Monthly Roundup #1

Link post

When I split entirely non-medical, non-Covid topics off from the weekly Covid post, I expected to have a grab bag of them on a weekly basis. Hence the name ‘weekly roundup.’

What changed were four things.

  1. The flow of worthy one-off items decreased and they seemed less time sensitive.

  2. The flow of larger topics in the news increased, including AI, FTX and Twitter.

  3. It became increasingly clear more focused posts have higher long term value.

  4. I started spending more time exploring longer-term questions.

The result of this was that it no longer made sense to do a full weekly roundup. It still does make sense to do this periodically, so for now I am going to move these posts to monthly. I may spin off some of the sections (e.g. the FTX one) as part of future posts, but have no present intention to do so.

Bad News

Business Insider reportedly claims that as long as they say ‘reportedly’ in their headline, they are in the clear even if they know the thing ‘reportedly’ claimed is absolutely false, because it is true that it was reported. Journalism.

Germany to reduce rather than increase military spending, despite government promises to the contrary.

European court decides Right to Privacy too important, takes down registers of beneficial owners.

I am generally skeptical of predictions such as Samo’s that Europe might soon be discussed as a ‘formerly industrialized economy’ and yet there is a growing movement in French towns to ‘bring back the horse and cart as an alternative to fossil fuels and a way to slow down urban life.’

The UK effectively charges a very large tax for access to its postcode address file, making it out of reach for many uses and also being effectively a large tax on business, especially small business, that requires the information. This seems like quite an insane place to collect this much government revenue.

Lucky person uses social position to warn others of punishment for focusing on luck.

Yes, the government will always use all the data available, such as using your treatment of NPCs in games like Skyrim and Fallout to predict future psychopaths.

IRA’s Made in America requirements going exactly how you would expect, except for the whole ‘excluding allies was a typo, sorry’ incident.

I was expecting In Defense of Being Alone to center around being alone being kind of awesome these days. All the games and books, all the shows and movies, the internet and even GPT at your fingertips. Whereas the defense was the opposite of this, that the political norm enforcement has made group participation worse and so opting out is a rational response. I do not buy that such costs are dramatically higher now than they were in most of the past, and political shibboleth enforcement does not seem to impose unique costs versus other forms.

I do think that the people choosing to be alone more often are mostly making a mistake because they are discounting the compounding returns and general human needs from in-person interactions. There is a balance for those of us who also need alone time, yet in person interactions are something I know I under-consume. Whereas for video calls I am not convinced which mistake I am making.

Math thread pointing out that our crop subsidies are simply not big enough to substantially impact our eating habits. Shifting $15 billion of payments from other crops into fruits and vegetables, even if all the benefits went to the consumer, would not move prices enough to visibly shift consumption. This mostly seems right if we model the impact of government interventions as lowering prices. The trick is that often what government is doing is instead restricting supply and raising prices. That is a very different story. The impact on the price of the thus ‘protected’ products often, as I understand it, shifts dramatically. As an example, here’s sugar prices. So, while I agree that ‘shift our subsidizes’ is not the answer,we can still blame the government for a bunch of our bad outcomes here.

People Are Were Trying To Destroy the Internet

Schumer planned to introduce the JCPA into the must-pass NDAA, going full ‘link tax.’ This would have created a media cartel of the big companies, who would get to dominate over smaller companies, that would extract payment for every link, that others would be mandated by law to ‘negotiate’ with, and have ‘must carry’ provisions, and this was described this as a first step towards linking no longer being ‘fair use.’ So full mafioso behavior, here’s the price and you have no choice but to buy the amount we tell you to.

There is at least one report that the ‘price’ of this deal was getting rid of the vaccine mandate on our troops. On the plus side, Facebook said it might respond by removing news from its platform.

A few days later, the JCPA was stripped back out of the NDAA.

This keeps happening.

  1. We are told various people are trying to destroy the internet.

  2. ???

