Month: November 2021

Immigrants keep us out of nursing homes

by  Tyler Cowen in

We examine whether immigration causally affects the likelihood that the U.S.-born elderly live in institutional settings. Using a shift-share instrument to identify exogenous variation in immigration, we find that a 10 percentage point increase in the less-educated foreign-born labor force share in a local area reduces institutionalization among the elderly by 1.5 and 3.8 percentage points for those aged 65+ and 80+, a 26-29 percent effect relative to the mean. The estimates imply that a typical U.S-born individual over age 65 in the year 2000 was 0.5 percentage points (10 percent) less likely to be living in an institution than would have been the case if immigration had remained at 1980 levels. We show that immigration affects the availability and cost of home services, including those provided by home health aides, gardeners and housekeepers, and other less-educated workers, reducing the cost of aging in the community.

Here is more from Kristin F. Butcher, Kelsey Moran, and Tara Watson.

A South Korean floating city?

by  Tyler Cowen in

The world’s first “floating city” will be built off the coast of South Korea — and it will be hurricane-proof, flood-proof and self-sustaining.

The United Nations-backed project will be constructed off the coast of Busan and was designed as a response to rising sea levels, Business Insider reported. It is expected to be completed by 2025.

The futuristic city will be made up of large hexagonal platforms that float on the water. The design will utilize a limestone coating that’s buoyant, despite being two to three times as hard as concrete, project leaders told the outlet.

The ultimate goal of the pioneering project is to create a flood-proof community able to withstand a Category 5 hurricane for 10,000 residents. It will also be able to produce its own food, energy and freshwater with communal farming at the “heart of every platform,” according to its designer Oceanix.

Here is more from The New York Post.

Jason Abaluck writes me about masks and the Bangladesh RCT study

by  Tyler Cowen in

This is all him, no double indent though:

“As a regular reader of your blog and one of the PIs of the Bangladesh Mask RCT (now in press at Science), I was surprised to see your claim that, “With more data transparency, it does not seem to be holding up very well”:

  1. The article you linked claims, in agreement with our study, that our intervention led to a roughly 10% reduction in symptomatic seropositivity (going from 12% to 41% of the population masked). Taking this estimate at face value, going from no one masked to everyone masked would imply a considerably larger effect. Additionally:
    1. We see a similar – but more precisely estimated – proportionate reduction in Covid symptoms [95% CI: 7-17%] (pre-registered), corresponding to ~1,500 individuals with Covid symptoms prevented
    2. We see larger proportionate drops in symptomatic seropositivity and Covid in villages where mask-use increased by more (not pre-registered), with the effect size roughly matching our main result

The naïve linear IV estimate would be a 33% reduction in Covid from universal masking. People underwhelmed by the absolute number of cases prevented need to ask, what did you expect if masks are as effective as the observational literature suggests? I see our results as on the low end of these estimates, and this is precisely what we powered the study to detect.

  1. Let’s distinguish between:
    1. The absolute reduction in raw consenting symptomatic seropositives (20 cases prevented)
    2. The absolute reduction in the proportion of consenting symptomatic seropositives (0.08 percentage points, or 105 cases prevented)
    3. The relative reduction in the proportion of consenting symptomatic seropositives (9.5% in cases)

Ben Recht advocates analyzing a) – the difference in means not controlling for population. This is not the specification we pre-registered, as it will have less power due to random fluctuations in population (and indeed, the difference in raw symptomatic seropositives overlooks the fact that the treatment population was larger – there are more people possibly ill!). Fixating on this specification in lieu of our pre-registered one (for which we powered the study) is reverse p-hacking.

RE: b) vs. c), we find a result of almost identical significance in a linear model, suggesting the same proportionate reduction if we divide the coefficient by the base rate. We believe the relative reduction in c) is more externally valid, as it is difficult to write down a structural pandemic model where masks lead to an absolute reduction in Covid regardless of the base rate (and the absolute number in b) is a function of the consent rate in our study).

