Wednesday assorted links

by  Tyler Cowen

1. Are economists smarter than epidemiologists?

2. Is China the big loser in the new NAFTA?

3. Goats prefer happy people.

4. This bride canceled her wedding after guests refused to pay the $1500 attendance fee.

5. Why are puffins vanishing? (NYT)

6. Elephant Debussy (short video, if only humans were so appreciative).

Comments

Jack pq

  Hide Replies

Maybe Pearl is correct, but his blog post is just one unsupported claim after another. He sounds like a preacher not a scientist. He ignores the possibility that econometrics provides other tools to conduct the analysis he believes can only be done using graphs. And he never replies to the crucial criticism that key assumptions simply cannot be made in economics, preventing use of graphs.

JFA

  Hide Replies

Yeah... when he kept referring to "graph-free minds" as if graphs and equations weren't two sides of the same coin, I lost interest.

Dzhaughn

  Hide Replies

I clicked through to look at the context of the Gelman quote, which surprisde me. For me, Pearl proves himself quite capable of dishonesty.

Paraphrasing: Pearl, loosely reading a comment thread, tries to bait Gelman into saying graphical models are the only good way to work with causal questions these days. Gelman says no, it is a good way but it is a high bar to say it is the only way. Pearl now takes that out of context to say Gelman is a Philistine graph-o-phobe, look see, they are everywhere.

Worship Pearl or be cast asunder! It's pretty ironic to see this bullying from a guy playing the Galileo-was-bullied card.

C

  Hide Replies

#4 - my take until proven otherwise is that entire story is a B.S. troll.

Uribe

  Hide Replies

Transparently fake. Don't see how Tyler fell for it.

agoodsubreddit

  Hide Replies

https://www.reddit.com/r/ChoosingBeggars/comments/9a8ppv/update_on_ultimate_wedding_choosing_beggar_from/

Uribe

  Hide Replies

I wasted 10 minutes looking for evidence there and found none.

Dzhaughn

  Hide Replies

What if the point of faking the story if to make $200/person seem reasonable, if not modest?

Not sure if I'm gay dude

  Hide Replies

I was raped too.

JWatts

  Hide Replies

"4. This bride canceled her wedding after guests refused to pay the $1500 attendance fee."

I saw this story yesterday. It's, if true, an amazing Bridezilla tale.

JWatts

  Hide Replies

On the happy side, the groom dodged a bullet.

cliff arroyo

  Hide Replies

Dude done dodged an effin' machine gun battery....
Ain't no way that girl's ever making any man happy.....

BC

  Hide Replies

Suppose not having a "blow out" wedding truly affected the bride emotionally. If mental health is a positive right, then why shouldn't she receive financial assistance to pay for mental health care? And, instead of forcing her to ask her family and friends for help, why not just require insurance companies to include emotional wedding coverage as part of everyone's health insurance? That way, it can be paid for out of insurance companies' profits.

If one disagrees with this chain of thinking, then where is (are) the logical flaw(s):

(1) Emotional health is not like physical health. For a mental health treatment to be viewed like a physical health treatment, the treatment should have a clear physiological basis, for example a drug that affects brain chemistry. Note that under this principle, emotional support animals would not be viewed like medication, for example.

(2) Health care is not a positive right, at least no more so than food, shelter, and clothing. Health care subsidies should be limited to only the truly needy poor, like food stamps and housing assistance, and even then with an expectation that such aid would be temporary and limited.

(3) Coverage mandates only mask who pays and receives subsidies. Subsidies are not "paid for out of profits". Subsidies should come in the form of taxes and benefit checks so that they can be transparently evaluated by the political system.

Ethan Bernard

  Hide Replies

Good afteenoon, utility monster.

EverExtruder

  Hide Replies

#2 I think that remains to be seen, but I hope so. It would also be great to revive TPP and start hitting China with anti-dumping rules in their own "backyard" so to speak.

#3 Most people also prefer happy goats. But not me. I like mine fainting.

#4 Wow. Dude. Run.