  3. The internet is not destroyed.

That doesn’t mean the internet is never destroyed at all. It does mean that the almost all of these bills end up either not passing, being thrown out or quietly get ignored. We will see what happens with the Texas and Florida laws, which seem to set up direct contradictions with the laws elsewhere. Various European laws that make very little sense do remarkable amounts of damage. GDPR is a nightmare.

I wanted my game, Emergents, to have the players chat with each other. Yeah, no. That’s not remotely legal in practice, letting people talk to each other.

So is this the boy who cried wolf? The boy who cried p(wolf)>0.01 while using words that did not make that clear? Is this a continuous self-preventing prophecy?

I don’t know. There’s a lot of wolf crying out there. There are a lot of very bad things introduced into bills that then get stripped out, and also a lot of bad things that get successfully slipped into bills on other occasions. It is hard for an outsider to know how close such things were to passing, or how important it is to raise public alarms.

I take the wolf cries collectively as an important fire alarm and also endorsement of the fire department.

Legislators really do come remarkably close to enacting remarkably terrible laws that would burn substantial percentages of the world’s value, in order to transfer some of that value from those they dislike to those they like, or to those who paid to play, or they do it simply by not understanding what they are doing. They also actually do this all the time, which is even more important to know, and it is also important to know that remarkably often they get close and then fail because people heroically stop them.

It is a procedural drama, 22 episodes a year, most look similar, some surprise you.

Then we read that Greece is going to outlaw password managers? Europe is going to allow people to allege that information about them is inaccurate and demand Google take it down or else? Yeah.

Good News, Everyone

What is a vegetarian? How do you define vegetarian? If a vegetarian has a 23 chance of having eaten meat in the past 24 hours, it means ‘rich enough to engage in virtue signaling,’ I suppose (original source)?

Only 1.34% of total firm wage bills go to employees engaged in regulatory compliance, says new working paper. I roll to disbelieve that this is so low even if we are talking only about employees whose sole or primary purpose is explicitly regulatory compliance, rather than adding up ‘things we do and spend time on to ensure we comply with the regulations.’ As someone who has run a business there is no f***ing way that the amount of time spent on regulatory compliance is in the low single digits, that’s a Seriously WTF Are You Even Talking About style of number.

Patrick McKenzie thread about Stripe efforts to combat credit card fraud, via introducing tiny amounts of friction to credit card testing when there is reason to suspect something fishy is going on.

Airline seat pricing theory and strategy, everyone needs to up their game edition.

I do not get the tiny price differentiations above. You are paying the complexity costs and feel-bad-about-paying costs of differentiated prices without making the prices remotely clear the market. Even those who do take the lower price won’t have that ‘save money’ itch properly scratched.

Nate Silver’s proposed strategy seems great. You have a substantial chance of an idle middle seat remaining unclaimed and willingness to switch should be very high, even higher if the person thinks they ‘saved money.’ If they refuse, now you don’t have to feel bad about inconveniencing them.

While I Cannot Condone This

A very easy philosophy of probability question with a clear right answer. In case you are wondering, when asked with the exact text from the comic, ChatGPT wimps out in a very invalid way and says it depends on what you mean by “equally likely.”

Let my hat be tipped to those who can pull this off.


I have learned that I simply cannot handle this level of meetings or structure. This is a relatively mild case, even, with Fridays at least somewhat allocated to deep work and evenings, dinner and bedtimes reserved. Still, seems to me like a nightmare.

There are those who have it far worse than I do. Venkatesh Rao says he can only do one thing per day – if he has a meeting, that’s his day.

The Efficient Market Hypothesis is False

The question, as always, is how false?

Scott Sumner suggests it is close enough that one can use avoid looking at prices in the supermarket, and order the most popular iteration of a consumer product on Cyber Monday, and the time saved will be worth it.

For consumer products my experience is that this is safe if and only if you are confident that your preferences are similar to those of others. Note that this is the kind of thing ChatGPT does quite well – telling you what the typical person will be saying in some spot, so you can compare to your own preferences. If all you care about are reliability, cost and generic ‘quality’ in roughly normal ratios, then mostly, sure, do some mix of ‘most popular’ and ‘highest rating’ and call it a day. When you have additional preferences, you can’t do that.