  1. It is certainly true that survey response bias is a potential concern. We have repeatedly acknowledged this shortcoming of any real-world RCT evaluating masks (that respondents cannot be blinded). The direction of the bias is unclear — individuals might be more attuned to symptoms in the treatment group. We conduct many robustness checks in the paper. We have now obtained funding to replicate the entire study and collect blood spots from symptomatic and non-symptomatic individuals to partially mitigate this bias (we will still need to check for balance in blood consent rates with respect to observables, as we do in the current study).
  1. We do not say that surgical masks work better than cloth masks. What we say is that the evidence in favor of surgical masks is more robust. We find an effect on symptomatic seropositivity regardless of whether we drop or impute missing values for non-consenters, while the effect of cloth masks on symptomatic seropositivity depends on how we do this imputation. We find robust effects on symptoms for both types of masks.

I agree with you that our study identifies only the medium-term impact of our intervention, and there are critically important policy questions about the long-term equilibrium impact of masking, as well as how the costs and benefits scale for people of different ages and vaccination statuses.”

That was then, this is now

by  Tyler Cowen

The word “pogrom”…entered the international lexicon in the 1880s. By the early 1920s, it had become so familiar that a champion racehorse in Britain was named Pogrom; it won the 1922 Epsom Oaks and the Coronation Stakes.

That is from the new and fascinating book by Jeffrey Veidlinger, In the Midst of Civilized Europe: The Pogroms of 1918-1921 and the Onset of the Holocaust.  This is one of the better books for explaining how the Holocaust was possible, in this case not focusing on Germany of course.

Bitcoin and Electricity

by  Alex Tabarrok in

How many times have you read something like this, “Bitcoin uses as much electricity as Malaysia or Sweden or Denmark or Chile….”. What a bore. Have you ever wondered, however, why the comparison is to countries? Why don’t they ever tell you what would seem to be a more natural comparison which is how much “Bitcoin” spends on electricity?

The reason is that electricity is incredibly cheap so Bitcoin electricity expenditures priced in dollars don’t look very large. Bitcoin uses something like 100 terawatt hours (TWH) of electricity annually (depending on the price of Bitcoin) but a TWH costs less than $100 million (10 cents per KWH times 1000000000). Thus, Bitcoin spends say $10 billion on electricity annually. (In fact, it’s less than this since bitcoin miners can be located in places where electricity prices are especially cheap.)

$10 billion in spending isn’t a lot. It’s less than the world spends on toothpaste ($30b), much less than the US spends on cigarettes ($80b), and considerably less than the US Federal government spends in one day ($18.65 billion).

If we think of the $10 billion spent by Bitcoin as a security budget (as the spending secures the blockchain) it also compares reasonably to US bank spending on cybersecurity. Bank of America alone spent more than $1 billion on its cybersecurity budget and the total financial security budget is much larger.

None of this proves that Bitcoin spending is well spent but it puts things in context. It is also true, of course, that most of the new crypto platforms such as Elrond (I am an advisor) use proof of stake which uses much less electricity than proof of work.

Still, the next time you read that Bitcoin consumes as much electricity as Sweden substitute Bitcoin spends as much on electricity as Americans spend on Halloween costumes.

Photo Credit: MaxPixel.

Best movies of 2021

by  Tyler Cowen in

Listed in the order I saw them, noting that foreign releases get classified by their USA release year.  And sometimes you will find a review of mine behind the link.

The Dig.  Archaeology and British restraint, circa 1939.

Minari, Korean immigrant family in Arkansas.

Promising Young Woman, black comedy, wow, brutal.

Another Round, Danish film about alcohol.  Against it.

Sweat, fame, social media, and gender, but insightful rather than the usual b.s.

Old, Shyamalan, plot twists and conceptual thought experiments about the biomedical establishment.  Imperfect, but delivers in some significant ways.

Green Knight

Night of the Kings, Ivory Coast prison movie.

Nine Days.  If you were choosing who gets to be born and walk on the earth, which interview questions would you ask them and how would you evaluate their answers?

“Ich bin dein Mensch” — should you date a robot?

The Many Saints of Newark

McCartney 3, 2, 1 (Hulu).  Self-recommending and consistently interesting, the best segment is when you get to hear just how much value Paul’s bass line added to “While My Guitar Gently Weeps.”

