Wednesday assorted links

1. Should we have laxer conviction standards for tougher punishments?  No, but a good brain twister.

2. “Does this mean inflation isn’t transitive?”  Correct, inflation measures are not transitive.  Larry Temkin should be happy.

3. Musk personally made sure the guy’s order was cancelled.

4. “Ravens do spy on each other, it turns out, and they can infer when other birds are snooping on them.

5. “Between 1993 and 2015, cattle killed 13 people who were out for walks in the United Kingdom. Dozens more walkers received broken bones or other injuries from the animals…Murderous cattle are an understudied phenomenon…”  Link here.

6. Haiti since the earthquake.


3. Yep, the new world of high tech CEOs will look an awful lot like the old world of petty feudal lords.

I'm curious, did feudal lords compete for peasants? I know that they competed for knights.

This is a real question, I imagine you couldn't really prevent your peasants from picking up and going, if they were willing to pass up on their right to work the land.

Nope you had to do with peasants that you got.

It varied, but generally there were free peasants (a few) and serfs that were bonded to the land. The Lord could not sell the serfs or kicked them out.. they belonged to the land. Likewise the serfs could not leave, and in some places like Germany (where there were tons of small lords) were pretty serious about runaway serfs.

"and in some places like Germany ... were pretty serious about runaway serfs."

They were effectively slaves. Technically, they had some specific rights, but there was little recourse. And generally their access to weapons was strictly limited.

I'm reaching way back to my college studies, but I seem to recall that after Spanish conquests of new land from the Moors, they had to offer good terms for settling the land to peasants from farther north. A large amount of German peasantry also settled in Slavic eastern Europe in the medieval period; I suspect they may have been lured there by easy terms.

So at least some of the time I think lords competed for peasants.

6. Haiti since the earthquake.

You know it is going to be bad going. You prepare snarky little comments in your mind, like "What? You mean Bill Clinton *and* Sean Penn didn't solve this one?".

But it is actually worse than that really.

A popular and bawdy showman, he appears in one typical video clip in a night club, dancing for the camera in a red bra and a yellow sarong. At one point, he feigns masturbating a giant phallus, then hoists an imaginary breast and licks it.

OK. I suppose Haiti will never be anything other than a source of cheap labor for Florida and France.

6. For the amount of money we have given Haiti the place should look like Switzerland now

Or at least a couple of Geneva villas.

It's too bad that money was given to "Haiti" and not to Haitians.

The domestic product of Haiti has bounced around a set point of 1% of the Swiss domestic product for a generation now. Official development aid is typically 11% of the Haitian gdp, or 0.11% of the Swiss gdp. I do not think the scale is such to make anyplace look like Switzerland, other than a small city, perhaps.

Hard to believe that Bill Clinton and literally thousands of people unable to learn any useful skills or get a real job in the real economy in the West have not turned Haiti into Switzerland by now. I mean how hard can it be? Surely someone with good intentions and an Ivy League degree in English literature should be able to do it.

Sean Penn? Not so sure. After all, I don't think Haitians need lessons in popping paps.

Miley Cyrus might make a lot of money in a debased modern world. But few electorates are stupid enough to elect her President.

Which is a problem, for it means the policies you have had for a long time now are not of the devising of a somewhat crazy young women who is desperately trying to prove that she is not Hannah Montana and who rode the whirlwind of fame while American voters cared about more important things, like the Kardashians. So what is your excuse?

Well Jim, since it would take about $5 trillion to get Haiti up around the wealth per capita of Switzerland, all I can say is, wow, you've been very generous.

"At one point, he feigns masturbating a giant phallus, then hoists an imaginary breast and licks it.
"What does the Haiti Earthquake have to do with Miley Cyrus?

Extremely hostile bulls chasing city slicker Manhattanites who had come to the country looking for a little bucolic peace and quiet was a pretty standard gag in comedy movies and New Yorker cartoons in the postwar era.

It must have happened to Fatty and Mabel a few times as well.

It is inherently funny like monkeys and people receiving injuries to their buttocks. Now if you have a fat man chasing a monkey into a field occuppied by an ill tempered bull... Now all you need is to have a pretty girl with a small dog try to ineffectually help him...

#5: the problem is people with dogs. The only time I've been chased by wild hogs is when a friend took his dog for a hike. The dog tried to socialize with the hogs. It happens also with other mammals, dogs are not trained to be careful around wild animals.

Dogs aggravate the situation, but aggressive bulls may attack humans unprovoked.