For supermarkets, it depends on your hourly rate, whether you intend to repeat the transaction, and to what extent you are debating which store to use.

In my experience, supermarkets absolutely do not price their things efficiently within a store, nor do different stores offer balanced trade-offs between quality, availability and price.

Instead, different stores target different market segments and have different specializations. When you match what you are buying to what they do well, you get a good deal. When you mismatch, you get a bad deal.

  1. You want to buy lower-quality goods, or fully generic goods, at lower-quality stores. They will offer the best prices on the same or similar brands, and their store brand will typically be a superior low-end choice.

  2. You want to buy high-quality goods at high-end stores. This is partly about ‘they will have them at all’ but it is also more than that. They will actively charge less for the same product, or a similarly high-quality product.

  3. Some stores are simply worse than others, often by quite a lot. In my neighborhood, Westside Market and Morton Williams are clearly overpriced, and dominated by Fairway. Back at my old apartment there was this place called Gracefully and walking in there was pure pain. Whereas Fairway and Whole Foods do different jobs, and Trader Joe’s is in an amazing spot on the price-quality curve for most of the goods they carry.

  4. Often places will have particular loss leaders or other intentionally aggressive pricing, along with other intentionally too-high pricing to take advantage.

  5. Between brands, stores will sometimes randomly jack up some of the prices and not others, on the theory that those like Scott who don’t check will sometimes buy the expensive ones (likely on the easier to reach shelves too) whereas those who check can feel smart. And not only between but also within brands, too. Sometimes there will be a ‘correct’ size to choose, and you’ll get a worse deal if you choose a different one.

  6. Why do I go into such detail and care so much? Partly that it is visible. Partly that I grew up in a family very concerned with not overspending. Partly as training.

Sadly, FTX

Things continue to happen or come to light, some of which is gathered here in grab bag form. I think I am close to finishing gathering my overall thoughts and may write that up at some point.

Why hasn’t SBF been arrested yet? Popehat explains that building an ironclad case takes time, and rushing carries a lot of risks, and we need to be patient. Juries need to hear exactly what happened if you want a conviction. Us knowing that criminal acts were performed does not make this requirement go away. Nor does this requirement make our knowledge go away, either.

“SBF won’t shut up and it’s driving lawyers mad.” From what I can tell it’s mostly driving uninvolved lawyers mad simply through the pain of being a witness, as anyone SBF does manage to hire simply fires him a few days later.

SBF continues his apology tour. Whatever your explanation for why, whatever the issues of ‘journalistic ethics’ or legal risk in whatever direction, the mainstream media continues to largely buy his story, including NYT, WaPo and WSJ and more. Here’s the Financial Times.

Who else is seeing fraud and saying fraud? At least CoinDesk. The New Yorker piece is being more ‘precise’ yet is both full of excellent detail and does not pull punches.

Dustin Muskovitz stepped forward on Twitter to offer some examples of reasonable media coverage.

Who are some others at least offering coverage that at least takes the misdeeds seriously, even if it’s not willing to go all the way? Vox (1, 2 and 3 on top of the Kelsey Piper story). NY Mag, NY Post (literal headline: ‘don’t trust mainstream media, SBF’s no schlub, just a crook’), CNBC in at least this instance, and to at least some extent Fortune Crypto (1, 2, 3). This is a third thing.

Patrick McKenzie offers a model of this as a game of Diplomacy, played in view of those who don’t play, where no one wants to act first until everything is fully fact checked. An alternative is that they are… playing Diplomacy, with those who get to play, and we don’t control any supply centers.

There may also have been a bit of regulatory capture.


Sunlight, sometimes effective in remarkably small quantities.

I was pointed at this ‘takedown’ of SBF/​FTX/​Alameda, from well before the collapse. Some of the points are excellent, raising a bunch of real red flags. However they are mixed in with a full-service hit job with everything he could come up with, including ‘look at this guy, who would fund him (which is ugly and awful in various ways),’ calling everyone involved ‘glorified interns’ who couldn’t possibly know how to trade because they lack impressive mentors and track records (way off base), and various forms of ‘well I don’t know things so everything is terrible.’ Some of the questions were clear cases of him Not Doing The Research.