#5 Years back I had a chance to try puffin. Delicious. I hope they don't because I would like the chance to try them again.

rayward

  Hide Replies

2. Balding lists no benefit to the U.S. None, other than to punish China. And Canada. Balding's big beneficiary? Mexico. Will the Wall stop Mexicans from coming to the U.S., or Americans from going to Mexico. Hey, amigo. Of course, Trump's transactional approach to trade is the divisive approach to trade. The TPP would have been a way to punish China (by diverting trade to other Asian countries), but Trump rejected it. Now, I understand that Trump may restart TPP. Cowen's latest at Bloomberg attributes (blames?) capital mobility for flat wages in the U.S. What if Trump's trade war blunts capital mobility? Who wins? Mexico? Hey, amigo.

JWatts

  Hide Replies

"2. Balding lists no benefit to the U.S. "

rayward, does this have to be spelled out for you?

If Mexican wages inflate, there's less illegal immigration from Mexico into the US. Which also places less downward pressure on low skilled wages in the US. Furthermore, a wealthier Mexico, is probably going to be both more environmentally friendly and less violent. Both of which are prone to spilling over into the US.

rayward, why do you hate Mexicans? Try not to be so racist. /jk

rayward

  Hide Replies

So Trump's policy is to build Mexico's economy so Mexicans won't come to the U.S.? Make Mexico Great Again. Will that be Trump's campaign slogan this year? MMGA.

EverExtruder

  Hide Replies

Tongue-in-cheek aside, that's not entirely a bad idea and there are aspects of it that could be considered as practical solutions to immigration. Not just Trump, but Mexicans should make Mexico great again especially in light of this https://www.economist.com/the-economist-explains/2018/05/09/why-mexicos-murder-rate-is-soaring.

People seem to forget just how truly desperate people (rich and poor) in other countries are to get to the US because economic opportunities simply don't exist at home. That is not necessarily the fault of the US however, and more arguments should be made to people in other countries that the better solution is to fix economic problems at home rather than packing your bags for the US or EU.

Hazel Meade

  Hide Replies

If it's that easy to change the country you live in, why isn't America a libertarian, or socialist, or whatever paradise yet? If your desired government is something only 5% of the population agrees with you on, what's your answer? Sucks to be you?

JWatts

  Hide Replies

"If your desired government is something only 5% of the population agrees with you on, what's your answer? Sucks to be you?"

Work with a bigger group to join a coalition. If you refuse to be in a coalition, then yes, it sucks to be you.

Hazel Meade

  Hide Replies

So Republicans living in California are morally obligated to stay there and try to change California's government, instead of moving to Texas, right? No more voting with your feet - stay put and spend the rest of your life arguing about politics instead of doing what you want with your life. That doesn't sound very libertarian to me.

JWatts

  Hide Replies

Sure they can. But they can't move to Texas and claim homestead public land. They have to abide by the laws of Texas to live in Texas.

Hazel Meade

  Hide Replies

Since when are immigrants homesteading public land?

But I guess you agree that people aren't obligated to remain in their home countries and "fix economic problems at home rather than packing your bags", so I'll consider that conceding the point.

Erik

  Hide Replies

It's my understanding that Mexicans are not the ones that are immigrating to the US illegally through Mexico

I.e. South Americans transit Mexico to enter the US, so improving circumstances in Mexico isn't as relevant as it may seem

JWatts

  Hide Replies

There is still sizable illegal immigration from Mexico. And even for the others, I imagine if the Mexican economy improves at least a percentage of illegals transiting through Mexico will decide to look for jobs there.

But that being said, all of this is rather a weak tea argument in favor of the new deal.

Excursive

  Hide Replies

Well, a wall across Mexico's southern border would be much cheaper.

Ari T

  Hide Replies

#1 Have you guys heard of the worm wars? It was fought between epidemiologists and economists in Effective Altruism research.

https://blogs.worldbank.org/impactevaluations/worm-wars-anthology

I wonder if this is related. Anyway that has to be one the most amusingly named debates.

clockwork_prior

  Hide Replies

'Why are puffins vanishing?'

Porgs are winning out, obviously. Who needs the NYT when the Force is clearly with the Porgs.

JWatts

  Hide Replies

"2. Is China the big loser in the new NAFTA?"

