« Could Anything Be Right? | Main | Existential Angst Factory »

July 18, 2008


"at the top of each walk of life we expect to find a disproportionate fraction not only of high ability folks, but also of high ambition and low qualm folks."

Including science, medicine, & engineering, no? This is why practices like data audits by the FDA may be important, even if rates of fraud are low.

Cycling fans might be interested in the discussion of Ricardo Ricco's ejection from the TdF at http://www.sportsscientists.com/; policing performance enhancing drugs is, of course, the same problem discussed here.

Some guesses:

i) Murder Is Wrong! (Efficient moral conditioning makes successful people discard this option to begin with. It might be okay to cut corners here and there, but murder is still Out Of The Question. It's not 'just business')

ii) Arms race explanation: If I start killing my competitors, they might start doing the same thing.

iii) A generally unfavourable risk/reward ratio. a) You give up a lot of control when you're not the one doing the killing. b) When you're, say, a rich and successful manager, you might stand to gain more from a murder than Average Joe. But you also have a lot more to lose. c) Top bosses (for instance) are not irreplaceable.

It seems that murdering business rivals happens often enough that it's not extraordinarily rare:


As for very large businesses, I think you bring up a good question. I'd like to see more study in the economics of collective restraint. I put this question in the same category of "Why do almost all white people refrain from checking "black" on college admissions applications?" I think the answer probably comes from the same place as how behavioral economists can get near perfect restraint from cheating if they have experimental subjects think about an imaginary honor code first. How that plays out in real world phenomena like corporate behavior would benefit from more careful study, IMO.

They fear getting caught?

They fear the associates they would have as a result?

Could be an agency "problem." Most gains from the murder of an outside rival accrue to the shareholder, and most of the risks go to the manager (including possible retaliation). That should suggest that fewer murders than expected will occur.

The opposite should be true for the murder of internal rivals.


"why don't more corporations have their competitors' leaders knocked off?"

Because to succeed I need to co-operate with my competitors as much as I compete with them - in terms of conferences, trade shows, lobbying political leaders to gain favorable regulation for the industry overall, forming trade associations to do this, developing joint standards for technology, etc.

Furthermore, since CEOs and like board members tend to be of the same class, they often knew each other at business school previously, live in the same or similiar neighborhoods, have the same friends, meet each other on vacations, become friends at conferences, belong to the same civic groups, etc. It's harder to kill a member of your country club, presumably, especially if you both went to Wharton together and your wives are best friends, than a stranger.

Naive answer: maybe it really is that hard to pull off a hired hit.

First, the police will look for those with motive. The business rivals pop up. So they audit that business's records for suspicious payments, and the hitman comes up. But given my experience with getting law enforcement to solve crimes, it may be in error to attribute such diligence.

@Robert: Good point, but what about concentrated-ownership firms, like partnerships and proprietorships?

@Hopefully: Whites don't check off "black" because someone will notice that they're white, and soon it will be all over the media.

I think frelkins' in right. It's the same reason wars were fought rather than kings assassinated: at the end of the day, the leaders all belong to the same class and hang out together. CEOs and board members probably see one another more often socially than they do people within their own company. The main concern when policing corporations is stopping cooperation and maintaining competition.

I would say that there are simply avenues with better cost-benefit ratios for criminal businessmen.

1. Hiring an assassin exposes the employer to blackmail. (Not to mention that the assassin has every incentive to snitch on the employer for a reduced sentence if caught red-handed.)
2. Embezzlement, price-fixing, and fraud can more directly and reliably enrich a crooked CEO, are much less likely to provoke a determined investigation (where murder will almost certainly do so), and carry much less severe average criminal penalties.
3. CEOs of large companies tend to have extensive security measures, increasing the likelihood of failure.
4. A murdered CEO will be replaced, and may not have an enormous effect on corporate performance unless he or she has an extraordinary role at the firm, e.g. Steve Jobs or Warren Buffett.
5. Legal substitutes exist, such as hiring investigators and leaking embarrassing information, to cause politicians or business leaders to lose their jobs.
6. Wealth has diminishing marginal returns, and prison or execution would represent a very sharp decline in well-being for crooked elites, so the gains from murder would have to be improbably high to give the scheme a risk-adjusted positive expected return.