One could however argue that, given the real red flags, the other stuff looms larger and the Bayesian likelihood ratios add up fast. There is some truth in that. This still felt more like a ‘smack down those kids who aren’t playing by my rules’ attitude than anything else. Then again, those kids very much did not play by not only ‘his’ rules but any sane (or ethical or honest or legal) set of rules, and very much didn’t in many ways know what they were doing. So there is that. And when you call total fraud and you’re right, that is worth a lot, including hinting that perhaps your reasoning is more effective than it looks, however ugly it might seem even in hindsight.

Zohar Atkins offers his thoughts. It is interesting to get this more philosophical and rabbinical perspective. I think it mostly misses the central things happening.

A link between FTX and binary options, illustrative of the general link between crypto and gambling for reasons, which is partly for reasons too obvious to explain and partly because both involve skill at money laundering. I do not think the linked post does the job of establishing the outright scam nature of this particular FTX operation, mostly asserting that binary options operations are inherently and often scams. Which, true, fair, point, still not conclusive here.

FTX had high ESG ratings, despite being perhaps the world’s largest zero-G business, and the E-implications of its focus on cryptocurrency. Almost as if ESG ratings are a measure of how much you work the system and are liked by power.

Another list of who and what might rise or fall in status. Such lists by default continue to be ‘what I would like to rise and fall in status’ to a greater degree than they are predictions. I worry mine would be no different. I also worry about focusing too much on what rises or falls instead of asking what actually happened or will happen and the gears behind it. Still, there has been some time, so what does seem to be rising and falling in status?

Falling in status:

  1. Mainstream media. Their failure to get the story in real time was surprising to many but something I expected. Their failure to stop printing puff pieces about it, not only NYT but also others. This will to some be where they spent their last gasps of credibility and exposed the nature of their game. Most of course will ignore it, as they always do. I am curious to what extent ‘look how they covered SBF afterwards’ becomes a general talking point over time.

  2. Utilitarianism. I wonder, though. The mad scramble for deontology has already died down, and in some ways equating utilitarianism with scams will make it more attractive to many rather than less. I predict more of a ‘except don’t get caught’ vibe to emerge in broader spaces over time.

  3. Virtue ethics. Sad but true. We ran the experiment, and it turns out that Tyler Cowen was wrong and no one is interested, virtue ethics is like a third party candidate, everyone agrees it works in practice but it lost the popularity contest because its arguments are too hard to understand and no one wants to throw their vote away. If this did not display why we need virtues, what will?

  4. Freedom. This will be blamed on people having freedom, and we will demand to have our freedoms taken away to prevent it from happening again. Same as it ever was. We will get more regulations of basically everything, on the margin, especially crypto and money, potentially CBDCs. This is what happens, people will say, when kids are allowed to Do Things.

  5. Effective Altruism. Lower status is not always harmful. FTX throwing money around like it did had large warping effects that made existing problems worse. The whole movement needed a Halt and Catch Fire moment.

  6. Crypto. All of it. Make no mistake. Everything in crypto falls in status when something like this happens. If you think ‘oh this only proves the case for Bitcoin’ or what not then I am sorry to report that is not how this works for most people. In the long term, perhaps you will be right. For now, no.

  7. Youth. The narrative of ‘crazy kids can’t do things’ or ‘crazy kids are always crooks’ is taking shape more than people are noticing. This is quite bad.

  8. Doing Things. People will be more inclined to assume anyone trying to Do Things is some combination of a fraud, clueless, reckless and doomed. Or to attack anyone they do not like, on this basis.

Rising in status:

  1. Twitter journalism and general amateur journalism. We had some very good coverage, and it was clear that the community was able to identify the best sources like Milky Eggs and Autism Capital. The system proved that it worked. More people will trust such sources next time, mostly for the right reasons.

