Introduction

In many fields of occupation, there are regular and agreed-upon cycles to the job market. For example, because college students generally graduate in June, many employers focus their efforts for hiring entry-level positions to begin work soon thereafter. Some employers are very highly sought after for their level of pay and the future career opportunities of their employees. Consider a less-sought-after employer who has difficulty offering salaries that are similar to those commonly offered in the market. Very often, in such markets, less-soughtafter employers start their process of seeking recruits earlier than their peers and make offers that expire before their competitors will make their initial offers. Such exploding offers are designed to put the recruit in a quandary. Consider a very qualified recruit with an exploding offer. He can take the offer and obtain the lower pay and a secure job with certainty. Alternatively, if he rejects the offer, he takes his chances and potentially fails to find a more desirable job. If the salary offer is low enough, he might have a very high probability of obtaining a more-lucrative offer when the real cycle of offers begins. Nonetheless, it can be very difficult to turn down a job without another offer in hand. People often face problems of decision under risk. In general, models of decision under risk consist of two-component models: a model of preferences over outcomes and a model of risk perception. Rational models of decision under risk depend heavily on the assumption that people understand the potential outcomes of any risky choice and the probability of each of those outcomes. In the previous chapters, we discussed several anomalies related to how people deal with probabilistic information. Behavioral models of risky choice generally base the component model of beliefs on behavior observed in experimental settings. For example, people seem to treat certainty in a very different way than probabilistic outcomes, an effect that might make exploding job offers more profitable for inferior firms. Or people may be reluctant to invest in the stock market despite higher average returns, opting for low-returning savings accounts. Interestingly, experimental observations of choice under risk sometimes contradict the common behaviors discussed in the previous chapters. Examining the economics of decision under risk is a bit of a challenge. In a standard consumption context, we may observe whether someone chooses to purchase an apple or not. We know with relative certainty that the person understood the characteristics of either choice and we can also know with relative certainty what those characteristics are. In the context of risk, we may observe whether someone chooses to purchase a share of a particular stock, and we can observe subsequent changes in the value of that stock. However, we will never easily be able to observe what the purchaser thought the probability of each possible outcome was or the actual probability distribution. By basing choice models on experimental results, theorists have hoped to sidestep this measurement issue. In an experiment, the experimenter has the freedom to select the probabilities and outcomes and specify them exactly. This control is a substantial advantage relative to field observations, but it also creates a potential weakness. The experiments can present risky choices that do not represent the real-world choices that people face. Many decisions are made without a clear understanding of what choices are available, what the possible outcomes of the choices are, or what the relative probability of those outcomes may be. For example, consider college freshmen considering their choice of major. Although it may be possible to determine all the possible majors available in a university, it may be difficult to know what sorts of knowledge and potential employment options would be available with each possible major. Such ambiguity about the possibilities leads nearly half of all students to change majors at least once as they discover new information that leads them in a different direction. Many students put off declaring a major until the last possible moment. Moreover, some students gain excessive credits in their search for the right major. This has led some universities (for example, the University of Wisconsin) to charge students who take too many credits, trying to encourage earlier decisions and earlier graduations. The latter portion of this chapter describes economic theories of how people deal with ambiguous choices such as these, as well as their impact on outcomes. The ease with which people can introduce and test models of decision under risk has led to the introduction of dozens of competing models. In this chapter and the next I present the most important behavioral concepts arising from the behavioral risk literature, including only a small selection of the models that have been proposed. The interested reader who wishes to see a broader treatment of models of decision under risk is directed to the literature review compiled by Chris Starmer.