Choice by chance : an exploration into the practice and virtues of using lotteries in public choices

Abstract

Sometimes distributional schemes for scarce goods and public roles include a lottery element. Lottery schemes are seen in a variety of contexts, from the selection of jurors and the conscription of soldiers, to the allocation of educational places and affordable housing. The use of lotteries in our allocative practices provokes two questions. First, what exactly is the practice of allocation by lottery, and how widespread and important is it? Second, under what circumstances and why would it be justifiable to use a lottery in a distributive scheme? The first part of this dissertation - which includes Chapters 1 and 2 - is devoted to answer-ing the first question. Chapter 1 gives an overview of the main applications and most important examples of allocation by lottery. It shows that lottery schemes have been and still are employed in a wide variety of, often very important, public choices. It also emphasizes the great variety in the details of lottery schemes, and their frequently high degree of complexity. Chapter 2 provides a characterization of a lottery, and distinguishes equiprobable from weighted lotteries. It argues that the process employed in any lottery has five main characteristics, and that the process's probabilities might be interpreted in multiple ways. I also introduce a distinction between simple and complex lotteries, and stress that a complex lottery can provide participants with unequal probabilities of success without being weighted. The second part of this dissertation-which includes Chapters 3, 4, 5, and 6 - covers various topics regarding when and why we would want to employ a lottery. Chapter 3 considers what general approach we should take to justifying lotteries. I argue for pluralism in the justification of lotteries: there will be varied and multiple reasons that can justify lotteries in different contexts, and these reasons are often grounded in the interplay between different institutional features. Chapter 4 analyzes what types of distributive procedures might be appropriate when claims to a scarce good are equal. I contend that only certain lottery procedures seem to fully respect equal claimants, because they are fair and meet other relevant standards for treating claimants properly. Additionally, I argue that in many, if not all cases, it would be fairer not to allocate the scarce good at all, rather than to hold a lottery. Chapter 5 entertains the position that a lottery's contribution to fairness plays a main role in its general justifiability. I maintain that this position is only plausible, if the contribution of lotteries to fairness is understood pluralistically and not exclusively related to respecting equal claims. I identify two additional ways of how lotteries might be understood to contribute to fairness. Chapter 6 considers whether inadequate epistemic capacity can ground the claim that all proposals for randomly selected citizen assemblies are epistemically indefensible. I argue that, with the appropriate design, a citizen assembly would at least not have substantially worse epistemic capacity than an elected assembly, and that it may even have more epistemic capacity given its advantage in terms of diversity. Show more