  2. Deontology. Deontology won the battle against virtue ethics without lifting a finger and is the new hotness. When things go wrong people demand rules.

  3. Regulations. Especially of crypto, also of other things. No one will notice that FTX’s long term game plan was widespread regulatory capture and it came remarkably close to working. Nor will they notice that FTX was not even close to complying with even basic requirements anyway, it’s not like our existing requirements were tried and failed.

  4. Proper Authority and Very Serious People. Similarly, this will be used by those in power to reinforce the need for their power, and the dangers of letting unauthorized personnel do anything at all.

  5. Age. The narrative of ‘crazy kids can’t do things’ or ‘crazy kids are always crooks’ is taking shape more than people are noticing. This is quite bad.

Send to the Balsa Research Department

David Shor claims systematic deception in the progressive information space.

This matches my understanding of the progressive information space. There is strong pressure to suppress information that contradicts progressive narratives, leading many people to believe and say that which is not. One of the widespread beliefs in such spaces is that such manipulation of the information space advances progressive causes.

As David Shor points out, believing that which is not has a way of making one make dumb decisions. When this feeds on itself, with people whose information space is distorted then further distorting the information space of others to prioritize things that don’t match anyone’s underlying preferences, things can spiral out of control.

I, too, would respond to someone being innocent of the crime they are serving time for by not seeking to keep them in jail, regardless of whether they have a technical ‘right to appeal.’ This is one of many reasons, presumably, I am not a prosecutor, as they seem to widely disagree and the Supreme Court has to decide what to do.

California homeless math.

From WSJ

California Gov. Gavin Newsom ….recently put a temporary freeze on $1 billion of state grants for city and county homelessness programs….the measures would have reduced homelessness statewide by 2% between 2020 and 2024

[California has] more than 116,000 residents sleeping on the street on any given night.

California has dedicated some $15 billion toward the issue since the start of the pandemic.

He then does the math.

$15 billion /​ 116,000 = $129,310.34

2% x 116,000 = 2,320. $1 billion /​ 2,320 = $431,034.48

I am not saying solutions are easy or cheap. It still seems like with this level of funding we can do better.

Lower-Income Americans are Taxed Much Less Heavily Than Lower-Income Europeans. The European numbers are kind of stunning.

The top 10 percent of households in the United States earn about 33.5 percent of all income, but they pay 45.1 percent of income-related taxes, including Social Security and Medicare taxes. In other words, their share of all income-related taxes is 1.35 times as large as their share of income. That is the most progressive income tax share of any OECD nation. In Germany, the top 10 percent earn 29.2 percent of the income and pay 31.2 percent of income-related taxes, 1.07 times their share of income. The French top 10 percent earn 25.5 percent of the income and pay 28.0 percent of the income taxes, 1.10 times their share of income.

Those European ratios are, in distributional terms, practically flat taxes.

Tyler Cowen defends Biden by saying that ‘perhaps 80% of the scorecard’ is how he handled Russia’s invasion of Ukraine and that China did not invade Taiwan.

There are two obvious ways to respond to that. The first is that it is weird to give credit both for China not invading Taiwan when it was never that probable (see prediction markets if desired) that China would invade Taiwan and nothing Biden did seems to have helped with that especially, and also give him credit for how he handled Russia given that Russia did invade Ukraine and it seems plausible that there were ways Biden could have perhaps prevented this.

I do agree that once the invasion happened, Biden has handled this well in terms of outcomes so far, yet it is not clear to me how we should judge his balancing of escalation risk versus intervention impact, or where to set the bar for comparison.

The main thing I would say is, no. You do not get to make that 80% of the scorecard. So much of the scorecard are things that most people don’t even notice, that’s why Balsa exists. I’m not saying I want to give Biden any particular ‘grade’ on that stuff. To be convinced of that perspective, I would have to think that otherwise things are basically good, and ‘don’t screw anything up too badly while not fixing any of the important problems either and in fact making many of them actively worse’ would put America and the world on a good path.

Why didn’t anyone else think of that?