Perhaps there are lots of corporate assassins, but they kill their victims in ways that make it look like their victims were not murdered.

Robert has a good explanation with the distribution of costs and benefits. I am also taken by a combined arms race/socialization explanation. Kings and presidents agree that assassination is out of the question; death is for those little people, not those at the top who meet cordially and resolve disputes with those deaths. Every other piece on the board is expendable, but the king is always checkmated, never captured.

I think Carl and others have given a full and complete answer, but just to speculate here's a minor point.
"People initially model others as similar to themselves. Less secretive people can acquire counter-evidence to this initial model, but to more secretive people the evidence is just more cover. As a result, highly secretive people, like, say, potential hirers of assassins, model people, especially small classes of people in similar situations, as VERY similar to themselves. In a world where other corporate elites *were* prone to hiring assassins it would be safer and smarter to stay low profile and not enter a social role that would attract assassins, so that is what such people do.

"If somebody kills someone, that's murder, you go to prison. If you kill 10 people, they send you to Texas, they hit you with a brick, that's what they do. 20 people, you go to a hospital, they look through a small window at you forever. And over that, we can't deal with it, you know? Somebody kills 100,000 people, we're almost going, 'Well done! You killed 100,000 people? You must get up very early in the moring.'"
-Eddie Izzard

I think he's hit on something. Our internal infrastructure is such that we can deal with small-time murderers better. People in charge of immediately dealing with murderers (ie police officers) are less accountable, as any reader of Radley Balko can show you. People at the top of the government are much less accountable. People at the top in countries that don't have a good legal infrastructure at all are much, much less accountable. If corporate rival murders do indeed happen less often than political assassinations, it might be evidence that corporations don't have as much power as many claim--even small time police get away with worse.

Perhaps the reason there are no (or not many?) corporate hit men is just a standard repeated prisoner's dilemma story? (i.e., count hit man hiring as a defection, discount rates sufficiently low that there's an equilibrium of no hit men?)

5. Legal substitutes exist, such as hiring investigators and leaking embarrassing information, to cause politicians or business leaders to lose their jobs.

Does the story about Larry Flynt at the Clinton impeachment allow us to calibrate this claim? I'm unclear on exactly what came out, but I'm surprised by how little info he seemed to get. Maybe he got a lot of leads, but only published the best (the Speaker).

I think that's a naive answer that reflects "honor code" socialization more than accurate appraisal of reality. I could explain the different contradictions underlying your answer, but I think you're already aware of them, you're just walling them off in your mind.

More generally, I think people rush in with "just-so" explanations whenever an open question like this is proffered. Another example is Tyler's recent question on why so few people choose cryonics. It's comical how few people in this commenting cohort will respond with "Hmm, that's a good question. We should engage in more empirical study." I think it's the rush to fill a status vacuum that causes people to rush in with explanations (I'm as guilty as anyone, although perhaps not in lack of meta-transparency, look how I rush in here with my status vaccuum explanation).

Anyways, I've been thinking for awhile now that a more general economics of general, class, and collective restraint would be very useful.

that's not how it's done. they don't hire assassins to kill the leaders, they hire assassins to knock off people holding power whose interests conflict with the business leaders' interests. duh.



Here's one fellow who was caught lying about his race (and grades, and test scores, and much else) and suffered for it.

On the other hand, apparently Ian Ayres, a Yale Law School professor and author of the book 'Supercrunchers,' lied on an admissions application, claiming to be black, in order to boost his chances. His career has certainly been advanced by that choice.

Given how little it seems to cost to have someone killed, why don't more corporations have their competitors' leaders knocked off?

Expanding on HA's point, a more useful question than "why...?" might be "how could we find out why...?" Ideally, this would force people to give not only their just-so stories (more charitably, hypotheses) but also empirical tests.

