Private housing completions have been steadily rising since 2010, but we still have a huge hole to fill:
https://fred.stlouisfed.org/series/COMPUTSA
Note that this is annualized monthly rate, so there were about 125k homes completed in May, not 1.5 million. And two-thirds of those units are single-family homes, so we need to be building much more multi-family housing.
I've been saying for years that abortion is good but Roe v Wade was a total hack job. The Constitution and rule of law matter more than any one decision.
I think about 1% of people in the US, tops, will acknowledge a distinction between a Constitutionally correct decision and their own personal policy preferences.
I looked into this a bit more. Initially I naively assumed that an airplane would be pressurized to one atmosphere, but it turns out they're actually pressurized to about 0.73-0.8 atmospheres at high altitudes. The reason for this is that this reduces the pressure differential by about a third, which presumably makes a material difference in terms of fuselage weight.
One of these is that they are subjected to the stresses of immense pressure charges. In an airplane, the high pressure is on the inside as the craft flies at high altitudes with very low air pressure outside.
I don't think this is even remotely comparable. The air pressure in an airplane is less than one atmosphere. Since there's no such thing as negative absolute pressure, there's less than a one-atmosphere pressure difference between the inside and outside. That's still quite a lot of pressure (7 PSI is about 1,000 pounds per square foot), but with the submarine, it's a difference of hundreds of atmospheres.
To me, I think it is rather obvious: folks who attain the position they do often do so because they’ll flout customs and norms and take big stupid risks that pay off.
This seems superfluous. Poor people make stupid, self-destructive choices that flout norms all the time. Do we really need a theory to explain why rich people sometimes do as well?
Jay's comments are generally pretty clear to me. It's not like he's making allusions to classical Sanskrit poetry. I don't know why you guys have so much trouble with them. But in this case I was responding to a comment by Lee.
An LLM predicts the most likely next token, given the previous n tokens of context (initially the prompt, and then the text it already generated) and based on what it saw in the training data. It's not that it chooses to make up fake citations or factoids. It's just that it gets to a point in the generated text where a citation seems likely to occur, and then it chains together a plausible-sounding citation given the context.
I've actually gotten ChatGPT to give valid citations in the past, e.g. here's the first paragraph when I ask it about the paper that claimed vaccines cause autism:
The paper that initially claimed a link between vaccines and autism was published in 1998 by Andrew Wakefield and his colleagues. The paper, titled "Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children," was published in The Lancet, a reputable medical journal at the time.
Buuurn!
At first I thought this was gibberish, but that actually was the title. I think the key is that it has to be a paper that's talked about a lot in contexts similar to the one you provide.
I've gotten valid citations for important papers in more niche topics, as well, so I think it's the context-relevant prominence of the paper that matters, not the absolute prominence.
We do tolerate a lot of mediocrity and good-enough performance. You're making it sound like everyone who's not in the top 5% of performers lives in poverty, and obviously that's not true. It really doesn't take that much effort to have a decent middle-class life, or even upper-middle-class.
This is such a strawman that I'm not even sure what it's standing in for. What, in concrete terms, are you objecting to here?
Obviously there are people who argue that the top spots should go to the top performers, but that doesn't mean that everyone else is condemned to a crappy life. Why would the top spots not go to the top performers? What other criteria would you use? Lottery? Hereditary sinecure?
One of the quotes they're being dragged for is "How about someone who's not a Jew?" in reference to a job candidate.
Here's the thing, though: A lot of the people who are rightly condemning this would be wrongly defending "How about someone who's not a white man?" Discrimination against white people, especially men, is not only tolerated, but actively encouraged. It's done out in the open, not in email between two trusted colleagues.
Supposedly this is justified because people with poor critical thinking skills think that overrepresentation can only be due to privilege and underrepresentation can only be due to oppression.
But you know who's much, much more overrepresented in law than white people? Jewish people! If law firms should discriminate against gentile white candidates to make the legal profession look more like America (they shouldn't), then they should discriminate even more harshly against Jewish candidates (they shouldn't do this, either).