Also under why didn’t we think of that? Inflating your GDP growth by lying about your GDP growth, which is why dictators get to report higher growth rates.

Pure corruption is bad. Government Officials Invest in Companies Their Agencies Oversee.

Then again, it’s important to know what real hardcore corruption can look like, such as a judge who challenged those in power being convicted of taking a bribe, from a couple she’d never met, to fund expenses she had receipts to show she paid for them, to influence a her verdict in a case she recused herself from and thus didn’t hear. From there, it does not get better. Before we say how impossibly horrible everything is all the time, we should remember how most places have been ruled most of the time through all of human history, and continue to be today.

Federal marijuana laws still matter, for example New York legal dispensaries have to sell product from New York state, which is widely regarded as inferior.

Science research plans all start with ‘we are going to fund risky science’ and always snap back to standard operating procedure. This thread has some theories on why.

If you found Arxiv, you may transform science for the better. It still won’t be that great a career move.

Economists send letter supporting price cap on Russian oil exports. I presume that a sufficiently strong cap would not be so different from a ban and am curious exactly what number everyone has in mind, that keeps Russia in a mood to accept it.

Reminder: If you frame a poll, you can mostly get it to say what you want.

California’s solution costs more money, so it is an inferior way to make the problem worse unless you think – and this is a remarkably close call – that it is funnier.

Also, time to order more popcorn, while I still have a market that lets me do that.

No rush. It’s going to be 6 to 8 years… to do the reports. Never mind the construction.

Paper: If you are going to grant private control of the airport, better to align incentives and sell the place outright. Private equity acquisitions improve on many fronts after they are bought. Note that this could involve selection effects.

Seattle may soon choose approval voting for its elections, amounts of money involved in the choice are very small so if you think this would be high value (or high negative value) it is potentially high leverage. I have been looking into approval voting and continue to think it would not go well in practice once people adapt to how it works, especially in partisan spots, due to effectively having to choose ‘vote in the primary’ or ‘vote in the general,’ irrelevant alternatives being highly effective at destroying their side if they are plausibly not irrelevant, and the most extreme factions more credibly threatening defection.

Rhine River is no longer deep enough so there is a proposal to deepen it, usual suspects are against ever doing anything for any reason, and suggest ‘ships adapt’ to whatever happens, after which I am sure they will go on to oppose the ships.

Firms given ability via the H-2B visa lottery to hire more low-skill non-immigrant workers increased production and did not substantially change employment of Americans. Given those with H-2B visas rarely overstay their welcome, it seems likely we could safely eliminate the cap and thus the lottery, and grant as many such visas as there are qualified applicants.

Biden Administration continues to try and make gig not happen.

Which is it? Either the gig workers want to be employees, or the gig workers don’t want to be employees.

I especially loved this turn of phrase.

The Dispatch has further coverage of this attempt to destroy millions of American jobs (post) so we can force people to be more legible and compliant.

My proposal is that we take all gig workers at risk of reclassification, let them vote on how they want to be classified, and call it a day. Who objects?

In other Biden attempts to make things happen by not letting things happen news:

New study attempts to quantify extent to which foreign workers coming here net create American jobs, finds clear positive effect.

How different groups in Michigan do on the SAT.

The mask comes off: Advocate explicitly says that the way to measure the value of a law is how much damage it does to people they dislike.


A good point that doing everything we can to tell kids that ‘baby will ruin your life’ might be extrapolated beyond one’s teenage years and be suppressing the birth rate.

Teenage pregnancy is way down, so the relative value of marginal reductions in teenage pregnancy is way down, whereas the marginal cost of reducing the birth rate after that as people ‘pursue their dreams’ until too late is going up.

As long as consumers don’t directly see the fraudulent or absurd health care bills, it’s hard to do anything about it, people don’t care enough. So we should… no, that’s crazy.

A modest proposal as an alternative solution to parked cars blocking bike lanes, legalize low-level vandalism of such cars. So much more fun than being a narc.

