Michael Lewis and the parable of the lucky man taking the extra cookie (2017) | Hacker News Reader

Michael Lewis and the parable of the lucky man taking the extra cookie (2017)

kottke.org

51 pointsTomte44 comments

The cookie study hasn't been published anywhere. Descriptions of the methodology vary (sometimes it's 4 cookies, sometimes it's 5 cookies). Super suss, and I wouldn't be surprised if the results don't reproduce. https://thenewstatistics.com/itns/2019/05/22/the-cookie-mons...

The paper can be accessed here [0].

[0] https://web.archive.org/web/20120915145553/https://gsbapps.s...

I see the "graph" on page 104, which shows (if anything) a slightly negative effect for men ("high power" men took slightly fewer cookies on average). I didn't see a detailed methodology anywhere.

Ugh, can't let the truth get on the way of a good story though, I guess.

The graph is also scaled awkwardly, starting at .5

That's not the point of his address though, if you read the whole thing. His cookie was getting sat next to the wife of the Salomon Brothers big shot. Whether the cookie study is apocryphal or not, his point remains.

Interesting find. But my go to research for selfishness and class is https://www.pnas.org/content/109/11/4086.

Very interesting paper. The categorisation of upper and lower class could be problematic however. Being a paper from the US, I presume the participants' class category corresponds to assets and income, ergo upper class being simply richer. In other countries, such as the UK, one's class has a totally different meaning.

For sure, but isn't this sort of an inherent definitional challenge in such research?

It's not a methodological issue. (The methodology itself, of course, has potential issues as well, e.g. as with the "luxury car" categorization.) It's foundational: can you really say, in a culture-independent way, that "higher socioeconomic class translates to more antisocial behavior?" Not without big caveats.

I wouldn't say it has a totally different meaning - upper class people are going to be more likely to be wealthier than the average but having lots of money doesn't make you upper class - at least in the traditional UK sense where upper class means titled aristocracy.

A similar paper you might like: https://psycnet.apa.org/record/2017-46375-012

The conclusion in that link seems to say that selfishness correlated with power only in women, in men it was inversely correlated with power.

EDIT: I read that wrong. In men is less correlated with power than in women, not inversely correlated.

Super suss, but rings so true.

The anecdote about the study doesn't pass basic common sense unless they happened to randomly select psychopaths for the leader every time (unlikely)

Imagine yourself in the situation of the leader and think about what you would do. Would you greedily grab the cookie and eat it "with gusto"? Or would you maybe propose any of the blindingly obvious and more fair solutions that immediately leap to mind - break the cookie into 3 equal pieces (duh), draw straws for it, rock paper scissors, ask politely if anyone else wants it, you know, like anything that a normal human being in a non-made-up story might do.

Absolutely, I recently read Rutger Bregman’s “Humankind” and it has made me extremely skeptical of these short “sounds nice” stories. Like with the Stanford experiments, they should be thoroughly reproduced or otherwise assumed to be made up.

Here's the paper but its quite long

https://web.archive.org/web/20120915145553/https://gsbapps.s...

This isn't the paper about the cookie study, this paper examines results from a number of studies, including the cookie one, which it identifies as Ward and Keltner 1998. That paper was never published.

https://thenewstatistics.com/itns/2019/05/22/the-cookie-mons...

It also doesn't claim anything like the effect that Lewis is talking about in his speech. The people in power ate, on average 1.25 cookies, which is not consistent with every single one greedily devouring the extra cookie. It also has 5 cookies for 3 people, so that at least one person could take an extra cookie without the social stigma of being the one who grabs the last cookie.

Worrying about the sugar contents I'd be happiest to pass off a second cookie to someone else anyway, even if I perceive it to be 'mine'.

This, also. Plus one for calling something a 'psychopath' when it's indeed Cluster B behavior pur sang.

The entire story lost credibility when it noted that nearly every leader drooled and had crumbs on their shirt.

I'd expect a few groups to have a person who passed on the cookie, leaving two each for the others.

> With incredible consistency the person arbitrarily appointed leader of the group grabbed the fourth cookie, and ate it. Not only ate it, but ate it with gusto: lips smacking, mouth open, drool at the corners of their mouths. In the end all that was left of the extra cookie were crumbs on the leader’s shirt.

I've never consistently seen the "highest ranking" in board rooms, meetings, whatever consistently taking the last sandwich, cookie or whatever was only one left of.

This feels like one of those things that is referenced in a dozen papers but has never been replicated.

Maybe if you are selected at "random" then you will always pick the extra cookie because you know you don't deserve it

Whereas if you "highest ranking" person you think you are deserving and expect more grace from yourself

Could very well be. And those are definitely interesting questions.