I'm not saying that they're actually super woke, of course, but this does highlight the intellectual inconsistencies and blatant hypocrisy of the charlatans preaching the systemic racism mythology. The continued success of market-dominant minorites undermines their narrative, so they quietly sweep it under the rug rather than do the work to bring their ideology into conformance with reality.
In US dollar terms (probably more relevant than PPP) the ratio of Mexico's gdp per capita to the United States' has been falling for the past decade, and is not significantly higher than in was in the early 70s.
Huh. I'm not really familiar with his work, but I had always assumed based on his reputation that he was a serious intellectual. But here he comes off as deeply unserious, and frankly a bit dim-witted.
You're off by an order of magnitude. GDP per capita is about $75k; 4% of that is $3k.
Of course, this is only explicitly means-tested programs. As Michael Cain points out, Social Security and Medicare are also anti-poverty programs, but since they're not means-tested, they're not counted. State and local spending aren't counted, either.
Also, about a sixth of GDP is depreciation, i.e. money we have to spend to replace aging and obsolete capital just to maintain current levels of productivity. So the percentage of net domestic product that gets spent on anti-poverty programs is about 20% higher than the percent of GDP that gets spent.
For the last 40 years or so, there's been a back-and-forth on tax policy where Republicans cut taxes on all income levels, followed by Democrats raising taxes only on the rich. As a result, effective federal tax rates have fallen dramatically over time for the lower brackets, while effective federal tax rates on the top 1% have gone up and down with no discernible long-run trend. The bottom three quintiles are now paying virtually no federal taxes other than payroll taxes, which on average aren't even enough to fund their expected retirement benefits.
along with the general strangulation of the Great Society in the crib
This never happened. From 1972 to 2012, means-tested federal spending increased tenfold in inflation-adjusted dollars (population increased 50% during this time), and went from 1% of GDP to nearly 4%.
I don't have more recent data on hand, but I guarantee you that it has increased further since then, possibly not as a percentage of GDP (as the economy was still in recovery in 2012), but certainly in terms of inflation-adjusted dollars per capita.
"Women and minorities hit hardest" is how the media say they care about an issue. When the media are behind an issue, government---especially Democratic government---is more likely to take action.
I could be wrong, but that doesn't sound right to me. I know that you can't deduct hobby expenses from your income from other sources like your full-time job, but surely you can use them to offset income from the hobby itself?
Maryland actually does quite badly when adjusting for student demographics, or at least did in 2015. It looks like the Urban Institute hasn't updated this for more recent years:
Or, rather, we didn’t attempt to solve it until the pandemic, where we discovered that pretty small payments could make a _huge_ difference in childhood poverty
That's because poverty is defined by having income over a specific threshold for your family size. Apparently many families with children and incomes below the poverty line had incomes that were just barely below the poverty line. With the additional handouts, they were then just barely above the poverty line. It's not like it moved families from deep poverty to the middle class. There was no transformative effect here---it was just the mechanical effect in which giving people money causes their incomes to be higher when you count the money you gave them.
there’s almost no evidence the problem has anything to do with ‘culture’ instead of just ‘literally not having any money or a hope for a future in which they might have money’.
The idea that racial gaps in outcomes are entirely mediated by poverty is a dogma with no basis in fact. It just isn't true. Look at the research instead of just assuming whatever facts are most convenient for your ideology. On crime and test scores in particular, white people from families with poverty-level incomes perform about as well as black people from families with low six-figure incomes.
Controlling for income does narrow the gaps significantly, but that doesn't demonstrate that income is a major causal factor. Income is to at least some extent acting as a proxy for heritable and culturally transmissible traits.
Furthermore, the intergenerational elasticity of income just isn't high enough to support the poverty hypothesis. Rank-rank intergenerational elasticity of permanent earning income is about 0.4, which means that parents with earnings at the 10th percentile (40 percentiles below the median) have, on average, children with earnings at the 34th percentile, only 16 (40 * 0.4) points below the median.
