Toxic Origins, Toxic Decisions: Biases in CEO Selection | Hacker News Reader

Toxic Origins, Toxic Decisions: Biases in CEO Selection

papers.ssrn.com

105 pointsmarojejian58 comments

archive: https://archive.is/https://www.ft.com/content/5ffe19c3-20ec-...

Paper: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5270031

OK, I'm very skeptical of the correlational results, and feel a bit guilty posting this. But... on the other hand:

1) This is pretty entertaining! 2) Pollution aside, I do buy into the narrative that selection bias for lucky risk-takers explains a lot of bad leaders:

>Superfund CEOs, by contrast, are more likely to take high-variance internal risks — such as aggressive restructuring — that sometimes generate standout outcomes. When these gambles pay off, they are difficult to distinguish from skill. Firms, observing only results, may systematically promote risk-takers...

> I do buy into the narrative that selection bias for lucky risk-takers explains a lot of bad leaders

It’s not even actual lucky high risk gambles that are successful that are rewarded. A lot of times, it’s just perceived success. Plenty of leaders fail upwards. Reward for incompetence and failure is a promotion that is mostly dictated by who you know in the organization. And perceived success is a huge factor in “who you know”.

Yep. Your startup fails? "Wow you founded a startup! Join our exec team!" You lost a billion dollars? "Wow you were trusted with a billion dollars! Join our exec team!" You presided over the death of a famous and beloved institution (say Radio Shack, Toys R Us, Yahoo, etc)? "Wow you worked at a famous place! Join our exec team!"

At a certain level, risks no longer are about potential upsides and potential downsides, they're about potential upsides and potential massive upsides. Risk taking has no negatives

Reminds me of the military where being present when a disaster caused by upper echelons strikes can lead to "careers due to silence".

Much more likely that what you see as failure by certain criteria was an overall success by the criteria of the people who make decisions about hiring them "upwards".

If running massive vanity projects for years, that get completely scrapped, while you get so far behind your competitors that you have to layoff half the org and then spend hundreds of millions to acquire those competitors is called success by some metrics, then I would say you’re prioritizing the wrong metrics.

If there aren't really any failure states for you on a personal level why even spend time planning?

You are not contradiction the parent commenter, you actually say the same. A key point was/is that they use the wrong criteria.

The parent commenter wrote

> it’s just perceived success

and the abstract from the paper says

> suggesting firms reward observed success without recognizing underlying risk tolerance

>It’s not even actual lucky high risk gambles that are successful that are rewarded. A lot of times, it’s just perceived success.

Or, perhaps relatedly, short-term success which sets up long-term failure - but by then the leader has moved on from the role.

Yep, always entertaining to watch these Marissa Mayer / Ron Johnson types take a step outside their perception bubble.

This is just the Peter principle - individuals will keep getting promoted until they're no longer competent.

GP is arguing that known incompetence is promoted for political reasons. Peter principle is far more idealistically meritocratic.

If the companies crash and burn, but the CEOs succeed more personally (money, sex, fame, whatever floats their boats), then it's a selfishly useful adaptation.

At work I’m watching a skilled, proven VP take a huge gamble that is out of his control even though he doesn’t seem to realize it - his confidence clearly blinds him and causes him to acknowledge the risk but claim everything is 100% according to plan. Even while many senior people are quitting due to this gamble, some of whom were less effective, but some are virtually irreplaceable.

I’ve worked with this guy for a long time and he’s a risk taker but nothing like this.

What seems to have triggered this is his fascination with LLMs. I have a feeling this is happening to a lot of execs in tech right now.

LLMs brought an environment of speculation/gambling about its impact in thr future. Pretty much every exec and investor is blinded by it.

Just keep in mind that lucky risk-takers who can do this on a risk adjusted basis are a good thing. The problem is when they can't risk adjust and burn everything down.

Another problem is when they are extreme narcissists and burn everything down themselves to fuel their own egos

> I do buy into the narrative that selection bias for lucky risk-takers explains a lot of bad leaders

It explains a lot more than bad leaders, it explains almost everything in society. The winners write history and you can't really win unless you take big risks.

> In short, Superfund CEOs display performance with essentially lower mean but higher variance.

Outcomes however are not evenly distributed, large success typically has outsized rewards.

This is why VCs have the investment strategies they do: invest in 10, 7 fail, 2 don't matter, and 1 is everything.

which makes sense in a portfolio, but a company only has one CEO in their basket. But I suppose if CEO elections represent stock holders with many companies, then perhaps it may still be optimized like a portfolio with multiple CEOs.

Correct.

The effect is more mild, not necessarily because the profile is different but because public companies don’t usually have 100x upside.