But what I find troubling is the massive extrapolation that happens next:

> This experiment helps to explain Wall Street bonuses and CEO pay, and I’m sure lots of other human behavior.

I'm not saying there is or there is not something wrong with CEO compensation. But doing an extrapolation to the scale of CEO pay is just nuts.

Would you step on an ant if somebody paid you 50 bucks? Ah then you could also be a hitman.

the highest ranking person in the room won’t take the extra sandwich, but will take the extra million to buy as many sandwiches as he wants.

in this isolated experiment the entire organization consists of 3 persons, one of them a leader, and 4 sandwiches.

real world orgs are much more complex, and sandwiches take many subtle forms

It really sounds like a fake story, and many behavioral psychology studies have turned out to be fake.

I realize this story is probably made up but I know one area where the pattern seems to ring true.

If you have three developers working for your company, all equally contributing effort but:

1. One lives in US, one from Eastern Europe, one in India.

I think we might find, more often than not, we don't divide the cookies equally among those developers and one of them quite greedily eats more than their fair share over and over.

We do a lot of things in life that we attribute to being "deserved" but none more flagrantly unequal than the seemingly arbitrary income disparity of country of residence.

Previous discussion (2012):

https://news.ycombinator.com/item?id=4065233

Thanks! Macroexpanded:

Michael Lewis: Don't Eat Fortune's Cookie - https://news.ycombinator.com/item?id=4065233 - June 2012 (180 comments)

Also:

Michael Lewis and the parable of the lucky man taking the extra cookie (2017) - https://news.ycombinator.com/item?id=16408758 - Feb 2018 (111 comments)

”... a considerate man who is aware of his status as a celebrity but doesn’t take advantage of it, and who is generous but careful with his presence.”

Keanu Reeves is Too Good for This World - https://www.newyorker.com/culture/culture-desk/keanu-reeves-...

For those complaining that the study is bogus or exaggerated, here's a more believable anecdote with a similar implication, recounted by the Freakonomics authors:

What the Bagel Man Saw (2004): https://news.ycombinator.com/item?id=28327228

I suspect the cookie thing might be about who's in decision-making mode.

Pertinent book recommendations:

Success and Luck: Good Fortune and the Myth of Meritocracy, Robert H. Frank [https://www.jstor.org/stable/j.ctvc77k7k] Great read, entertaining and insightful, many anecdotes, and also simulation results.

The Tyranny of Merit: What’s Become of the Common Good?, Michael Sandel (of Harvard ethics YouTube superstar fame) [https://www.harvardmagazine.com/2020/09/montage-michael-sand...] Still on my agenda. Podcast with the author and Sam Harris: [https://samharris.org/podcasts/221-success-failure-common-go...]

So true. But knowing this - whether you are the one with the extra cookie or not - doesn't matter.

I only have two pieces of advice: 1. If you're the one with the extra cookie, just eat it, everybody expects you to do that anyway; 2. If you're the one without the extra cookie, just grab it, then eat it, see 1.

Your advice is to be selfish and not care what anyone thinks of you?

> Your advice is to be selfish and not care what anybody thinks about you?

Well if one wants to be cynical that's how big fortunes and generally big advantages to self happen.

That is behaving selfishly and somehow manage to avoid repercussions.

Just two examples from stuff that is in the news these days:

1) The grift happened during the War on terror (especially 2001-2010) which somehow was never investigated

2) Members of the rock band "The Rolling Stones" . They cumulatively slept with thousands of married women and women in relationships. They all somehow managed to avoid the husbands and boyfriends looking for their wives. Most of them had homicidal intent against Mick Jagger and their unfaithful wife.

No, on the contrary. It's not very clear from my post, but I warn people not to expect any gratitude when they selflessly give their cookies away. And if you do share your cookies (which is a good thing!), be careful who you share it with. That's all.

Having been in the position of selflessly giving my cookies away with no expectation of reward - the reward was amazing, and worth way more than the cookies.

Perfect! I live for these moments.

Besides, the so called 'leaders' in this experiment were pretty selfish. Making all kinds of gross noises while eating their cookie. Did they care what somebody else thought of them?

> Besides, the so called 'leaders' in this experiment were pretty selfish. Making all kinds of gross noises while eating their cookie. Did they care what somebody else thought of them?

That bit made me suspect the study - that the "leader" ate the cookie is believable, that they did so in a semi-vindictive manner is not believable.

I want to see the actual study.

> I want to see the actual study.

https://www.gsb.stanford.edu/faculty-research/working-papers...

That's not the paper cited in the study. It is also not a study, but a meta-study (review of existing literature on the subject).

In short, it doesn't address the concerns I list in my post.

I find all aspect difficult to believe. I call bs.

A bit of artistic license i suppose ... its supposed evoke a feeling of disgust against the leader"

Reply on news.ycombinator.com