Thursday assorted links

1. A modeled argument, by Daniel Susskind at Oxford, that technological unemployment will be worse than you think.

2. Oldie but goodie: Iceland’s natural experiment in supply-side economics (pdf).

3. “Escaped livestock are relatively common in New York City, which is home to dozens of slaughterhouses.”  Location economics are usually a bit stranger than you think, link here.

4. Ted Lowi has passed away.

5. Will a revised Trump travel ban hold up?  And Cass Sunstein is recovering from his car accident.  But thinks there should be fewer of them.


I'm sorry Cass Sunstein was struck by a car in a snowstorm while he was a pedestrian, and I wish him a speedy recovery. Had that happened to me, my response would not have been to write an article advocating additional government programs to try to reduce traffic accidents, but maybe that's just me. Hammer, nail may be at work here.

Can you spot why 5a and 5b are linked then?

If I was struck by a car, I would advocate for the same thing I would advocate for if I'm not struck by a car...the soonest possible implementation of computer-driven vehicles on roads.

There are commercials that say, "Friends don't let friends drive drunk"...but there will never need to be a commercial about not letting computers drive drunk. Or computers driving while texting. Or any of the other hundreds of dumb things humans do that cause accidents.

AI cars can't drive in the rain let alone in the snow.

When AI cars work, people will use them. They don't work yet.

"AI cars can’t drive in the rain..."

Apparently they can. Or do you think this video is faked?

"...let alone in the snow."

What you mean is that the can't drive as well as humans in the snow. Not that they can't drive in the snow. They're learning fast, and it won't be long (probably less than 10 years) that they'll be better than most humans driving in snow.

"When AI cars work, people will use them. They don’t work yet."

No, people won't use them if it's not legal to use them. That's why Cass Sunstein should have come out forcefully for policies that speed up the use of autonomous vehicles. It's the quickest way to significantly reduce fatalities.

Let's say that it isn't a matter of learning, but that cheap sensors could be embedded in roadways and cars could also communicate with each other, so they could be alerted to conditions, to obstacles ahead, to drivers around them braking or veering and even to traffic conditions on alternate routes. (All this is easily doable, currently).
Then the question is: when we talk about 'infrastructure spending', what does that mean? Yesterday's infrastructure or tomorrows? Same questions, by the way, with power generation, news delivery, purchasing of any kind, work environments...
The real issue is that we have a government that is fighting past wars and has no clue about a better future.

"Then the question is: when we talk about ‘infrastructure spending’, what does that mean? Yesterday’s infrastructure or tomorrows? Same questions, by the way, with power generation, news delivery, purchasing of any kind, work environments…
The real issue is that we have a government that is fighting past wars and has no clue about a better future."

Yes, that's a key problem with Donald Trump. His whole shtick is to return the U.S. to some mythical golden past. He's going to get the federal government to build bridges and roads...not figure out how to double or triple the capacity of current roads with autonomous vehicles. He's tells coal miners, "you're going to be working your asses off" without thinking whether new developments in photovoltaics (e.g., perovskites) would be a better way to generate electricity than new coal-fired power plants. He wants to build a wall on the Mexican border, without thinking about whether a fleet of drones might do a much more cost-effective job.

Too many people walk at night without any type of reflective clothing oblivious that most drivers' night vision is quite poor. I'm glad Prof. Sunstein is OK but he should remember that glo in the dark vests and sashes are really cheap and will aid in your being seen.

+1. Also remember that pedestrians should walk facing traffic, bicyclists with. Facing traffic you are never blindsided, have that last chance to dive out of the road.

"I’m glad Prof. Sunstein is OK but he should remember that glo in the dark vests and sashes are really cheap and will aid in your being seen."

Yes, I have flashing LED belts that can be worn over jackets and are visible from 100+ yards. I've been passed by bicycle riders while jogging who've commented on how great the belts are for being visible after dark.

While the description of the accident is not detailed, it does appear that lack of visibility was a cause (there may have been other causes as well). In which case there was a cheap, easy, decentralized solution. (Not that I disagree with government run, centralized, expensive safety programs in all instances.) Wear reflective clothing and/or carry a light. I walk quite a bit and always have a light handy; I am continually surprised how many people do not.