I know of people in Greece who were castrated by bulls charging.

And the other day I tried to corral a domestic pig, about 6 (?) or so months old and probably weighing 100 pounds. It was impossible. The S.O.B. was pure muscle. it was scary. We finally shooed it back to its pen with treats and mild threats, but trying to manhandle even a small hog is very hard to do. I've been injured (my hands bloodied trying to hold onto a rope around their neck, and once it got dangerous as the rope almost wrapped around me) twice trying to do this--they are pure power, with a low center of gravity. Contrary to popular opinion, hogs don't have that much fat on them, they are mostly muscle (at least the ones I've seen here). In fact, I hate hogs so much after these episodes I actually want to go see one get slaughtered. They squeal like pigs, so loud, when they go to the slaughterhouse here in the Philippines (which is found in every block it seems). I am positive--I've never asked--they don't use bullets (certainly they don't use western style stun guns) but instead a sharp knife to the jugular and it takes a while--several minutes I surmise--for the pig to bleed out, all the while it is screaming at the top of its lungs; you hear this all the time in the barangays. I want to watch--even participate with my steely knife--since I hate these beasts. They are so selfish, so rude, so inconsiderate, and such...pigs. They eat their own feces when mixed with food. They're fouler than fowl...well, they're pigs after all!

How Hemingwayesque!

About his parents' farm: "The boar-pig gnawed at the forehead, shoulders and arms of the man who was feeding him. It was a new, huge boar-pig that had been brought in to improve the entire herd of pigs. The fellow was frightened to death and sobbed like a boy. He too was taken to the hospital."-- My Life-- Trotsky, Leon D.

"Murderous cattle"? Given what we do to them, surely it's self-defense?

5. Around two dozen people are killed by cattle in the US each year, less than half of them by bulls. One of the reason cattle are so placid is that they are dangerous enough as it is. Groups of cows are especially dangerous when you are on foot.

Now considering that in North America walking through some farmers field harassing their cattle is pretty much unimaginable. Heck walking close to cattle of BLM is something I have almost never seen. So we manage 20+ deaths a year mostly from agricultural workers...

In Europe, and especially the UK, there are all sorts of ancient rights of way exist, and lots of Europeans, Brits in particular seem a little too calm about approaching cattle.

Also some of that advice, as the article alludes to, is particularly bad.

"Walk boldly through the middle of a herd. "

That really, really bad advice. Cows are large animals that often press on each other even when they aren't moving. It's ridiculously easy to get sandwiched between two large heifers. Then if the herd starts moving, there's a good chance you'll get trampled under foot. In addition to that, don't get near the rear of any large animals or with a couple of feet to the side. They kick. A cow can easily snap your leg in two with a side swipe. The swipe has a lot more reach than ill-experienced people tend to assume.

You shouldn't walk in front of a concentrated herd either. If they're disbursed for grazing it's safe, but if they are still bunched you are better off walking around the sides and rear.

#1 is a perfect example of why Utilitarianism is a disaster, it always gets hung up on details, and those details always ignore context. That it is a particularly bad example of apllied reason is irrelevant.

Or in short no one should discuss this sort of thing without meditating on the career of Shang Yang. It saves a lot of time.

I think it is less about utilitarianism and more about susceptibility to certain brain loops.

My point was they go together, the problem is that Utilitarianism is like central planning it runs into a massive knowledge problem.

Might not a simpler utilitarian say "I am here to act on a jury because justice benefits us all?"

Or "as soon as I muck up justice, I trigger unpredictable consequences, and likely greater loss for society as a whole?"

And if you believe that the inscrutability and incommensurability problems are tractable, you are effectively saying that moral liberty exists only where there is ignorance: liberty is a problem to be solved.

2. Not much has changed with college freshman since I was in college: still spending too much time in the dorm late at night contemplating nonsense. But I digress. Is it true that technology improvements are underestimated given the zero cost of Facebook? No, the cost of Facebook, in loss of privacy, manipulation, and waste of time, is underestimated. What is a Markets and Social Systems Engineering Program? It sounds like something college freshman would come up with in the dorm late at night after too many beers.

1. Landsburg has it backwards: the harsher the penalty, the less likely jurors vote to convict. But Landsburg seems to live in a different universe, or a universe with a different reality. For example, his advice to Sen. Cruz to bundle his opposition to ethanol subsidies with opposition to all subsidies. Sure, it's logical, but it defies reality: Cruz doesn't oppose all subsidies, only the subsidies he opposes. Does Cruz support lax enforcement of tax and environmental laws (yes he does), even though doing so subsidizes firms which benefit (to the detriment of everybody else).