"general, class, and collective restraint"

I meant "individual", not "general".

@HA: I think that's a naive answer that reflects "honor code" socialization more than accurate appraisal of reality.

No, being the first person to so brazenly claim to be black, and being seen in person as white, would most certainly get me a lot of attention. (That may, however, be due to my well-documented bad luck though.)

I could explain the different contradictions underlying your answer, but I think you're already aware of them, you're just walling them off in your mind.

No, you couldn't, no I'm not, and that's a pretty cop-out reply. Mods, warn HA about his tone.

It's the same reason wars were fought rather than kings assassinated

It is not always as easy to assasinate someone as you might like. The US tried several times against Saddam during the invasion, for example, and missed every time.

It is easier to kill off the followers simply because there are so many more of them.

The premise that less scrupulous people get to the top disproportionately seems to make sense, but I wonder also if it's mistaken. To get to the top you might have to have a great deal of social intelligence, which might include empathy and genuine sympathy (make friends, take care of friends, avoid making enemies, and forgive or help enemies when you've beaten them, to turn them into friends, or prevent bitter revenge-fantasies from growing in their minds), and you certainly will have competitors for positions seeking to make you look bad, so the best defense against that could possibly be a clean record.

I think of an alpha male in a tribal environment. Can he really afford to be a really massive jerk? Maybe, but everyone's got to sleep some time ;)

I remember the Fourth Checkraise linking to a post on this once. They made the point that business murders tend to take place at a very low-level. Above that there would be too much to lose and likely a more thorough investigation. I don't take seriously the idea that it's common but we just don't know because the assassins are so skillful. A few would slip up and we'd know that they try to cover up their crimes by making it look like something other than foul play. I also think that the death-rate of corporate executives, despite their age, is much lower than for drug dealers.

There are many good explanations already, but here's one that's been missed:

The murderous, risk-taking people aren't going to wait until they're a CEO, slowly working their way up the ladder. People always see their current situation as high stakes, and act consistently. The real crazies weed themselves out by being consistently crazy.

As a result, all CEOs are risk averse. The risk-taking that we admire in CEOs is really only risk-taking in contrast to extremely risk-averse CEOs. All they really risk is money and reputation.

Perhaps individual leaders are relatively replaceable in a corporate setting. Once a good leader has stepped up and instituted valuable changes, these changes are usually enshrined as policy. They are not merely held in place by the will of a strong leader, like is ofter the case with leaders (especially in dictatorships, by definition) or crime lords. My company recently got a new CEO. Time will tell if he's any good, but the changes he's made have already been written into policy. If it turns out that these were amazing decisions and our competitors decide to kill him, they will already have been enacted -- and someone else will merely stand up and continue his work.

This may also be the same factor at work with relatively representative states. Four of our presidents have been assassinated, but to the extent that any of them were attempts to change policies (perhaps the attempts on Lincoln before the end of the Civil War) they did not work. The "martyr" syndrome may be at work as well -- if you kill a dynamic leader pushing a new and popular strategy, those who replace him will may be even more zealous about prosecuting it. Assassinations only work in situations where the person you kill is the necessary link in the event or policy you wish to stop. Optimally, his organization should contain people with a higher aversion to risk, so that when one of them steps in to replace him, they will quietly rein in the things you want stopped, either because they don't want to get killed or because they didn't really support them in the first place.

It's not simply a matter of the cost of a killing. If corporate assassinations became commonplace then you'd have CEOs spending exorbitant quantities of cash to protect themselves from assassination.

I think Carl's #4 is a big reason. Few real benefits are likely to come from a short period of "chaos", and few CEOs are spectacular enough that they are irreplaceable. The actual hit is unlikely to have a positive news effect, as the "newsvertising" of Acme's CEOs "strange death" will probably fall under the "No press is bad press" rule unless it is coupled with an expose of exploitative and unscrupulous practices (e.g. child labor).