I wish the Professor a speedy recovery.

I have a magic crocodile tooth necklace that protects me from being hit by cars. Putting it on before I leave the house fills me with confidence that I am in control of whether or not bad things happen to me.

If it glows in the dark with magic LED lights on the crocodile teeth, your confidence is justified.

I think equally important is if you are walking around at night in poor visibility conditions, don't assume that drivers can see you. Be ultra-cautious crossing roads, don't be distracted, pay attention to where you are going and look out for cars that might not see you. It may be the drivers fault if he hits you, but you're still the on who is laid up in a hospital bed, or worse.

Additional data point. In a recent conference call, Travellers indicated increasing losses in Bodily Injury coverage driven by distracted driving / texting / smartphone while driving.

I'd be willing to bet that he was walking on a dark road wearing a dark coat, dark pants and dark shoes, and quite likely a dark hoodie or cap. In other words, he was all but invisible to the driver.

Maybe we should require that guys named Cass Sunstein wear reflective clothing when they're walking on dark highways in snownstorms?

5a. What would actually be best bang for the buck in stopping terrorism? I would think that it would be more staff and technology applied to social networks. I would hope this could be done with existing privacy, but I worry that if you do want to spot radicalization you won't just ask for passwords in customs. You might have to open all phones to the NSA/FBI.

Containment. There are over 50 majority Muslim countries. If any Muslim is unhappy in his current land, he can find another that suits his tastes. There's no need to be importing a populating of hostile foreigners. Keep the Muslims bottled up in their home lands and Muslim terrorism drops.

And yes, I would be OK with banning Islam in America and deporting all oft the Muslims currently here, including Louie Farrakhan, if St. Kitts will take him.

Wow. Not only was that not an answer to my question, not only was it remarkably immoderate, it was very unconstitutional.

What are your values, sir?

Constitutions can be changed. The current one used to permit people like you to own people like me as pets. The Constitution is not a holy relic. It was not handed down by God. It's words on paper. That's it.

There are days when you come off as pure parody account. As when you suggest that because we used to more overtly discriminate by race, we should now do more (not less) discrimination by religion.

Here's my "two cents" on the best way to forefend terrorism: a concealed-carry pistol and a rifle behind every blade of American grass.

It's not about religion. It's about peace, prosperity, and public safety.

Do you also stink, er, think that the Aztec "religion"/human sacrifice is protected by the Constitution/First Amendment? Try to publicly sacrifice a ram to Wotan (or the great god Gitchy-Goomy) and see what jail you get to inhabit and for how long. .

None of it concerns "establishment of religion" (the precise meaning/words in the First Amendment), it's about life, liberty, and the pursuits of happiness. Many adherents of Islam practice it in a way that is antithetical to America, American liberties and to our way of life.

Plus, "If you think it’s a good idea to bring in refugees who hate your country, your religion and your culture… you might be a moron!" Jeff Foxworthy

If you think refugees have committed acts of terrorism, you are definitely a moron.

Maybe another trillion dollars spent on massive privacy invasions which can be easily repurposed to society wide oppression? To save maybe .. a life or two?

But as for a million or two lives lost to tobacco ... FREEEEEEDOOOOOMMMMMM!!!

Well, yes, if islamic terrorists were blowing themselves up in empty forest preserves harming no one else, that would be quite a different thing, wouldn't it?

Again, I think the reason Tyler puts two things under item 5 is the irony. He expected readers to reject calls to save tens of thousands of lives with the power of the state, while endorsing state power to save a much smaller number.

Similarly, if people who smoked did not in fact smoke, then the point of comparison would not be relevant.

FALSE FLAGS, that is the only thing you need to know about terrorism. The obvious solution to stop terrorism is to stop financing terrorism and Islamic extremism (Hear that CIA/FBI?) But clearly they are too useful to topple enemy governments and to make sure that americans/westerners live in fear of Islam, which is an important mechanism of control. Obviously, what started in 1980's with Bin Laden, with ours and Saudi money didn't stop and its not going to stop any time soon...

There's hope.

At least we can tell them how dumb they are.