"Does Cruz support lax enforcement of tax and environmental laws (yes he does),..."

I doubt that. I imagine Ted Cruz supports lower taxes and less environmental regulation. That's not the same as lax enforcement.

#1 - "Namely (at least in this very simple model), the harsher the prospective punishment, the laxer you should be about reasonable doubt." My criteria would be the more heinous the crime. Another aspect would be a murder is more visceral and understandable than a massive fraud with losses to many (widows and orphans) investors, e.g., Bernie Madoff - not really a good example due to massive press. Absent the death penalty, Madoff's sentence is as harsh as a murder.

Another aspect: do judges properly explain the standard for "reasonable doubt?" In accounting, US GAAP defines "probable" the threshold for recognizing asset impairment, as "more likely than not." Believe me, there is a lot of wiggle room for judgment in that definition.

The approved jury instruction in federal courts explains:

"(4) The government must prove every element of the crime charged beyond a reasonable doubt. Proof beyond a reasonable doubt does not mean proof beyond all possible doubt. Possible doubts or doubts based purely on speculation are not reasonable doubts. A reasonable doubt is a doubt based on reason and common sense. It may arise from the evidence, the lack of evidence,or the nature of the evidence.

"(5) Proof beyond a reasonable doubt means proof which is so convincing that you would not hesitate to rely and act on it in making the most important decisions in your own lives. If you
are convinced that the government has proved the defendant guilty beyond a reasonable doubt,
say so by returning a guilty verdict. If you are not convinced, say so by returning a not guilty verdict."

It is a qualitative standard that focuses on the issue of the state's burden and evaluation of doubts which are reasonable, not merely speculative. I don't believe quantitative standards like 98% or 80% probability are ever used, they are confusing.

"you would not hesitate to rely and act on it in making the most important decisions in your own lives"

I feel like this standard would vary greatly. I would much prefer a quantitative standard. We should clearly state what we think the appropriate odds of false conviction should be.

I think the short response is that jurors are not very numeric. Perhaps they could be screened by only allowing jurors to serve who can pass a test that examines their understanding of probabilities (if heads come up three times in a row, what is the probability of heads on the fourth toss?)

It is better to let 10 guilty people go free than to convict one innocent person []. Quantitatively, that means that one should be at least 91% certain. That's a lower bound.

On the other hand, some have suggested that, if even 1 out of 10000 Syrian refugees is a terrorist, then we should deny freedom of movement to the other 9999. By that standard, juries should convict if they are 0.01% certain.

Huh that seems nonsensical. Active terrorists can potentially do a lot of damage and almost always cost a very large amount of money.

How much money has the US spent on upgraded security after the shoe bomber? The number is probably in the billions of dollars.

So there is no implication of a 0.01% certainty factor. It's instead a cost benefit analysis. How much are the 9,999 non-terrorists going to benefit everyone else in the country versus how much the 1 terrorist will cost.

If even 1 out of 10,000 Americans in your neighborhood is a domestic terrorist, then all of you should be kept in a concentration camp. Unless we have a way to reliably vet you and your neighbors to guarantee that none of you will "potentially do a lot of damage", then we have to weigh the cost against the benefit of you to the rest of us in this country. The benefit of freedom to 9999 of your neighbors doesn't count in the cost-benefit analysis.

I support more Syrian refugees coming in to the country, but I'm amazed at how horribly bad some people's arguments for it are.

Refugees have no right to enter the US and no due process rights if denied. The ratio could be 1:1,000,000 and it would be entirely fair to the refugees.

Putting someone in jail is not anything like the same moral calculus.

2. Inflation is not transitive, and 30 year bonds are less about inflation expectations than economists think. (Any real bond is in a portfolio, and real expectations are about composite performance. Certainly anyone buying 30 year bonds now is doing so because they need them for a strategy, not a return.)

"I came away about 80% sure that the defendant was guilty and 100% sure that I’d vote to convict him. "

Appalling. Chilling. To convict a person of murder is to completely and utterly destroy his life (and in some states it means sentencing him to state murder). Landsburg would accept a ratio of 1 innocent life destroyed for every 4 true murder convictions!? And if our criminal justice system were weak enough that the best strength of evidence we could ever achieve was 50%, he'd go with that and convict people of murder at a 50% confidence level!? Wow.