There's a novel called "Jennifer Government" that poses this question. One young buck rises to prominence in his corporation by corporate violence, and plans to eliminate the already-marginal government so that he can be more ruthless.

I don't remember if the book makes the point, but reading it, I kept asking myself, "Why do these corporate executives go along with this, when it's clear that it will lead to corporate war, possible death, and less profits for everyone?"

Some companies do engage in this kind of behavior. We call them mafias. In some countries, like Russia, the political and judicial climate may put someone who doesn't engage in violence at a competitive disadvantage. In some countries, like Japan, the mafia has (according to Mike Vassar) an accepted, circumscribed, and in many ways useful role in society.

So, we are looking at a stable system with multiple possible stable states. You can get into a somewhat stable state, like Russia is, in which some corporations do use violence. The question could usefully be rephrased as: What's the relevant difference between countries that accounts for whether corporations use violence? I suspect it may be historic. In a country that has an existing set of large corporations that doesn't use violence, those corporations will use their power to protect the judicial system that protects them. If the existing corporations use violence, they'll use their power to weaken the judicial system.

It's also worth noting that, in the US before World War 2, corporations sometimes used violence against workers. It's probably harder to prevent corporate violence against workers, than corporate violence against other corporations.

Ideally, we would like to map out the entire state space, and figure out what the easiest path is from various existing states to various desired states.

BTW, Carl's reasons are all good ones.

GavinBrown wrote:

The murderous, risk-taking people aren't going to wait until they're a CEO, slowly working their way up the ladder. People always see their current situation as high stakes, and act consistently. The real crazies weed themselves out by being consistently crazy.

Interesting. What do these people have in common?

  • Alexander the Great
  • Genghis Khan
  • Napoleon Bonaparte
  • Josef Stalin
  • Adolf Hitler
  • Fidel Castro
  • Kim Il-sung

Answer: They all rose to rule their nations, or conquered other nations, while only in their 30s or 40s.

So, these ones didn't weed themselves out. But it does support the idea that power-hungry psychopaths are risk-takers who will gain power either quickly or not at all.

I have no idea how much my experiences are indicative of anything, but some of the most moral and responsible people I know are CEOs, presidents or executives. They'd never murder a rival. If they were that sort of person, I'd think they'd have trouble attracting and leading employees. Who wants to work for a murderer? Mobsters presumably only put it with it because of how much they can profit (in terms of money and power). Most people are not that desperate for either that they'd risk being involved in violent, organized crime.

The book "The authoritarians" (theauthoritarians.com) demystifies this for you, guys. The personality type of the highly-successful (social dominator) is more strongly correlated to sociopathy than the personality type of the regular folk. If they can get away with assassinations, they will have no qualms -- because, for them, it really is a matter of not getting caught instead of not feeling guilty. The last psychiatrist (thelastpsychiatrist.com) has more information on that too.

Quick fact check on Phil Goetz's list: Alexander the Great assumed his father's throne and took control of his army in 336 BC at the age of 20. He did most of his conquering in the following decade -- by 326 BC he was fighting battles in India. He died at the age of 33.

(This is not a criticism of Phil's main point.)

David Friedman is skeptical about The Authoritarians.

I'd guess it's because while a head of government is at the top of the only pyramid in town, and a head of organized crime is at the top of the only pyramid he can climb, CEOs are mobile. It's not unusual for a "career CEO" to head up several different companies in his lifetime, and barring legal woes (ala Ken Lay) they usually remain financially secure even in the event of their company's downfall, walking away with notoriously large retirement deals or severance packages. Nowadays, a CEO has no reason to assassinate another CEO because he rarely has a personal stake in the success of his company, unlike a politician who protects his party or a mobster who protects the family business. If he leaves well enough alone, someday that rival company may headhunt him on the recommendation of the retiring CEO he might have, well, "retired".

Shareholders, on the other hand, may have a genuine interest in killing off the competition, but they're less likely to fit the character profile.