Lamb (Icelandic), one of my favorites for the year.

Dune

Get Back, directed by Peter Jackson.  Amazing and epic, my pick for best of the year.  And the whole world is responding.  And read this on the miracle of the production side.

Cinema is not back to where it used to be, but this list is way better than what last year was able to deliver.  And as always I will let you know if I see anything notable between now and the end of the year (notable films often cluster in December, though they tent not to be my personal favorites, furthermore the release schedule remains somewhat disrupted).

What else was there?

Your political views are not your own

by  Tyler Cowen in

In a unique sample of 394 adoptive and biological families with offspring more than 30 years old, biometric modeling revealed significant evidence for genetic and nongenetic transmission from both parents for the majority of seven political-attitude phenotypes. We found the largest genetic effects for religiousness and social liberalism, whereas the largest influence of parental environment was seen for political orientation and egalitarianism. Together, these findings indicate that genes, environment, and the gene–environment correlation all contribute significantly to sociopolitical attitudes held in adulthood, and the etiology and development of those attitudes may be more important than ever in today’s rapidly changing sociopolitical landscape.

Here is the full piece from Emily A. Willoughby, et.al.  Via the excellent Kevin Lewis.

An update on the mask debate

by  Tyler Cowen in

I am long since tired of this debate, and I see that a lot of people are not joining it in the best of faith.  I can pass along a few updates, namely this study, with some critical commentary attached.  And here is more on the Bangladeshi mask RCT.  With more data transparency, it does not seem to be holding up very well.

That said, I am not sure that either calculation really matters.  Any good assessment of mask efficacy has to be radically intertemporal in nature, and I mean for the entirety of the pandemic.  “Not getting infected” now may well raise your chance of getting infected later on, and that spans for longer than any feasibly designed RCT.  And have you heard about the new “Nu” variant?  It may turn out not to matter, but it does remind us that the pandemic is not over yet.

As a simple first approximation, think of the real value of masks as “a) how many infections are delayed for how long, plus improvements in treatment in the meantime, plus b) how many infections are avoided altogether.”  Even a well-designed RCT is going to focus on a version of b), but only for a limited period of time.  The extant studies don’t at all consider “plus improvements in treatment in the meantime,” or when some of those protected by masks for say a year or two might nonetheless later catch Covid later yet.  So those RCTs, no matter what their results, are grabbing only one leg of the elephant.

To make matters more complicated yet, a “very small” efficacy for masks might (yes, might) translate into a much larger final effect, due to effective R (sometimes) being greater than 1.  So finding a very small effect for masks doesn’t mean masks are only slightly effective.  As the pandemic is ending, you might (again might) have had one less “pandemic cycle” than if you hadn’t tried masks at all.  You can think of masks as a kind of lottery ticket on “one big gain,” paying off only when the timing is such that the masks have helped you choke off another Covid wave.  Again, the RCT is not capable of estimating that probability or the magnitude of its effect.

Yet another part of my mental model of masks has evolved to be the following.  You have two sets of countries, countries that manage Covid well and countries that don’t, argue all you want who goes into which bin but that isn’t the point right now.

Now consider the countries that don’t manage Covid well.  They might wish to stretch out their epidemics over time, so that better treatments arrive, subject to economic constraints of course.  But the countries that manage Covid well probably want the poorly-managed countries to reach herd immunity sooner rather than later, if only to lower the ongoing risk of transmission from a poorly-managed country to a well-managed country.  And to lower the risk of those countries birthing new variants, just as southern Africa now seems to have birthed the Nu variant.

So we have two major points of view, represented by multiple countries, one wanting quicker resolution for the poorly managed countries but the other wanting slower resolution.  Does any study of masks take those variables into account?  No.  Nor is it easy to see how it could.

To be clear, I am not arguing masks don’t work, nor am I making any claims about how much masks may or may not protect you individually, or the people you interact with.  I am claiming that at the aggregate social level we are quite far from knowing how well masks work.