Only one at a time, but a CEO is still chasing that monumental success that their reputation can ride on later.

Alphaville is usually free to read.

> selection bias for lucky risk-takers explains a lot of bad leaders

For example, it explains Elon Musk entirely.

No, Musk has not just been lucky. He founded three companies which came to be valued over a billion dollars, each in a different industry; this is more than luck.

Which three? Cause Tesla wasn't founded by him, neither was Twitter.

I’d argue that he did found the Tesla as it exists now, but if you want to discount that, you could choose XAI, SolarCity, or OpenAI.

Musk did not found SolarCity; that was Peter and Lyndon Rive.

I don't know what "found the Tesla as it exists now" means. Founding and leading are distinct. Musk has been a key leader at Tesla; he was not a founder of it or any merged or acquired company.

Founding is not the same as incorporating or being the first to work on. For example, the founding fathers of the USA were not the first people to live in the thirteen colonies, and many of them were only involved with some part of the origination of the country.

I have to imagine you know that isn't what anyone else means by "founding a startup" and that the original 13 colonies weren't the United States (that's why they had to "join or die," they viewed themselves as separate and independent states, even in competition with one another). You've invented a new definition to suit your rhetorical purpose. I think you can make your argument without this frankly dishonest tactic. I've seen people make the same point here on HN many times.

I think Musk is a grifter but even I think those picks are unfair to him. Respectively that's a company with no significant accomplishments yet (it's only two years old so I don't think that's even a knock against it), a company he didn't found but purchased under shady circumstances and has had a lot of scandals, and a company he "founded" a lot like he "founded" Tesla.

If you want to showcase a company he unambiguously founded which is unambiguously successful, why wouldn't you pick SpaceX?

Pedantry, this counterfactual idea that tesla or EVs would be as big as it is in America without Musk is just absurd. The company was bought when it had like 3 people and a 250k concept car which was completely redesigned and rebuilt under Musk. The real reason why tesla is important is because it was the first one to execute big on charging networks, no other manufacturer had the balls to even conceive of a nation wide network.

> Pedantry

You took a lot of words to agree that Musk didn't found Tesla. That other stuff would better be argued with someone else who is disputing it, because the person you replied to was talking about founding companies.

I believe the pedantry label was sufficient acknowledgment of fact, while also pointing out that in the context of the larger conversation we are really talking about whether his leadership decisions led to success.

I disagree with both the premise and the conclusion.

As far as I can determine, Musk is the sole founder of only two companies -- SpaceX and The Boring Company. The former is clearly valued at >$1B; the latter is not.

He is also the cofounder, with many other cofounders, of a variety of other companies: Zip2, X.com, OpenAI, Neuralink. OpenAI is clearly valued at >$1B; Musk was one of eleven cofounders. My assumption is that your third company is X.com; Musk was one of four cofounders, and the company then merged with Confinity (also multiple cofounders), then took the name PayPal (which had been a Confinity product). PayPal is clearly worth >$1B today. I would find it misleading to say Musk "founded OpenAI and PayPal" given the above, but up to the reader.

Whether this is "more than luck" -- in particular, whether it's actually due to Musk's good leadership -- is far from proven. OpenAI, for example, had $1B in capital pledged at founding, suggesting it was already valued at over $1B at creation time. And the skill sets required to found a later-successful company versus to lead one are distinct. Musk might well be a great founder but bad leader.

(Of course, the obvious intent of my original post was to be a snarky dig at someone I view to be an atrociously terrible leader whose success has been due to a combination of others succeeding despite his influence and simply going all-in with huge amounts of capital every time. If you're not already inclined to view Musk that way, and you believe he's actually a successful businessman who is brilliant if eccentric, then a joke post on HackerNews won't change your mind.)

Your 'snarky' intent was not obvious to me, and would be in clear violation of the HN Guidelines:

>"Be kind. Don't be snarky. Converse curiously; don't cross-examine. Edit out swipes."

Thing is, snarky often gets the votes. So the unofficial guideline is the opposite of the official one.

I remember reading that people who were hypersensitive would avoid overstimulation.

Meanwhile, people with lower sensitivity needed more stimulation and became more prone to crime and risk taking.

I wonder if superfund sites decrease nerve efficiency leading to less stimulation and therefore risk taking.

EDIT:

A 2015 longitudinal study based on army medical records of Swedish men showed a correlation between low resting heart rate and violence and criminality, with the authors theorising that lower sensitivity to stimulation resulted in increased likelihood of risk-taking and sensation-seeking behaviour – effectively a low sensitivity counterpart to SPS."

https://en.wikipedia.org/wiki/Sensory_processing_sensitivity

I live in Silicon Valley, and one thing I noticed while house hunting is that the entire Google Campus, including the entire early campus sits atop a superfund site due to groundwater contamination from solvent tank leaks from early chip manufacturing sites.