Well, most readers and contributors here buy into the argument that those more highly paid people in the private sector are just ever so much more competent than their boobish counterparts in the private sector, so that means that they are probably much better at doing it and getting away with it. So, James Miller is probably right; they are better at making it look like there was no murder, which is indeed the best way to get away with murder.

I doubt that killing the CEO of any average multi-billion dollar company would have a sufficiently pronounced negative effect on that companies performance to make such a move attractive for competitors, given the risks involved.
I'd bet that most business related assassinations are targeted at rivals within a given company.

nobody to fix prices with!

My father used to tell me: "Son, any fool can commit a murder, but it takes an artist to commit a natural death."

If you put me in a room and told me that on command I would turn a key and kill 250,000 people, none of whom I have ever met or talked to and most of who I would probably like, I would more than likely not have a problem turning the key.

If you told me to take a gun and walk into the next room and kill someone I knew I would probably kill you.

You figure it out.

Accountant seems to be headed in the right direction. Milgram showed pretty convincingly that a tiny bit of distance and a tiny bit of coercion would breing reasonable people to do some fairly horrific acts. What would probably seal the deal? A blind vote.

1. Are there business domains in which killing your competitor is acceptable? If so, then we compare such domains with those in which it is unacceptable. E.g. Drug Cartels vs. Information Technology.

2. There is a military strategy that says it's better to wound the other guy than kill him. Wounded soldiers require many more of the enemy's resources than a dead soldier. So it's better to force a competitor to "leave for personal reasons" than kill him outright: golden parachutes, etc. make a larger impact on your competitor than just forcing them to promote the successor.

3. Do unto others... killing your competitor would require a huge investment in personal security for yourself. (See #1.)

4. With sufficient resources, murder can be made to look like natural causes. So it happens more often than we know.

Is your basic premise actually right? Is the murder of businessmen less common than the murder of criminals and politicians (in countries where murder is common). A quick Google suggests that the number of businessmen assassinated in Russia (for example) is huge.

There may be a reporting bias as well - are Western media more interested in political murders in far away places than business murders in far away places?

A response to the comment by "hopefully anonymous"

> "Why do almost all white people refrain from checking "black" on college admissions applications?"

I don't know, but here in the UK, a reasonable number of (white) criminals describe themselves as "black."

This is because arresting officers have an obligation to allow the person arrested to make their own determination of ethnic origin on the arrest report.

I'm told (anecdotally, by contacts in the police, rather than with statistical rigor) that repeat criminals have learnt that describing themselves as "black" serves to confuse subsequent juries about the police witness statements, as in "Officer, you described my client as a Black man, about 25" - how can this possibly be the man we now see before us?"

I was thinking the same thing during John Chambers (CEO of Cisco) recent press conference in Second Life. I mean, there was the perfect opportunity for somebody to assassinate a competitor without even breaking the law. It would have really disrupted their press conference. You could leave a paper trail and gradually reveal more clues about who did it in a way that would lead to really cool publicity for your organization...

The best part of running a company is playing the game. Once you hit a certain net worth level it becomes less about the money and more about "winning". So I think you'll see the same breadth of behaviors among business leaders as you do in professional atheletes. "how far" someone will go to win is inversely related with their "need to win" (self-esteem). In this regard I think that business leaders and atheletes are somewhat self-selecting. These are high achievers by nature and are probably less likely to stoop to base behaviors (Tonya Harding/The Mafia being such examples in the sports/business world).

So, the satisfaction in running a business is in winning but as we all know there is little glory in winning when one cheats. I have to think that most business leaders will have realized this by the time they are running billion dollar companies. That said, I think that most individuals are willing to bend rules if they feel that the competition is already doing so. I think this is a slightly different psychology than the "everybody is doing it" justification that we hear about so often. I think it's more of a rationalization. "I have to break the rules otherwise _insert rationalization_".

Ever consider that maybe it happens all the time, but they use difficult to detect measures like drug or electrically induced "heart attacks."

One of the most common jobs written off as evil. Investing

Huh, funny you should mention that. How do you think I make my $.

The comments to this entry are closed.

Less Wrong (sister site)

May 2009

Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30