Couldn't you just not carry a phone? Mail it ahead of time to your destination if needed

1. The condition of the economy reminds me of outpatient surgery centers. They typically have a useful life of about 15 years, after which they either are reloaded or they collapse. The reason is old age, namely the old age of what had been very productive surgeons. And so it is with the economy. As Cowen et al. search in vain for explanations, the underlying problem is too few people have too much of the income and wealth. If economic Armageddon awaits, how do those with wealth expect to remain wealthy? With stock in Walmart? Some of Cowen's friends are shifting to real estate, large tracts, here, there, and everywhere. But what's the real value of real estate? Indeed, the wealthy in early America came to realize that real estate is as much a liability as an asset. Like those surgery centers, the only solution is to reload. In the case of surgery centers, that means bringing in the young surgeons to replace the old. Believe me, it's an unpleasant task sending the old surgeons out to pasture. In the case of the economy, the solution is just as unpleasant, but necessary. And the solution? Ask Cowen's colleague Peter Boettke. It would be nice if the economy could be reloaded by offering discounts on U-Haul trailers, just as it would be nice to reload surgery centers by giving the old surgeons a couple weeks vacation. But for the old surgeons and the economy, the End is Nigh.

"If economic Armageddon awaits, how do those with wealth expect to remain wealthy? With stock in Walmart?"

Walmart is in danger. ("Grave danger," to borrow from "A Few Good Men.") Probably 90% of its brick-and-mortar stores will be closed in the next 30 years. Will Walmart be able to successfully transition to a completely online retail world? To me, the crystal ball seems cloudy on this question.

They are having a hard time competing for talent from they bay. Their corporate culture is, as one would expect, having a very hard time dealing with the compensation structure.

Their hope is that coding retail websites becomes far more of a commodity in the future and that their enormous logistical network will stay an advantage.

I've had generally good experiences shopping online for Walmart stuff. I probably purchase more through Walmart than through Amazon at present. But I have no idea of the relative profitability of each side of the business. It seems like I buy big packs of toilet paper and paper towels from them, online, for the same price as I would pay in store. My naive assumption would be that having UPS or FedEx bring me a huge box filled with a package or two of paper products isn't great for margins, but I don't know.

I hadn't known, but apparently is enough of a powerhouse that had to reduce their minimum order for free shipping to $35 to keep pace.

Thinking that WalMart is going to disappear is silly, both because physical stores are really important for many things, and that walmart is going into the online world with guns blazing.

Walmart very recently dropped to $35 for free 2-day shipping on certain items (a decent selection but nothing close to Amazon Prime). I've been paying for Amazon Prime since maybe 2009 so I didn't know their minimum order for free shipping had ever risen above $35.

Last year Walmart had tried an Amazon Prime-like subscription service that gave free two-day shipping on select items (I can't remember if there was any order minimum). It was $50 a year and didn't come with anything extra like Prime, so I don't think it was very popular (though I tried it because, as I said, I order a good chunk of Walmart stuff).

Within the past month, they gave up on this subscription idea and issued full refunds to everyone who had signed up for it, going instead with the free 2-day shipping on a $35+ order of a decent selection of items. For example, on Tuesday I bought some motor oil, herbal tea, diapers, printer ink, and laundry detergent, all qualifying for the 2-day shipping.

However, my experience has been that Walmart is not as reliable about 2-day shipping as Amazon. I very, very rarely get a 2-day Amazon order later than the 2nd day, and have been using the service for 7+ years. I have already had a few Walmart items come a day late (or possibly more), and it looks like my order from 2 days ago may not make it today.

So still has some kinks to iron out, for sure. My prediction would be that they will find a comfortable spot in which they are somewhat less reliable than Amazon, but noticeably cheaper for lower-priced goods.

"Thinking that WalMart is going to disappear is silly,..."

I'm not sure who you thought wrote that Walmart (the company itself) is going to disappear, but I know I didn't. The question was whether buying Walmart stock was a good idea. I did write that the brick and mortar stores would be closed/repurposed, but that wouldn't make the company itself to disappear.

"...both because physical stores are really important for many things,..."

I don't agree with that. Not long-term. I think that in 3 decades, 90% of Walmart's physical stores will be closed/repurposed.