I say it is third doses we should be doubling down on, not masks.  To be clear, I am fine with wearing masks myself, I am used to it, and I dislike it but I don’t hate it.  On this issue, I am not one of those people translating his or her own snowflake-ism into some kind of biased policy view.

But the emerging science on third doses is much stronger, and most countries have been dropping the ball on that one.

Simple advice for watching and understanding on-line chess

by  Tyler Cowen in

Yes, the computer evaluations are extremely useful.  But they are measuring the quality of the position when two computers are playing.  Yet most of the games you care about tend to be two humans playing each other.  And those humans do not play like computers.  The computer might say the game is even, and maybe it is with perfect play, but one side can be much harder (easier) to play than the other.  So I suggest this trick.  Go to analysis.sesse.net, which covers top games (only).  Scan down the vertical list of all possible moves and consider the distribution of outcomes.  If the top move is great for White, but all the others are not, robustness is low, especially if the top move for White is not super-obvious (such as recapturing a Queen, etc.).  If all the sequences look very good for White (Black), you will know that for humans the position probably is somewhat better for White than the single computer evaluation number will indicate.  Robustness against human error will be present.

For the Carlsen match, here is a good Twitch stream, currently with Caruana as commentator.

Air Pollution Reduces Health and Wealth

by  Alex Tabarrok in

Great piece by David Wallace-Wells on air pollution.

Here is just a partial list of the things, short of death rates, we know are affected by air pollution. GDP, with a 10 per cent increase in pollution reducing output by almost a full percentage point, according to an OECD report last year. Cognitive performance, with a study showing that cutting Chinese pollution to the standards required in the US would improve the average student’s ranking in verbal tests by 26 per cent and in maths by 13 per cent. In Los Angeles, after $700 air purifiers were installed in schools, student performance improved almost as much as it would if class sizes were reduced by a third. Heart disease is more common in polluted air, as are many types of cancer, and acute and chronic respiratory diseases like asthma, and strokes. The incidence of Alzheimer’s can triple: in Choked, Beth Gardiner cites a study which found early markers of Alzheimer’s in 40 per cent of autopsies conducted on those in high-pollution areas and in none of those outside them. Rates of other sorts of dementia increase too, as does Parkinson’s. Air pollution has also been linked to mental illness of all kinds – with a recent paper in the British Journal of Psychiatry showing that even small increases in local pollution raise the need for treatment by a third and for hospitalisation by a fifth – and to worse memory, attention and vocabulary, as well as ADHD and autism spectrum disorders. Pollution has been shown to damage the development of neurons in the brain, and proximity to a coal plant can deform a baby’s DNA in the womb. It even accelerates the degeneration of the eyesight.

A high pollution level in the year a baby is born has been shown to result in reduced earnings and labour force participation at the age of thirty. The relationship of pollution to premature births and low birth weight is so strong that the introduction of the automatic toll system E-ZPass in American cities reduced both problems in areas close to toll plazas (by 10.8 per cent and 11.8 per cent respectively), by cutting down on the exhaust expelled when cars have to queue. Extremely premature births, another study found, were 80 per cent more likely when mothers lived in areas of heavy traffic. Women breathing exhaust fumes during pregnancy gave birth to children with higher rates of paediatric leukaemia, kidney cancer, eye tumours and malignancies in the ovaries and testes. Infant death rates increased in line with pollution levels, as did heart malformations. And those breathing dirtier air in childhood exhibited significantly higher rates of self-harm in adulthood, with an increase of just five micrograms of small particulates a day associated, in 1.4 million people in Denmark, with a 42 per cent rise in violence towards oneself. Depression in teenagers quadruples; suicide becomes more common too.

Stock market returns are lower on days with higher air pollution, a study found this year. Surgical outcomes are worse. Crime goes up with increased particulate concentrations, especially violent crime: a 10 per cent reduction in pollution, researchers at Colorado State University found, could reduce the cost of crime in the US by $1.4 billion a year. When there’s more smog in the air, chess players make more mistakes, and bigger ones. Politicians speak more simplistically, and baseball umpires make more bad calls.

As MR readers will know Tyler and I have been saying air pollution is an underrated problem for some time. Here’s my video on the topic: