Login 

Share

DANIEL ELLSBERG AND JOHN MAYNARD KEYNES MEET AT AN URN: New research on the impact of ambiguity and complexity on decision-making

New research identifies a perception-based trait that lies at the heart of a decision-making paradox attributed to a 1961 study, ‘Risk, Ambiguity, and the Savage Axioms’ by Daniel Ellsberg, who is best known for leaking the Pentagon Papers, a top-secret study of US government decision-making about the Vietnam War.

The new experimental study by Soo Hong ChewMark Ratchford and Jacob Sagi, published in the September 2018 issue of the Economic Journal, finds that more than two thirds of participants who understand and think carefully about a choice between a bet with known odds and an equivalent bet with unknown odds pick the former. These people, who exhibit behaviour that Ellsberg labelled as ‘ambiguity aversion,’ are younger, more educated, take more time and write more in reflecting on their choices.

In contrast, participants with less understanding of the choice between known and unknown bets tend to choose randomly. They exhibit behaviour that is less ambiguity aversion and more ‘complexity avoidance’.

The researchers’ findings have implications for the design and formulation of policies relating to the perception and communication of risk, including the comprehension and attentiveness of the people at whom the policies are targeted. They also point to the value of minimising ambiguity in policy design.

The study begins with Ellsberg’s thought experiment, which involved betting on the colour of a ball drawn from either an urn containing 50 black balls and 50 red balls, or another urn containing 100 black and red balls in an unknown ratio. Ellsberg theorised that people would tend to choose the known bet in the first urn and avoid the ‘ambiguous’ bet in the second urn, where the winning odds are uncertain.

Forty years earlier, John Maynard Keynes in his Treatise on Probability suggested that people may favour the known bet even if they also assign 50-50 winning odds to betting on the unknown urn. Keynes famously asked, ‘If two probabilities are equal in degree, ought we, in choosing our course of action, to prefer that one which is based on a greater body of knowledge?’

The new study investigates two questions:

• Do people avoid the unknown bet because of its greater complexity rather than its ambiguity?

• Relatedly, for those who do recognise this ambiguity, how might the different perspectives of Keynes and Ellsberg influence their choice behaviour?

To address the first question, the authors screen subjects for comprehension and attentiveness before presenting them with more complex versions of Ellsberg’s known and unknown bets as illustrated in Figure 1 below.

Deck #1 with ten red and ten black cards corresponds to the known urn while Deck #2 comprising twenty cards without knowing its colour composition corresponds to the unknown urn.

Like the known bet, Option B pays $100 if Red is drawn from Deck #1 regardless what is drawn from Deck #2. Option A is akin to the unknown bet. The subject wins if Red is drawn from Deck #2 and the specific amount is $100 or $101 depending on whether Red or Black is drawn from Deck #1.

The authors find some support for the idea of complexity avoidance among low-comprehension subjects who tend to choose randomly between Option A and Option B, even though the same subjects often exhibit ambiguity aversion when facing the simpler original version of Ellsberg’s two-urn problem. By contrast, 69% of high-comprehension subjects continue to exhibit ambiguity aversion in favouring Option B.

To address the second question on the influence of how the unknown bet is perceived, the authors classify high-comprehension subjects as Keynesian or Ellsbergian depending on whether they assign 50-50 winning odds to the unknown bet or consider such winning odds to be indeterminate.

They find that Ellsbergians outnumber Keynesians by two to one. As anticipated, Ellsbergian subjects are far more likely to avoid Option A. Comparing the two types, Ellsbergian subjects are younger, more educated, take more time and write more in reflecting on their choices, despite Keynes’ perspective being seemingly closer to subjective probability thinking.

In applications, the reported findings point to the need to incorporate considerations involving comprehension and attentiveness, and to minimise ambiguity in the design and formulation of policies relating to the perception and communication of risk at the corporate, market and public levels.

You Have to Recognize Ambiguity to Avoid It’ by Soo Hong Chew, Mark Ratchford and Jacob Sagi is published in the September 2018 issue of the Economic Journal.