"...and that walmart is going into the online world with guns blazing."

Good for them. Online is the future of retail...especially when AI-driven delivery vehicles become common.

Shopping online at Walmart is not bad. The truly awful part that they have not ironed out is when you pick up the item at the store. This has been a truly awful experience at every WM store I've tried.

Walmart Grocery pickup, on the other hand, is a beautiful thing from start to finish.

I have had decent luck with store pickup the few times I have tried it. But the closest store to me is in a relatively rural location and the pickup desk is on its own in the back. So I was the only customer there each time, and I was able to take a shortcut through the oil change/tire shop. Based on my other in-store experiences, though, I could imagine pickup taking a turn for the worse if any of the above weren't true. Which is why I try to buy for home delivery whenever possible.

I am talking about an item in-store, not grocery. I have never tried Grocery Pickup and actually forgot it existed. I should check if my local store offers it.

Assuming that people want to spend 100% of their time holed up at the office, in transit, or in the home bunker, then in 30 years time people will never need to shop.

Besides the ginormous vulnerabilities associated with a system that would make this possible, I think ... people might just want to get out a litlte.

But if going out can be made sufficiently unpleasant, perhaps people can be persuaded to completely atomize themselves, for mazimium ease of complete slavery and milking without even really realizing the completely enslaved situation?

Or ... maybe people will just wnat to get out a bit. And so there will be more brick and mortar than many people guess.

"Besides the ginormous vulnerabilities associated with a system that would make this possible, I think … people might just want to get out a little."

Yes, the Walmart stores will be repurposed into indoor go-kart racing and paint-ball facilities.

People like to shop, but Wal-Mart is not for shopping.

Walmart is for restocking....they should have been working on the last mile solution for some time now....and automatic order refulfillment.

The reference to Walmart stock wasn't a reference only to Walmart stock, but the folly of relying on asset values, financial asset values, for one's wealth if economic Armageddon awaits. How does an economy sustain itself if fewer and fewer capture more and more. It can't. That's why an increasing number of very wealthy people are shifting their wealth to "hard assets" like real estate, real estate located far from the Hell-scape of economic Armageddon. Do any readers know Peter Boettke's solution? That's more interesting than whether Walmart's future will be Sear's past.

"How does an economy sustain itself if fewer and fewer capture more and more."

The many legally extort money from the few, in the form of taxes.

3. I thought all Americans walked around with AR15s, surely a bull is an easy target?

The stockbrokers didn't want to harm their sacred cow.

1. "In a dynamic version of this model, the endogenous accumulation of capital drives labour out the economy at an endogenously determined rate, and absolute wages fall towards zero. In the limit, labour is fully immiserated and 'technological unemployment' follows."

That paper was one depressing read. Made me glad I'm old!

I don't know--is not working a bad thing? Thats what people aspire for their entire lives. The trick is equalizing the returns to capital among everyone.

#1 -- so much theorizing about how recent changes in the effectiveness of AI techniques will affect the labor market. Perhaps we should allow the engineers to simply work on the problem, make the world better, and the react? Or maybe we should use this as yet another reason to push a failed ideology down people's throats?

#2 -- great reading. Wish the experiment had been longer.

#5 -- I really enjoyed regulators taking credit for airbags, crumple zones, safety belts, etc, when they had little to no input into these technologies. Laughable. I suppose history is written by those with the guns!

Wasn't in Charles Murray who showed that most safety regulations come along after the fact. I may be wrong about who studied it, but it was an eye opener.

Engineers are not renowned for thorough consideration of social impacts of their work.

It's not hard to imagine an army of engineers working their way to "better better better", only to turn around and find ourselves enslaved by an AI who's had just some handful of lines tweaked to "maximize mind control powers attributed to individual X" (or the AI itself, as the great fear is), as opposed to "smoothly implement these trillion microimprovements, such as a wee stimulus to hurry up the slackers on their way back from break".

a) luckily we really, really, do not care, one iota what you and your ilk think/want/desire.

B) the idea that some ethicist will perform better analysis as they what AI should and shouldn't do is laughable. And/or that they will be better at predicting the consequence of the decisions made during implemention.