I think their is some truth to this, but mostly it's a weak tea argument about a 2nd order effect. Most of the gain to the US is going to be a slightly better deal for the US. Though, I'm happy to have manufacturing jobs go to Mexico over the China. But historically, Mexico, hasn't been able to run manufacturing facilities nearly as efficiently as China. I'm not convinced a bureaucratic dictate is going to fix this type of fundamental issue.

rayward

  Hide Replies

Why doesn't Trump just order Tim Cook and the other CEOs who shifted production to China to "bring those jobs home". WWPD (What Would Putin Do). He would order them to "bring those jobs home". The reality is that Trump isn't all that concerned about American labor. The trade war is just part of the reality show.

JWatts

  Hide Replies

"Why doesn't Trump just order Tim Cook"

Because he's a secret Nazi, rayward. He couldn't maintain his membership in the secret fascist club if he came out of the closet.

JWatts called Trump a secret Nazi

  Hide Replies

Trump seems to be ordering everyone else about how they should do their jobs. Harley Davidson, Washington Post, Google, Amazon, anyone? The Secret Nazi Trump shouldn't act so coy and embrace his inner Josip Stalin.

EverExtruder

  Hide Replies

Wow a nazi and Josef Stalin! What an internet treat! I've been waiting for that one for quite some time. You missed the trifecta of adding a Hitler or Mussolini reference though....but A for effort.

Hazel Meade

  Hide Replies

The price of cars will go up. That's a better deal for the US auto industry but not "the US". Stop conflating the interests of US manufacturers with the US as a whole.

a clockwork orange

  Hide Replies

The Oprah Industry is nothing to scoff at. In the 2000/10's the incidence of sex crimes on campus went up. So too did gang violence. Why isn't Oprah Winfrey talking about law and order? Why isn't Aziz Ansari doing a tell all?

''It's a new era on college campuses, where everybody - students, faculty, staff - is trying to pay greater attention to it,'' said William Schafer, director of the office of student conduct at the University of Colorado at Boulder. If one were to compare Louis C.K. to your everyday sexual assault misses the point. Shows how out of touch Academia is. Shows how sick celebrity really is. ''I would say 100 percent of our sexual-assault cases are alcohol-related,'' said Mr. Schafer of the University of Colorado.

Hazel Meade

  Hide Replies

#4. At $1500 per ticket it should only have taken 40 people in attendance to pay for the entire tab. That's clearly way too few people to have a large extravagant wedding of the kind she wanted. She should have invited 300 people and charged them $200 each in lieu of a wedding gift. This would be more economically optimal considering gifts are often not what the receivers want anyway. Just say that the best gift they can give is to help pay for the wedding. Plus the guests get some value for the $200 fee - food drink and presumably entertainment.

Becky Hargrove

  Hide Replies

Great video. I never had a cat that appreciated this Debussy favorite either, at least that I remember. Some would get upset and walk across the piano keys.

Hazel Meade

  Hide Replies

If the price of cars manufactured in North America rises (which is what the agreement will do), then that will hurt US auto exports to the rest of the world, which might actually benefit China in foreign markets. The US cannot assume (as it did in the 60s or the 80s) that the US market is the only market in the world that matters.

Slocum

  Hide Replies

I think it's mainly been the German and Japanese transplant manufacturers who've been exporting cars from the U.S. plants. But the point holds -- if production costs in the U.S. go up due to tariffs or changes in NAFTA, they'll stop or cut back on doing that and shift that production elsewhere.

JWatts

  Hide Replies

I agree, but these changes are so minor that it's unlikely will see much in the way of price change.

These changes are certainly going to cost less than the savings by Trumps placing a hold on the CAFE standards. There was a reason that the Obama administration ensured that the really drastic MPG requirements were in the 2017-2025 time frame.

To be clear, I think these changes are unnecessary. But on the other hand, I also think they are largely cosmetic and won't have much of a change.

Hazel Meade

  Hide Replies

ut these changes are so minor that it's unlikely will see much in the way of price change

So are you are in favor of the $15 minimum wage? Changes so minor nobody will notice, right?

JWatts

  Hide Replies

"Changes so minor nobody will notice, right?"

Exactly. Or more likely a lot of press will be spilled on the changes, but they'll be too minor for most people to even notice.

