What a story!

My succinct Strait of Hormuz synopsis:

I'm delighted that our paper, "Risk Matters Less When Options Are Apples-to-Oranges: The Translate-and-Accommodate Model" (Evan Weingarten, Yuval Rottenstreich, and George Wu) has just been accepted at Psychological Review.

papers.ssrn.com/sol3/papers....

1/N

Thanks. Will do.

Ok, that's an overview of the paper.

There's more, we think, in the paper.

papers.ssrn.com/sol3/papers....

N/N

It also can generate a "reverse uncertainty effect," a risky option valued above its best possible outcome (Goldsmith & Amir, 2010; Shen et al, 2015). According to our model, it isn't one bizarre pattern of risk preferences or the other, but both kinds of uncertainty effect can co-exist.

20/N

The graph below plots how combinations of tau and alpha map into different preference patterns. The green is what we document in our paper, risk mattering less.

The orange is the uncertainty effect. Importantly, this region gets larger as unimodal risk aversion increases.

19/N

We found that people value a $100 Barnes & Noble gift certificate at $45, a $50 gift certificate at $26, but a lottery that offered equal chance at both at $16!

This result has since been replicated.

18/N

The Uncertainty Effect: When a Risky Prospect is Valued Less than its Worst Possible Outcome Abstract. Expected utility theory, prospect theory, and most other models of risky choice are based on the fundamental premise that individuals choose amon

It turns out that certain combinations of tau and alpha are consistent with risk mattering less.

Surprise: the model also provides an explanation for the uncertainty effect, a finding that Uri Gneezy, @johnlist.bsky.social, and I wrote up 20 years ago. academic.oup.com/qje/article-...

17/N

We build an algebraic model that includes a translation term (tau) and an accommodation term (alpha). Tau = 1 and alpha = 1 corresponds with unimodal choice.

I won't explain the notation (you can start at page 67 of papers.ssrn.com/sol3/papers....) but you can see that it is pretty simple.

16/N

This theorizing builds on incommensurability and similarity, studies of judgment given alignable versus non-alignable differences, and classic results on anchoring and adjustment. It also builds on empirical findings on affect-driven probability neglect by Pachur, Hertwig, Suter, and others.

15/N

We posit that "some things are lost in translation" -- in other words, translation is somewhat crude rather than nuanced.

We also suggest that translation is likely to be primary (or come first) and then risk accommodation is secondary.

14/N

Unimodal choices only require risk accommodation. When considering a choice between a sure 1 unit and a 50-50 chance of 0 or 4, the gamble has an advantage (4 vs. 1) and a disadvantage (0 vs. 1).

Crossmodal choices require this PLUS translation of apples and oranges.

13/N

What's driving this pattern? We model these choices with a simple model that builds in two psychological processes: risk accommodation and translation.

12/N

I won't go into the details, but there are other studies and analyses that rule out the role of noise (think, regression to the mean) and preference heterogeneity (think, a subset of people who like Purell and hate Scotch Tape and others who like Scotch Tape and hate Purell).

11/N

There are plenty of other studies in which we vary the parameters of the gamble or the probabilities.

10/N

Here, we see what we mean when we say "risk matters less ..." -- the red bars (crossmodal preferences, Purell vs. Scotch Tape) are less sensitive to risk parameters (here, the attractiveness of the sure thing) than the blue bars (unimodal preferences, Purell vs. Purell).

9/N

We also manipulate the relative attractiveness of the gamble to the sure thing. For example, favorable involves 3 units for the sure thing compared to 0 or 4 for the gamble. Unfavorable involves 1 unit vs. 0 or 4.

Here are choice percentages aggregated over different product combos.

8/N

We don't use apples and oranges in our studies, but household products like travel-sized Purell (P) vs. a roll of Scotch tape (S).

We compare a sure thing (e.g., 3 Ps) and a gamble (say 0 or 4 Ps). We then look at all combinations of choices, P-P, P-S, S-P, S-S (Sure Thing-Gamble).

7/N

Then the independence axiom implies that you should be indifferent between 1 apple for sure and a lottery that offers 0 or 3 oranges.

This basic condition is violated. Risk matters less in apples-to-oranges comparisons than applies-to-apples comparisons.

6/N

Suppose that you are indifferent between an orange and an apple and n oranges and n apples. Suppose also that you are indifferent between 1 apple for sure and a lottery that offers 0 or 3 apples (same for oranges).

5/N

We find that "risk matters less" when choices are "apples-to-oranges" than "apples-to-apples." More precisely, people are less sensitive to relevant parameters, such as quantity, quality, and probability.

4/N

This distinction was introduced by Cubbitt, McDonald, and @danielread.bsky.social (2018) in a study of time preference.

pubsonline.informs.org/doi/abs/10.1...

They extended this to risk in 2023:

onlinelibrary.wiley.com/doi/full/10....

3/N

The paper compares risk preferences in "unimodal" settings (apples-to-apples; think the canonical monetary lottery paradigm) and "crossmodal" settings (apples-to-oranges; think a some sure cash versus a lottery over the duration of a vacation).

2/N

I'm delighted that our paper, "Risk Matters Less When Options Are Apples-to-Oranges: The Translate-and-Accommodate Model" (Evan Weingarten, Yuval Rottenstreich, and George Wu) has just been accepted at Psychological Review.

papers.ssrn.com/sol3/papers....

1/N

Congrats to my colleague @erikakirgios.bsky.social!

Late 19th century military building designed with castle-like features. A huge advertisement for a Chevrolet car dealership is painted across its facade. Another side advertises roller hockey. The building faces an empty lot. A lone pedestrian walks across the desolate scene.

The old Illinois National Guard Armory, Chicago (1966)
Photo: Richard Nickel / Art Institute of Chicago

IL-02

Can you spare a minute to help this campaign? STOP the Demolition of the Chicago Stock Exchange Trading Room!

Help save the fabulous Chicago Stock Exchange Trading Room that the Art Institute of Chicago wants to demolish for a new building. Dedicated architects and philanthropists spent years and more than a million dollars to reconstruct it at the Art Institute. Sign the petition c.org/L5GJtwLRTM

This is one of my favorite places in Chicago. And I have a LOT of places that I love.