The rationality of qualified lotteries

Abstract

A lottery (or random selection) is often considered to be irrational. However, a qualified lottery can lead to a second‐order rationality on an institutional level. The main idea is to make use of uncertainty, either by exploiting existing fundamental uncertainty or by deliberately enlarging uncertainty through lotteries. In both cases, decision quality may be enhanced by increasing diversity and risk diversification, by decreasing biases and noise, and through a strong disempowering effect by the shadow of uncertainty. The paper shows the different ways in which qualified lotteries have been used in history, analyses why this procedure has been forgotten despite its clear advantages, and compares it with competitive selection mechanisms.

Judgment under radical uncertainty: Epistemic rational heuristics

“…Hypothesis formulation and validation can actually be quicker and more reliable if judgments interact, and complementary bits of information are brought into the field of attention. In other words, “more is less” (effort and time), rather than “less is more.” A case in point may be again drawn from the studies on military pilots (Grandori, 2015): pilots never fly alone, and they are obsessively trained to listen to co‐pilots, to neglect hierarchical grades and status games, and to behave like a single expanded brain, whereby scanning for possibly relevant signals is augmented and the likelihood of error reduced. Qualified lotteries : rather than counting on the wisdom of crowds, Frey et al (2023) continue to build on ancient Greek wisdom (Zeitoun et al, 2014) in reminding us of that what we may count on is the wisdom of qualified crowds , in which the wisdom of knowledge and the wisdom of randomness are combined, whereas the random draw of decision makers can tone down fights based on selfish sub‐goal pursuit and pave the way to knowledge based, free‐minded “deliberative decision‐making” (Elster, 1989; Estlund, 2008). …”

Artificial intelligence and radical uncertainty

“…Future research could explore how the integration of AI changes the costs associated with comprehensive search or how human creativity might be elevated by the rapid and inexpensive provision of 'good enough' inputs. Frey et al (2023) develop the idea of using qualified lotteries to harness uncertainty instead of fighting it. Similar to qualified lotteries, AI may conceal information that feeds into decision-making.…”

“…Frey et al (2023) develop the idea of using qualified lotteries to harness uncertainty instead of fighting it. Similar to qualified lotteries, AI may conceal information that feeds into decision‐making.…”

The quest for non‐Bayesian decision‐making heuristics: Towards a logic of imagination

“…This statement not only means that indicators of utility are controversial or weak, it accounts also for the belief that utility evaluations cannot be made better by some additional information; that is, learning itself is rejected. This direction of work is explored by the paper by Frey et al (2023, this SI) on the rationality of qualified lotteries. Such extreme heuristics target situations that may appear in human resources management or when there is the need to select important executives while acknowledging that, after a first shortlisting, judgment about the future behavior of remaining candidates is difficult and controversial.…”