This isn't a $15 minimum wage. It applies to 45% of the labor force for manufacturing cars, not somebody running the cash register at McDonalds. Since most manufacturers already pay more than this, the correct comparison would be to the Obama minimum wage hike to $7.25. A lot of people claimed that would have dire consequences. But the effective minimum wage was already around that level so there was no significant impact.

Look, I don't support these changes. I'm generally not a fan of government regulatory socialism. But I'm not immature enough to believe that just because I find something distasteful, it will turn out badly.

I'll take this deal in exchange for blocking the insane CAFE standards that were scheduled to start last year any day. It's not a perfect world, but it's not a calamity either.

Artimus

  Hide Replies

#4 The bridezilla story is originally from The Daily Mail(Fail) which has a deserved reputation for rather dubious reporting.

Uh bruv

  Hide Replies

It's originally from a Reddit post, bruv

Rob

  Hide Replies

1. Are economists smarter than epidemiologists?

There are a few issues muddying this question. As an economics grad student who studied under Gelman and the son of an epidemiologist, I hope I have some clarification to offer.

First, an obvious and slightly pedantic point, intelligence is a distribution. What do we mean by saying, "economists are smarter than epidemiologists?" Are we saying on average? At the margin? Among professional economists and epidemiologists? The distribution of intelligence amongst economists is far wider and probably with fatter tails, but I would expect the average to be lower, since almost all epidemiologists have a PhD or MD/MPH. On to the more substantial points...

Second, I wasn't introduced to Directed Acyclic Graphs until well into my exploration of Bayesian methods in a class on missing data. The professor (Ben Goodrich, http://www.columbia.edu/~bg2382/) is a political scientist, like Gelman, not an economist, but I take it that economist is a stand in for quantitative social scientists as a whole. My dad, who teaches at UMD Medical School, was familiar with DAGs, but was perhaps educated before they fully took hold of epidemiology along with advanced computational methods. His superior statistical thinking and ability to reason from first principles outstrips all doctors and researchers I know around my age, which I attribute in part to he lack of reliance on computers. He could cogently follow and explain anything back to me better than I understood it from my Bayesian Data Analysis class, although his lack of ability to code is a major constraint. In this age, he would never get as far as he has. This leads me to believe that because Epidemiologists work in a far more well controlled environment that is well suited to frequentist methods and more narrowly defined, they are both less likely to be charlatans and geniuses than economists. Some people are just so far out on the genius scale that they are indistinguishable from charlatans (Gelman is one of these, who I firmly believe is a genius). Economics is a field that is easy to take advantage of, like entrepreneurship. It may even be the case that one person's genius is another's charlatan, such as Steve Jobs.

Lastly, the cultural point is big. Economists (post Friedman at least) come from the worlds of math and finance. These are virtuosic fields in which individuals are engaged in one-upmanship. Never mind the impracticality and inaccuracy of Black–Scholes (https://ideaexchange.uakron.edu/cgi/viewcontent.cgi?referer=https://www.google.com/&httpsredir=1&article=1410&context=honors_research_projects). It's elegant! Nobel prize worthy. The life of an epidemiologist is much more mundane. They are more likely to have studied visual models like cell diagrams and 3d molecular models in chemistry. My dad still has a 3d model of a heart sitting on his desk, even though his job is all databases and SAS. When getting a dual degree in Business (MBA) and International Affairs (MIA), I noticed a cultural gulf between the programs that is best described by a quick example. In business school, we had a number of bid points and the market sorted out which classes we got based on our bid allocation. In international affairs, seniority determined our sign up schedule, in theory, but in practice, there was also a way to get whatever class you wanted through personal relationships, cajoling, spending social capital, etc. Conway's law was in full force. I would be surprised if it wasn't as strong in debates between economists and epidemiologists.

west coast observer

  Hide Replies

excellent comment, thanks.

KL

  Hide Replies

Terrible comment!

"I would expect the average [intelligence amongst economists] to be lower, since almost all epidemiologists have a PhD or MD/MPH."

Almost all academic economists all have Ph.D.'s too.

That undergrad critique of the Black-Scholes formula is dreadful; it didn't even use options data!

Rob

  Hide Replies

Thank you west coast observer.