C) also, no one cares, one iota, about what you can imagine.

D) luckily given the trump administration it is unlikely that you and your meddling kind will have any influence.

And that's why engineers are engineers, and not managers or presidents.

I don't what your post refers to or even means.

Related idea: It is not lack of technical solutions, but lack of political will, that is behind most unsolved problems.

I learned that from engineers.

#1 so we should never plan for anything?

Couldn't Trump just direct the State department to not issue visas to anyone in the 7 countries? Does he not have this authority? Could this be challenged in court?

People already with visas wouldn't be barred, but it would stop those numbers from increasing.

I believe people at state are generally expected to act in accordance with the law and its interpretation according to courts, not dictats which are inconsistent with both of the previous.

You'll be happy a million times over that it works this way. Just not this time.

1. Worse than WHO thinks? Certainly not worse than YOU think, Tyler. You've been beating this drum for years now, while the American economy has added 16 million jobs (all private sector) since February of 2010.

Certainly not worse than, say, Kurt Vonnegut, who wrote the dystopic Player Piano in 1952, just to pick one top of mind example.

This time it's different? Maybe. But big thinkers have underestimated the resourcefulness of human beings for hundreds of years running now. That's a heavy-duty sandwich board.

If I remember correctly, you worry (as do I) about middle class wage stagnation. Are you really sure these are unrelated?

Certainly not worse than I think either. Artificial intelligence does not have to pass the Turing Test and does not have to model human intelligence precisely, in order for it to take over most jobs.

Maybe most jobs that exist today, but most jobs that existed 100 years ago no longer exist, yet here we are.

Where are we?

We are in a funny post free market, post invisible hand, world.

Voters rejected the idea that things are going swimmingly, and they rejected all wonky (now called elite) solutions (right and left) to those problems.

And so we get (even up until yesterday's news) walls and trade wars.

If the wonks are right, and automation is a large factor in economic discontent, then obviously these solutions will fall short.

Do you mean you agree, or disagree, with Brian Donohue?

And, weren't walls and trade wars the wonky solutions, around 275 years ago?

I am not sure where Brian places higher weight, on job creation optimism, or stagnant wage pessimism.

And I limit myself to current, ideally tested, wonkiness. The result of natural economic experiments, etc. We should do what we test and find we can.

Show me a trade war, anon. Looks like Trump wants to tweak existing ones. I think he's not as smart as he thinks nor as dumb as many leftists imagine: he wants to use mild -- if loud -- threats to gain marginal advantages in trade deals. So?

I think the biggest surprise relates to people starting to buy washing machines, etc., with their additoinal money instead of ploughing it all into more babies.

The escape from the Malthusian trap. To which we may return ...

5b. Cass is a smart guy. But why no mention of driverless vehicles? This is the clear solution to this issue.

Everyone interested in vehicle related deaths, injuries and property damage should focus on making this happen ASAP. Speeding this up even a few months will save thousands of lives in the US alone, tens of thousands worldwide.


Although a couple of companies have finally realized the size of the prize and are moving full steam towards it. Tesla is the poster boy for this. Putting full 'enhanced autopilot' hardware in every new car and transmitting the data back to home base was quite smart.

Regulators can't do much/anything to make the technology progress faster, and with a trump presidency their leeway to block any successful innovation will, likely, be limited.

At least with the human driver, a biohacked driver is still very likely not to swerve enough to take out a random target.

I fear that autonomous cars will not be truly autonomous, thus leaving numerous systemic-level vulnerabilities, alongside others like ... someone might hack a car and get it to run you over or something ...

All this about Trump's 'travel ban' and how he can make it legal. Yet how about something about, errr, why?

As far as I can tell this ban is about f*cking up a few people's lives for the sake of seeming to do something about a marginal problem.

If you're a 'clash of civilizations' type person this 'ban' does nothing as most Muslims live in other countries (for example, Turkey, India, Egypt, Indonesia) not covered by the ban.

The ban itself referenced 9/11, but there hasn't been a 9/11 attack in the US since, well 9/11. Obama decapitated Al Qaeda and its model of directed attacks by 'agents'. Needless to say if this is your measure perhaps the obvious question is what would have happened if this policy had been in effect before 9/11. Well 9/11 would have happened.