Are you implying that 80% certainty is not beyond a reasonable doubt? Or do you think the US standard for criminal conviction is appalling?

20% chance of innocence seems like a reasonable doubt

Yes, absolutely! 80% is WAY too low (and 20% of people locked away in prisons is much too high). Nobody would accept an 80% standard of statistical significance in science -- why on earth would we accept it when imposing life-shattering punishments? I have to say, too, that given what we now know about the unreliability of eye-witness identification, the serious flaws in the way police conduct lineups, instances of coerced confessions, abusive prosecutors withholding exculpatory evidence, phone videos of lyings cops, state crime lab scandals, etc -- my beyond-a-reasonable-doubt levels have gone up a LOT more.

I have served on a Jury and, bizarrely, most of the other jurors seemed to think that the defendant had to prove his innocence. I stuck with "not guilty" because the cost of a mistake was too high - the man goes to prison.

FWIW, he's not applying the correct standard. I don't believe any courts or evidence professors believe numeric probabilities are appropriate to explain burdens of proof. He would have to explain what evidence he finds lacking, or what witness is not credible, etc. If he is merely sitting there with a lot of background information about corrupt lab workers or the imperfection of memory, he is not confronting the evidence before him.

I think the real message of that article is that Steve Landsburg wants to ensure that he will never need to serve on a jury.

He's not the only juror. Maybe he also accounts for this in coming up with his 80% number.

"State murder' = a criminal penalty of antique provenance very seldom applied (and then only to reprobates who've had a personal trial. A nonsense term favored by those who fancy rough trade.

I think the term 'state murder' is a bad description. Nor would I be applying it if I felt it was necessary.

How would you describe execution of an innocent person who ended up being strapped to the gurney via some combination of police/prosecutor/crime-lab malfeasance, defense incompetence, and 'relaxed' standards of reasonable doubt (a la Landsburg)? Is it not murder if the state does it (provided the process has at least the appearance of propriety)?

2. If consumption of complementary internet service captures consumer surplus from the internet, wouldn't price only be a measure of willingness to pay if internet provision were near-perfectly monopolistic?
Because in practice, LTE prices are falling with competition in mobile comms. Ashok only pays $100/month for LTE because he's on a traditional carrier that bundles its phones in the price. He could switch to Google Fi or Republic Wireless at any time and pay $20-$40/month for data.

So does the emergence of cheaper cell data as a result of competition mean that the internet is worth less than it was in 2005?

Hedonic pricing.

#4 . I am not sure this is new. It was discussed in " Mind of the Raven"

#3) "In fact, says Alsop, Musk has decided that Alsop 'can’t own' one of his cars."

What prevents Alsop from finding someone to buy a Tesla on his behalf and reselling the car to him? People buy Teslas right --- they don't just license the use of them, with no sublicense rights, like some software?

Nothing. But Musk just cancelled his pre-order. There's no indication that he's actively seeking to have the guy banned from ever buying a used one.

Behavior around cattle: I grew up on a farm, and mostly remember cattle as fairly docile, but probably the most dangerous episode (outside of one highly aggressive animal that we shipped off) was on a 4th of July when I went back atop a hill/high point to watch the fireworks in a nearby town. The cattle (steers in this case) were mostly calm and just stood around eating or watching in the near dark while I watched the fireworks. The cows knew me, so no real surprise. However, after the fireworks were over I started making my way back through the herd (mostly spread out through the field), and something spooked them. I don't know if it was me or something else, but I suspect it was my presence. One cow gets spooked and it multiplies to others, and in the dark I think it's worse than in daylight. Anyhow, it didn't take long and I was hearing thundering hooves of part of the herd, and it seemed to be coming nearer. At that point I just start sprinting as fast as I could to get to the barn a good distance away. It could've easily turned badly if I got caught up in the stampede though.

#1. I wonder how Japan is faring after their earthquake and tsunami?

Re: item #1 ... I previously developed a Bayesian model of criminal and civil litigation that is relevant to this discussion. (The link to my model is here: In brief, my Bayesian model includes a scenario in which the outcome of a trial is purely random and in which the moving party is either “risk-averse” or “risk-loving” (i.e. in which the prosecutor is 90% confident the defendant is guilty or only 60% confident). The surprising result about my Bayesian model is that in either scenario, the posterior probability the defendant is, in fact, guilty is pretty high.

Comments for this post are closed