There were even reports from the early 2000s of workers getting nauseous within the offices, and the former YouTube CEO passed away recently prematurely from lung cancer (not saying that's related, but if we're looking at correlational studies...)

That would be a wild RTO-related lawsuit.

Were you a Google employee forced to return to office? You may be entitled to compensation.

https://www.mv-voice.com/news/2013/02/21/toxic-vapors-found-...

Same for several Netscape buildings on Ellis St back in the day. Not sure which companies are in those buildings now but I remember the pulme mitigation pipework being installed in the parking lots. Former Fairchild site.

I remember before google the area near rengstorff/shoreline was a huge hill over a garbage heap with lots of PVC vent tubes sticking out of the ground.

It was a huge waste disposal area.

"By matching the locations of Superfund sites and their pollutant accumulation periods with the birthplaces and birth years of these American-born CEOs, we identify 734 unique Superfund CEOs and 2,267 non-Superfund CEOs"

Certainly a bigger sample than I was expecting.

1 in 4 people live within 3 miles of a superfund site, so it's actually spot on.

We really have tried to destroy ourselves over the years.

There's a Superfund site about 10 miles from my house. Luckily it's close to the headwaters for a major source of our drinking water.

Cancer rates here are really high.

Surely they're not connected.

A system can do many dice throws and is better off for doing high risk ones .For the company its fatal, for the parallel exploration of the many possible scenarios of the slime mold expanding, its informative. Even the "boring"-no risk taken CEOs have their place, they ready the canon for the living canonball by driving a company to desperation.

I am going to develop real estate on a toxic dump. It will be beautiful. Prices will shoot up and I'll get rich. All the neighborhood kids will become CEOs.

Meta comment - [update: oh, I see they changed the link to the paper too. i guess that makes it more defensible. but the original FT story covering the paper did add value, and now it is more lost ]

It's curious someone changed the title of this, and I'm of two minds about it:

Pro change:

- Yes, the original title "Pollution causes CEOs: study"is baity.

- Yes, the new title is the actual title of the underlying study, which is less baity.

In general, this is the kind of thing I love about HN.

Con change:

- The original title is the literal title of the FT article, whch is what the post is on, and it does provide commentary (some snarky) on top of the paper.

- The original title does more elegantly and effectively convey the conclusion of the research, whereas the paper title does not convey this at all. It's more crappy intellectualized science writing.

I think this is one case where the more engaging title is also the more effective one.

I scanned the article. Oddly enough, it is about the toxins in the environment where a CEO was born. From the article's conclusion:

>Superfund CEOs—those born in counties later designated as Superfund sites—excel in internally focused management domains where risk-taking remains adjustable and containable, yet struggle with externally focused policies where risks are immediately exposed to market consequences. [...] Superfund sites represent some of the most hazardous contaminated areas in the U.S., and our sample primarily includes executives born before industrial chemicals were widely recognized as developmental toxicants. [...] This historical setting provides a unique opportunity to examine how early-life conditions interact with career selection mechanisms that systematically filter executives based on risk-taking outcomes.

I initially misread this to say that CEOs born in super fund sites had some sort of genetic change that caused them to be riskier due to environmental toxins.

You're not wrong.

It is about where the CEO was conceived and born. From the article's introduction:

>Our approach addresses these issues by exploiting prenatal exposure to pollution from Superfund sites as an exogenous source of variation in executive risk-taking behavior. [...] we control for a wide range of fixed effects, including firm, year, industry-year, CEO birth year, birth county, and headquarters state, ensuring that comparisons are made among otherwise similar CEOs. [...] we show that selection mechanisms in promotion can amplify behavioral traits shaped by early-life conditions, even when firms are unaware of those traits.

Personally, I think it sounds like hogwash -- "statistically significant" findings that have little bearing on reality.

More likely, those born in heavily industrial areas are more likely to have careers in industry.

I had a hard time understanding. At first I thought it meant babies born near Superfund sites were riskier CEOs.

Then I thought it meant managers who were in charge of divisions that created Superfund sites before being promoted to CEO made riskier decisions as CEO.

And now I'm back to "pollution babies take greater risks". Which to my surprise was a surprisingly reasonable hypothesis.

We don't talk about our mutant superpowers outside of the group chat.

im not able to read it any other way. can you explain what it actually says?

I totally did too, thanks for noticing that mistake and warning us!

Reply on news.ycombinator.com