How about the attacks that happened in the US since then? About half of the post-9/11 attacks were non-Islamic terrorism or 'mental illness terrorism' (think school shootings, the theater shooting, Giffords). We have, of course, imaginary attacks that exist in Trump's imagination (Bowling Green). The Islamic attacks that have happened have been dominated by American born, 2nd generation or more citizens. In fact, has a single attack actually come from a person who came from one of the 7 nations? Just one example, has a single terrorist from Iran *ever* conducted an attack in the US? Ever?

What if you instead were talking about a policy to prevent mass shootings from the mentally ill. "These 7 countries have a mental illness rate of 7 per 100,000 people therefore they are banned". What about all the other countries? "Errr well...let's start with these 7 they have some really crazy people look at this guy on Facebook, bipolar, schizo, delusional, he's a real doozy". Why don't you simply prevent people with mental illnesses from buying lots of guns. "errr we'll get to that after we figure out these 7 countries".

How about Europe? Perhaps you could say these countries have refugees that while they haven't attacked the US yet this could be the *next* thing in terrorism?

OK so why are green card holders who even stopped in one of the 7 countries ever targeted? Why not simply concentrate on Syrian and Iraqi refugees? What the hell does Iran have to do with it?

Here's an issue. The courts will probably end up limiting Presidential authority here because we have a dingbat running the show which might (slim chance) come back in the future to hurt the US when and if it ever gets a serious President again.

In the meantime we have a retreat from leadership. Notice how no one makes a positive case for Trump's policy. No one even tries to assert it's sensible. Instead we get defenses based on strawmen ("well we can't let everyone in") or defenses based on stretched, very stretched, analogies to Obama (last serious President). ("Well if Obama once said to keep an eye out for these 7...").

Where is someone, anyone, with an actual positive case? Even this administration seems unable to make one other than to grunt "just trust this guy".

"As far as I can tell this ban is about f*cking up a few people’s lives for the sake of seeming to do something about a marginal problem."

As far as I can tell, you never even attempt to listen to the other side.

"Notice how no one makes a positive case for Trump’s policy. "

It's pretty obvious. Trump wants to reduce the chance of terrorism in the US, specifically, the widespread Islamic terrorism that has become prevalent over the last 20 years. He specifically used the list of terrorist countries that the Obama administration had already compiled. You might argue that his method won't accomplish much, but the argument that the ban " is about f*cking up a few people’s lives" is disingenuous. It indicates that you don't listen to the other side, don't care about their concerns and just want to ignore the results of the election.

"You might argue that his method won’t accomplish much, but ..."

See when someone asks for a positive case for some policy, if you find yourself saying something like this you already lost.

I don't think the Trump ban is an effective policy.

But I didn't go start raving mad when the Obama administration sent the Dear Colleague letter to every college in the country and turned Title IX into a sexist kangaroo court system, either. And that was a far worse policy.

From Cass on traffic fatality numbers:
>The United States should not accept that level of human tragedy.

Well, of course not. Not NOW. Auto fatality numbers have been predictable for decades, but some time around Jan 20 I think we all agreed that THIS IS HORRIFIC AND THE FEDERAL GOVERNMENT IS RESPONSIBLE!!!

Maybe next week, Cass will catch on to the tens of thousands of Americans killed by heroin -- each and every year. That could be safely ignored in the 2009-2016 period, but now, I think the US should not accept that level of human tragedy.

Auto fatality-related stuff has been getting pumped out since not long after WWII.

And come Jan 20 ... now all of a sudden you observe it and see a conspiracy?

"2. Oldie but goodie: Iceland’s natural experiment in supply-side economics (pdf)."

That's a fairly impressive spike. Usually those kind of studies have fairly marginal effects, but going to a zero Tax rate will apparently cause a bump in employment.

"5. Will a revised Trump travel ban hold up?"

Wow, that's some bad journalism.

"And of course the real-world effect of the Trump ban is to exclude immigrants from seven majority-Muslim countries, none of which has sent terrorists to the U.S."

Really? Did the author do even a cursory check?

"Six Iranians, six Sudanese, two Somalis, two Iraqis, and one Yemeni have been convicted of attempting or executing terrorist attacks on U.S. soil "

Yes, good point. The 9/11 hijackers were never convicted of terrorism so no reason to include Saudi Arabia or Egypt. Clearly the ban was well thought out and a sensible policy.

I would support increasing the ban to include Saudi Arabia and Egypt. The Trump administration decided for political reasons to keep the list to the one the Obama administration had previously compiled.

"Yes, good point"

I'm glad you agree.

As I noted before, no positive case for the ban to be made.

For the study period of 1975 to 2015, that's like ... one every few years (this includes attempts, which are reputed to often include what are basically setups).

Across 350 million Americans ... I dunno, I still say go out and buy yourself some anti-slip materials for the bathtub, go to bed feeling safer, and forget about those terrorists who act outside the purview of the state.

What am I missing? Here is the relevant portion of Cass Sunstein's (here's to his speedy recovery) column:

In 1930, . . . the number of deaths per 100 million vehicle miles traveled [was] a whopping 15.12. By 1940, that figure was reduced dramatically, all the way down to 10.89. Over the next 20 years, the death rate was cut in half, reaching 5.06 in 1960.

One of the country's regulatory success stories, the National Highway Traffic Safety Administration, was created a decade later, and by 1980, the rate had fallen to just 3.35. Over the next two decades, it continued to fall every year, ending up at just 1.53 by 2000. Until 2014, there was continuing progress, producing the all-time low of 1.08."

So the death rate fell from 15.12 to 5.06 BEFORE the NHTSA was established, then it continued to go down, and now it has gone up some (with the NHTSA still in place). How does this unambiguously demonstrate a "regulatory success story"?

It's not clear that the decrease before the NHTSA wasn't also regulatory. No doubt building better roads, agreeing on sensible traffic laws and enforcing them, dramatically decreased highway deaths. It does seem like there's a limit to how much you can lower deaths by looking at factors other than the cars themselves.

The main reason it recently ticked up was that miles driven also rose for the first time in years. But at an economic blog you should know that economic trends regularly have noise and one or two observations does not disprove a trend.

Wait till we have at least have two or three observations before making such claims.

By the way, do you still believe in the Peltzman effect that seat belts actually cause more deaths?

Do they still teach that at George Mason?

The rate is per 100 million miles so more miles driven shouldn't cause the rate to tick up.

True it's 3 data points but presumably all the other years follow the same pattern. I doubt 1981 suddenly reverted to 1930 levels of highway death per 100M miles driven.

Congestion, etc.

Also, road quality is deteriorating, right? So 100 million miles drives in 2014 might be reasonably expected to by safer than 100 million miles driven in 2017, assuming higher congestion and lower quality roads.

Is there significantly more congestion and poorer roads in 2017 than 2014?

I think not. But considering more people and more economic activity, in a situation of no notable expansion of transporation arteries ... there's gotta be an effect in that direction. And the observatoin to explain is not that large of a difference.

"So the death rate fell from 15.12 to 5.06 BEFORE the NHTSA was established, then it continued to go down, and now it has gone up some (with the NHTSA still in place). How does this unambiguously demonstrate a 'regulatory success story'?"

Because that's what he unambiguously wants it to demonstrate.

I'm the cuck of the walk!

No, the real Art Deco has an impressively large vocabulary, and a capacity for interjecting facts into arguments dominated by opinion.

Wow. Look how siloed your readership is. They don't even know who Ted Lowi is. Shame.

Congress should pass a law that anyone walking in low visibility conditions within 500 yards of a motorway should be required under penalty of a fine or imprisonment to wear high visibility clothing and anyoneone carrying a minor in said conditions should also be required to also wear a sign stating "baby on board" also subject to criminal penalties. Given that motorways are instrumentalities of interstate commmerce the US Department of Justice should be empowered and properly funded to set up a "task force" to combat the problem identified. In honor of the noted professor who brought this grave problem to light, the law should be known as "Cass's Law."

Did Cass get nudged?

"Airbags and seat belts" seem to be inseparable. How much additional benefit do airbags provide to someone already wearing a seat belt?

Comments for this post are closed