In the U.S., you are legally permitted to sell your blood plasma for money, but it's called "donation".
In recent years, the numbers of places set up for donations has skyrocketed, and the amount they're compensating donors has followed suit.
Let's dig in.
Most of the visitors to these donation centers are highly local (A)
People are also more likely to visit donation centers in census block groups (CBGs) that are marked by poverty (B)
Why?
In surveys, donating plasma is predicted by being poor, Black, male, married, having kids, being a student, etc.
In short, people who could really use an extra $150 twice a week for a minimal inconvenience are more likely to donate.
We know this is true because we also have survey data indicating people's stated reasons for donating. Few people are donating altruistically. The top categories by far are about money!
Take a look:
Now, before getting to the juicy result, I want to show one more thing: the impact of COVID stimulus checks on plasma donations.
When the checks went out, the number of visits to plasma donation centers cratered. Donations plummeted because people had the cash they needed.
Now here's the kicker: When plasma donation centers open up, local inquiries into predatory payday and installment loans falls off.
People are seeking credit and donating blood might be how they get it.
If we stratify these trends by age, we see that those with ages less than or equal to 35 - the less well-established - are the ones deciding to use blood plasma donations to offset the need for quick, dangerously high-interest cash, not those greater than 35 years old.
When you look at payday transactions rather than inquiries alone, you get the same picture, albeit with more noise.
Young people really do seem to be defraying the need for credit by selling (sorry, donating!) their blood plasma.
The reasons people sell their are also, evidently, not just to cover essentials.
One of the clearest-cut impacts is that entertainment establishments see an increase in visits after blood plasma donation centers open up.
That last part clarifies something: people would prefer not to get risky, high-interest loans, and they really want a little bit of extra cash. So while they will seek out those loans if push comes to shove, they're more likely to frivolously pursue blood donation.
And that's good! We need blood plasma donations, so if paying people a bit of money makes that possible, so be it.
If we take away that possibility, we can also see that it would make people's lives worse.
How far can we take this? Maybe we can learn from Iran.
In the U.S., about 0.5-1% of the federal budget goes to dialysis:
In Iran, there is a legal, regulated market in selling kidneys and it's such a good deal that the government even pays for the operations. It beats paying for dialysis!
People are more than willing to sell their bodies in different ways that help their fellow citizens, from selling their blood plasma to pawning off a kidney.
Given so many people want to do that, and so many people would benefit from it, the question is, why not?
Frankly, I think we should just do it. The blood plasma donation model has been such a success and it's more than evident that organ payments could be too.
In the U.S., immigrants commit crimes at a lower rate than natives do.
But this wasn't always so true!
In the 19th century, immigrants and natives were much more similar in terms of how often they committed crimes🧵
One possibility?
Changing racial composition in "US-Born". This might happen because Blacks--who do crimes at higher rates than Whites--were a growing share of that category.
But that isn't it. Subset to Whites, same result, albeit with different timings and magnitudes:
Another possibility?
Changes in the sourcing of immigrants. Immigrants might come from places with less crime than they used to.
Alas, this is wrong. If anything, they come from places with more crime today. All sorts of adjustments don't change the main picture here.
A few days ago, Biden commuted the death sentences of almost every federal death row inmate.
Every single person whose sentence Biden commuted was verifiably evil and clearly earned the death penalty.
Let's go through all 37🧵
Shannon Agofsky drowned a bank manager alive, received life in prison, and in prison, kept talking about how he was itching to beat up other prisoners.
Then he killed a fellow prisoner by stomping his neck in and causing him to drown in his own blood.
On camera. Guilty.
Billie Allen killed a bank guard during a bank robbery, using a semi-automatic weapon.
Allen and his accomplice also stole two vans to use as getaway vehicles the night before.
He was inspired by the movies "Set It Off" and "Heat" and he was caught red-handed. Guilty.
"Without Mohammed, Charlemagne would have been inconceivable."
This quote describes Pirenne's thesis that Antiquity—the period when economic activity concentrated in the Mediterranean—ended because the rise of Islam destroyed the flow of trade across it.
The decline in trade that resulted from differences in faith had profound consequences for the economic geography of Europe.
Byzantine economic activity depended on trade, and it collapsed, whereas the Frankish economy, which was never trade-dependent, transformed.
The Byzantines' minting stalled and the Arabs' and Franks' increased (perhaps partly because they were cut off from one another!), providing each of their states with divergent trends in seignorage revenues and a widening gulf in the ability to fund the government.
Robustness tests are supposed to show a study's results hold up no matter how you reasonably change the specification
But we live in a world with p-hacking, and people p-hack robustness tests
Compared to unshown robustness tests (blue), what we get is suspiciously significant!
This is the distribution of z-values for different tests in economics papers, coupled with the robustness tests their authors presented, and other, plausible robustness tests they didn't.
Clearly people p-hack, and they p-hack tests that are supposed to make us think they didn't
It's sad this is the case. Were it not, it might be useful to get a surprising, marginally-significant result, and then show that it holds up across different permutations of the results
But because the robustness tests shown are selective, their potential utility is unrealized
Let's talk about the glass delusion, the Middle Ages' bout with a mass psychogenic illness marked by people believing they were made of glass.
Glass was a valuable commodity in Europe. It was primarily owned by the noble and well-to-do, and it had a notable purpose in alchemy.
Its perception as the technology of the time was as one that's both fragile and valuable, like the nobility.
Glass was the relatively novel technology people knew, and they knew things could be transmuted into glass. Delusional people also thought transmutation could affect them.