidence does not affect individuals’ reliance on their confidence when choosing compensation schemes, but significantly increases their willingness to seek performance feedback prior to making their decision.
This paper considers decision problems involving partial information, and presents a conception of commodification of information under which a buyer with limited probabilistic sophistication can reduce any resulting ambiguity through purchases of new probabilistic information. In the specific formulation articulated herein, a lottery of information batches helps overcome the induced holdup during the contracting process. Under the proposed formulation, additional noise in the lottery leads to a reduction in the randomness of the buyer’s ambiguity by the introduction of an expectation operator for random sets of beliefs. We show that the buyer’s utility function for information, formalized as Hurwicz expected utility, is monotone, continuous and linear in information. We then discuss the value of information within the corresponding decision problem and contrast it with the influential formulation of Radner–Stiglitz.
How can we understand rational choice when decision-makers face genuine non-comparabilities and incommensurabilities? This paper examines two fundamental but distinct coherence requirements for rationalizing relations: quasi-transitivity (which ensures that the betterness relation is transitive) and Suzumura consistency (which prevents exploitable cycles in goodness relations). While each condition individually relaxes transitivity in plausible ways, their combination has remained unexplored despite addressing complementary aspects of coherent behavior; quasi-transitivity ensures path-independent decision-making, while Suzumura consistency rules out money pumps. We provide a novel analysis of when choice patterns can be rationalized by goodness relations satisfying both properties simultaneously. Our argument offers a plausible foundation for decision-making, when transitivity is dropped under circumstances where incommensurability or non-comparability may be present.
The Owen value is a well-known allocation rule for TU-games with coalition structure. It can be characterized by the axioms of efficiency, differential marginality within unions, differential marginality between unions, and the null player property. In this paper we weaken the first three axioms into gain-loss, weak differential marginality within unions, and weak differential marginality between unions, respectively. We show that these three weaker axioms, together with the dummy player property which is stronger than the null player property, yield a new characterization of the Owen value.
The idea that people make mistakes in how they pursue their own best interests, and that we can identify and correct for these mistakes has been central to much recent work in behavioural economics, and the ‘nudge’ approach to public policy grounded in it. The focus in this literature has been on individual choices that are mistaken. Agreeing with, and building on the criticism that this literature has been too quick to identify individual choices as mistaken, I argue that it has also overlooked a kind of mistake that is potentially more significant: irreducibly diachronic mistakes, which occur when series of choices over time do not serve our interests well, even though no individual choice is mistaken independently of being part of the series. I argue for the claim that people make such mistakes, and reflect on its significance for welfare economics.
It is well known that ex-ante social preferences and expected utility are not always compatible. In this note, we introduce a novel framework that naturally separates social preferences from selfish preferences to answer the following question: What specific forms of social preferences can be accommodated within the expected utility paradigm? In a departure from existing frameworks, our framework shows that ex-ante social preferences can be reconciled with expected utility at the level of game utility, provided the corresponding non-expected utility component is absorbed by selfish utility. We also show that when a player’s preferences in both the game (against another player) and the associated decision problem (against Nature) conform to expected utility axioms, the permissible range of social preferences becomes notably restricted. Only under this stronger joint requirement do we reaffirm the existing literature’s key insight regarding the incompatibility of ex-ante inequality aversion with expected utility.
There is evidence from experiments, using real-effort tasks and tournament incentives, documenting women performing worse than men under competition. We question whether this perhaps is due to the very particular way in which competition is typically studied in lab experiments. We alternatively model competitive behavior in firms by letting experimental participants compete in bidding for salaries (prizes) of different sizes in flat or steep organizational hierarchies. In our representation of job competition via a competitive bidding process, bids sum up, in a stylized way, various monetary aspects of how one competes for higher positions in organizations. Higher bidding for positions corresponds to behaving more competitively. We mostly find no statistically significant differences between women’s and men’s bidding. Women do win the top positions significantly more often, but there are no differences in earnings, the difference between salaries and bids. These results may function as a counterbalance to results finding women less willing to compete in head-to-head (real-effort) competition. One approach, real-effort versus pure choice, is not superior to the other. They likely yield complementary insights on the same issue, like we believe they do in this case.
This paper aims to formalize the abstract algebraic difference between Keynes’s and Ramsey’s theories of probability. Drawing on the foundational paper of Birkhoff and von Neumann (1936) on quantum mechanics, the algebraic-axiomatic properties underpinning the relation between belief and probability in Keynesian and Ramseyan theories are identified. The paper demonstrates that a specific class of abstract algebras - bounded distributive lattice - can represent Keynes’s problem while sharing key properties with traditional Ramsey’s probability theory. By introducing the notion of interval probability measure, and assuming a model of uncertainty, keynesian uncertain beliefs can be represented as isomorphic probability intervals, offering a coherent resolution to Ramsey long-standing challenge.
Dekel et al. (1998) showed that standard state-space models preclude non-trivial unawareness. We argue that their introspection axioms may be too strong and instead explore two new axioms on unawareness, weak KU introspection and U-monotonicity. We show that even if we relax their axioms, when employing these new axioms, we arrive at the same conclusions as theirs. Results illustrate the relationships among various axioms that have not been previously explored. Based on them, we also discuss briefly about the possibility of representing non-trivial unawareness within the standard state-space approach.
If we break an instant runoff voting (IRV) election into stages or rounds by tallying votes, identifying the plurality winner, eliminating the candidate with the fewest votes and redistributing their votes, then repeating this process, we obtain a series of plurality winners (i.e. so-called transient winners), one for each stage. This process models the situation that arises in several real-life scenarios such as the Tour de France cycling race and in some political leadership contests. But how many different transient winners could be obtained in this way? Here we explore upper bounds for the number of different transient winners that could be obtained in single peaked models of IRV elections, and show that these bounds can be met. Through simulation we demonstrate that the number of possible profiles that meet the bound is likely very small.
Intransitive indifference is a well-documented phenomenon in which the decision maker is forced to choose between alternatives with a subtle difference. In this paper, we establish a general model that extends the semiorder/interval order approach (e.g., Fishburn in J Math Psychol 7: 144–149, 1970a. https://doi.org/10.1016/0022-2496(70)90062-3 ; Luce in Econometrica 24: 178–191, 1956. https://doi.org/10.2307/1905751 ) by adopting the set of lotteries as the domain of choice and a direction-dependent just-noticeable difference function. The model can distinguish two classes of intransitive indifference, that is, imperfect discrimination, which is relevant to previous studies on intransitive indifference, and uncertainty about tastes, which is relevant to incomplete preferences. The main theorem axiomatizes the essentially unique expected utility with direction-dependent sensitivity representation. The key axioms for this characterization are irresolute independence, wherein mixing alternatives with another alternative may change a strict preference to indifference while preserving indifference, and strict preference convexity, which derives the convexity of strict upper and lower contour sets. We also obtain two special cases of our model—one-directional and categorical sensitivity—which highlight the two classes of intransitive indifference, and discuss a possible change in the domain of choice to a vector space.
Building on work by Bradley and Stefánsson (British J Philosophy Sci 68(2): 485–533, 2017), we consider counterfactually-dependent preferences in a dynamic setting. In particular, we consider the interaction between such preferences and rational intention. We point out a hitherto unrecognised role for intentions: they can fix what would have been and thus influence the rationality of choosing certain actually available options. We furthermore argue that orthodox sophisticated choice can be understood in two ways in this setting. Actualist sophistication faces the problem of being overly restrictive and of potential probabilistic incoherence. Possibilist sophistication avoids these problems but faces the problem of being overly permissive and of violating the spirit of forward-looking consequentialism. Indeed, we show that possibilist sophistication can rationalise behaviour otherwise associated with resolute choice.
Individuals facing economic scarcity often struggle with delaying gratification, a pattern typically attributed to cognitive load, self-control, or risk preferences. However, the role of uncertainty remains underexplored. This study tests the preregistered hypothesis that uncertainty intensifies the negative effects of scarcity on delay of gratification. In a between-subjects experiment with Colombian participants (N = 230; M age = 24.4; 58
To aggregate rankings into a social ranking, a natural approach is to use scoring systems such as Plurality, Veto, and Borda that assign scores to candidates. We distinguish three types of methods built on scoring systems: ranking by score, ranking by repeatedly choosing a winner, and ranking by repeatedly choosing a loser. The latter method captures the frequently studied voting rules Instant Runoff Voting (IRV), Coombs, and Baldwin. We compare these classes of methods axiomatically, referencing prior results. In an experimental analysis, we show that the three types of methods produce different rankings in practice. We also provide evidence that sequentially selecting winners is most suitable to detect a ground truth ranking of candidates. For different rules in our classes, we then study the (parameterized) computational complexity of deciding in which positions a given candidate can appear in the chosen ranking. As part of our analysis, we also consider the Winner Determination problem for IRV, Coombs, and Baldwin and determine their complexity when there are few voters or candidates.