Case that New York subways and major bus lines should run every six minutes during off-peak hours. I strongly endorse. Poor subway frequency greatly erodes value and pushes people towards taxis, the cost here is quite small even if it isn’t mostly or entirely made up by additional fare revenue, which simple projections say would be likely.

What happened to high speed rail in California? Among other things:


Post that points out that for the proposed cost of the HSR line, you could serve the same number of people for 40 years using a fleet of 737 jets without charging fares. This is not the knock-down argument it sounds like it is.

Works in Progress looks at New York’s architecture and building principles, is largely impressed, is puzzled by the appeal of a gridiron. I am puzzled by anyone who thinks that gridirons are not obviously correct. Sometimes you have existing stuff and you’re stuck with something else, as Manhattan is below 14th Street, and it means navigation becomes non-obvious and space is used inefficiently. Which is good if The British Are Coming and terrible otherwise. The author does note that other places will often require crossing fewer streets to travel a similar distance, which they accomplish by making many things impassible. It is possible that this is sufficiently good that there is a compromise position that is superior, where the gridiron is modified to reduce crossings by having a pattern where some streets don’t go through but every 5th one does, or something, although I doubt it.

Biden Administration faces four lawsuits against Grand Theft Education, tries to modify program to deny some of them standing. This does not inspire confidence. The whole set of rules here is so bizarre.

Scott Alexander presents: A Columbian Exchange. I am going to position myself opposite Matthew Yglesias and firmly in the ‘we have too many holidays’ camp, our schedules keep getting constantly disrupted for things no one cares about and it isn’t worth it. Go big or don’t tell everyone to stay home, I say.

Arizona’s school choice system framed as a ‘regressive tax cut.’ As I understand it, if you choose to opt out of the public school system you are given $7k/​year you can spend on your child’s education, replacing 90% of what the state would have spent on your public school. You still have to pay all your taxes, and you don’t get any of the local or federal money, so this doesn’t refund more to those in richer areas.

On the face of it this is progressive, since it is a conditional fixed payment. No one would say UBI isn’t progressive, and the rich are paying more taxes in to generate the $7k/​pupil (and also more for the local and federal costs) and then getting the same $7k/​pupil out. The argument for regressive is that wealthy parents are more likely to choose private school, so they collect the payment more often, and what matters is comparing this new world to past law where parents have to pay fully for both schools to send their children to private school (or pay for public school and the costs of home schooling).

One could thus fairly say, the reason private schools get chosen more by wealthier parents is largely that they are being put at a fixed-sized price disadvantage. If you reduce or eliminate that price disadvantage and allow fair competition, more poorer families will choose private school. This is especially true if the $7k can handle full tuition, such as with religious schools, or the marginal cost of home school, and the $7k makes a much bigger difference for them either way.

It is possible one would want to make the payment actively larger for poorer families to fully balance this out. That would seem fine. Figuring out all the numbers involved and the projected behavioral adjustments from various choices is something Balsa would be interested in doing long term, but probably isn’t its low hanging fruit.

I am curious what this change in law does to property values. A lot of the value of property in rich areas is access to superior public schools. Private schools are the most attractive when wealthy parents are living in poorer areas with relatively poor public schools. In Manhattan you go expensive private if you can, in Scarsdale you go expensive mortgage and then send the kids to public school as a benefit, so the relative value of living in a place like Scarsdale declines here.

Case by Michael Shellenberger that we should mandate treatment of the dangerously mentally ill. It certainly does not seem like ‘this person is too sick to be held accountable for their actions, and not sick enough for us to prevent those actions from being taken’ is a viable position to be taking on these matters.

Claim that after transfers, many American incomes are startlingly equal.

You can now legally jaywalk in California, you still won’t do it enough, you cowards.

Scott Sumner reminds us that, at least as of November 2022, the Fed does not seem to be tightening all that much.

The social cost of carbon varies dramatically based on the discount rate. Trump used 7%, which seems clearly too high on purpose. Biden is using 2%, which seems low without being absurd, and combining that with a bunch of other arbitrary choices gets $190 per metric ton. A rule of thumb is each $100 translates to $1/​gallon in the cost of gas.