This paper investigates the influence of strategic choice and strategic luck on individuals’ preferences for redistribution, extending the existing literature on inequality and fairness in experimental economics. While much of the previous research has examined non‑strategic settings where individuals’ decisions are independent and luck is typically conceptualised as a lottery, this study introduces a strategic environment where incomes are shaped by both individual choices and the decisions of others. Furthermore, luck is redefined as a factor determined by others’ choices rather than a random lottery. Each participant participates in two blocks: Block 1 introduces a novel strategic environment based on a pure coordination game (PC), followed by redistribution over 30 rounds; Block 2 employs a standardised adaptation of the experiment from Cappelen et al. (2013) to measure preferences under non‑strategic choice and luck. The findings reveal that factors such as initial income distributions exert more significant effects on individuals’ preferences for redistribution compared to the impact of strategic and non‑strategic choice and strategic and non‑strategic luck.
Sheheryar Banuri, Amir Jafarzadeh, Robert Sugden
This paper investigates whether individuals’ redistribution preferences are influenced by initial inequalities resulting from inequalities of opportunity, a key concept in debates on fairness. Drawing on the distinction between procedural fairness (focused on equal access to opportunities) and ex‑post fairness (focused on the fairness of outcomes), we design a novel experiment to test whether individuals’ preferences for redistribution change when opportunities are equal or unequal. We conducted the experiment within a new game‑theoretical framework based on Sugden & Wang (2020), in which participants engage in a card game with varying opportunities and then decide how much to redistribute in a dictator game. The results show that participants’ redistribution preferences remain unchanged despite manipulation of equality of opportunity, suggesting that they prioritise outcomes over procedural fairness. This finding aligns with some research in the literature, arguing that most individuals focus more on outcomes than fairness in process. By exploring the relationship between procedural fairness and redistribution preferences, this paper contributes to the ongoing debate on the role of equality of opportunity in shaping perceptions of fairness in economic contexts.
Amir Jafarzadeh, Robert Sugden
It is well known that in Pure Coordination (PC) games in which strategies are distinguished by payoff-irrelevant labels, players are often able to coordinate their strategy choices by choosing the most ‘prominent’ label. The same prominence cues are known to influence behaviour in other games with multiple Nash equilibria. However, game theory has lacked a predictive theory of the determinants of prominence. In this paper, we develop and implement a general method of defining and validating empirical concepts of prominence, and apply it to three general concepts discussed in previous literature. We use an experimental data base of $2\times2$ PC games constructed randomly from 60 familiar labels belonging to 10 familiar categories. We find that participants can coordinate successfully in most categories. The frequencies with which labels are chosen are well predicted by ‘book rank’---the category-specific frequency with which each label’s text string appears in recently published books. This is a measure of each label’s mere appearance in discourse, in contrast to the less predictive concepts of ‘search rank’ (being the object of current interest) and ‘web rank’ (being highly ranked in terms of popularity, fame or significance). The predictive power of book rank suggests that players of PC games align their expectations by using intuitive mental processes of interactive alignment rather than by conscious use of principles of rationality. We find further support for this conclusion in the shortness and stability of players’ response times and in the absence of individual-level heterogeneity in use of prominence concepts.