KL, after rereading my rambling polemic, I understand your pushback, although I think the two points you pick up are the wrong points in my argument to attack.

First, the post is not only about academic economists, but about how economics is taught. Here are three economists without PhD's that I found in ten minutes of googling:
- Bank of America Chief Economist, Michelle Meyer
- Royal Bank of Canada Chief Economist, Craig Wright
- Morgan Stanley Chief Economist, Ellen Zentner

There is, however, an issue with my line of reasoning, but it is the opposite of the one raised. The problem is that degrees tell us very little about intelligence, except that the degree holder probably passes the minimum threshold.

Yes, the paper I cite is not terribly high quality, but it illustrates the point that economists are obsessed with the prestige of mathematical prowess rather than practical applications. Epidemiologists, who deal in millions of lives, have no such luxury. Black-Scholes derives its predictive power because everybody uses it, so option data is self-fulfilling. There are many better critiques of B-S including by Taleb (https://pdfs.semanticscholar.org/38dd/59e768d13c3e4d9b8ef40e124e09526421f8.pdf) and Hurvich (http://pages.stern.nyu.edu/~churvich/Forecasting/Handouts/Scholes.pdf). Ultimately, it is the practitioner who says it best, "Where Black-Scholes really shines, however, is as a common language between options traders." (https://www.quora.com/Is-the-Black-Scholes-formula-just-plain-wrong-It-seems-to-me-that-we-are-forming-a-model-to-fit-the-market-data-It-seems-more-like-data-fitting-Any-thoughts). The value comes from the network effects of standardization, not the accuracy of the model.

All of this is neither here nor there, when it comes to the main issue at stake. The argument boils down to this, economists claim they learn nothing from drawing causal diagrams, while epidemiologists get a lot of value out of them. Framed this way, I believe Tyler is being Straussian and actually poking fun at the field of economics. It is clear that economists are not more intelligent than epidemiologists, just more arrogant.

KL

  Hide Replies

"the paper ... illustrates ... economists are obsessed with ... math".

No. The paper shows one confused undergrad tried to empirically invalidate Black-Scholes without options data.

You ignore decades of improved option models and the actual empirical options literature by financial economists.

Rob

  Hide Replies

And you ignore that practitioners ignore this literature.

Dzhaughn

  Hide Replies

#6 What we need here is a piano tuna.

mkt42

  Hide Replies

1: That is an interesting, but several years old, discussion. I have a vague recollection that shortly after that a comment in Gelman's blog did a nice job of elucidating how Pearl and Imbens/other econometricians have a different framework for studying these issues and that a lot of the seeming disagreement disappears after the arguments get translated into the other framework. Perhaps analogous to deciding whether we should draw graphs depicting Marshallian demand curves or Hicksian demand curves.

But I forget the details, and don't recall where in Gelman's blog the comment was.

David

  Hide Replies

If you’re thinking of the same comment as me, then I believe it was about how statisticians model the data while economists model the underlying data generating process

west coast observer

  Hide Replies

Hey, wait, old Nafta also disadvantaged other countries. The reason we cannot get mangoes from India is that Nafta says can get them only from Mexico (no competition from Canada on this). So how does new Nafta differ from old Nafta in discouraging trade with other countries? We need some numbers.

Barkley Rosser

  Hide Replies

The deal is NAFTA plus TPP plus an auto deal. China might be slightly hurt, with US auto industry, mostly parts, gaining a bit, with higher prices for cars. But the country that may get hit harder is Japan, with Toyota and Nissan needing to Us more North American parts in the US assembly plants.

Thiago Ribeiro

  Hide Replies

"But the country that may get hit harder is Japan, with Toyota and Nissan needing to Us more North American parts in the US assembly plants."

Maybe they should not have promoted death marches?

Yancey Ward

  Hide Replies

He should have played Welcome to the Jungle.

Axa

  Hide Replies

#2: It's to disrupt/troll the supply chain of the Germans.

Just look at the fraction of components imported by BMW, MB or VW group to the NAFTA area. Plants in South Carolina, Alabama and Puebla will need source more components locally.

Dan in Euroland

  Hide Replies

Heckman took down Pearl here: http://www.nber.org/papers/w19453

Comments for this post are closed