And this includes the influence of hereditary and cultural factors. The true causal effect of parental income is even smaller.
What this means is that when poverty is purely exogenous, attributable entirely to external causes like discrimination, we should see aggressive regression towards the mean once the exogenous factors are alleviated, with convergence occurring within a generation or two. We saw this with Asians. We saw it with Jews. We don't see it with black Americans.
The evidence strongly points to endogenous causes for the persistence of black-white achievement gaps. The only
Wealth, by the way, has even lower intergenerational elasticity, net of wealth attributable to children's earnings. "Generational wealth" is a meme, not a serious hypothesis. Inheritance explains at most about 10-15% of the black-white gap in median net worth; in fact, the gap barely shrinks at all when comparing white households with no inheritance to black households that have already received one. In reality, it's almost entirely attributable to current-generation earnings, marriage rates, and age (the median black American is several years younger than the median white American).
No, it would not, and the fact that you remembered it as such indicates that it went right over your head.
What you want Hanania to be saying is that the police should just go round up random black people and throw them in jail. This allows you to dismiss it as racist, and avoid grappling with the actual substance of the point.
What Hanania was actually saying is that in order to reduce crime, we need more cops on the beat in high-crime areas, and we need to arrest and incarcerate the people committing those crimes. Because of racial gaps in criminal offending, this will necessarily mean policing predominantly black areas more heavily and disproportionately incarcerating black people. He's being blunt---maybe even a bit trollish---but this is a fact that needs to be confronted. We can't solve the crime problem if we can't accept the disparate impact (weirdly, lefties much more sanguine about the disparate impact of crime on black people; I guess those aren't the #BlackLives that #Matter).
No, as much as you might like it to be, the disparate impact is not the point
On “Open Mic for the week of 6/26/2023”
Private housing completions have been steadily rising since 2010, but we still have a huge hole to fill:
https://fred.stlouisfed.org/series/COMPUTSA
Note that this is annualized monthly rate, so there were about 125k homes completed in May, not 1.5 million. And two-thirds of those units are single-family homes, so we need to be building much more multi-family housing.
"
I've been saying for years that abortion is good but Roe v Wade was a total hack job. The Constitution and rule of law matter more than any one decision.
I think about 1% of people in the US, tops, will acknowledge a distinction between a Constitutionally correct decision and their own personal policy preferences.
On “The Titan and the Comet”
I looked into this a bit more. Initially I naively assumed that an airplane would be pressurized to one atmosphere, but it turns out they're actually pressurized to about 0.73-0.8 atmospheres at high altitudes. The reason for this is that this reduces the pressure differential by about a third, which presumably makes a material difference in terms of fuselage weight.
"
I don't think this is even remotely comparable. The air pressure in an airplane is less than one atmosphere. Since there's no such thing as negative absolute pressure, there's less than a one-atmosphere pressure difference between the inside and outside. That's still quite a lot of pressure (7 PSI is about 1,000 pounds per square foot), but with the submarine, it's a difference of hundreds of atmospheres.
On “Asteroid City, In Which Wes Anderson Is Now Just Screwing With Us”
The real trick of The Grand Budapest Hotel was how the main character changed from South Asian to Mediterranean as he aged.
On “Open Mic for the week of 6/19/2023”
This seems superfluous. Poor people make stupid, self-destructive choices that flout norms all the time. Do we really need a theory to explain why rich people sometimes do as well?
On “The Good Place, Ethics 101, and Spoilers, Spoilers, Spoilers”
Existence is pretty good, but the portions are too small.
On “Thursday Throughput: Chat GPT Is Not Your Research Assistant”
On second thought, a planet that took 14 years to create may be much, much older than 14 years.
On “Open Mic for the week of 6/12/2023”
Jay's comments are generally pretty clear to me. It's not like he's making allusions to classical Sanskrit poetry. I don't know why you guys have so much trouble with them. But in this case I was responding to a comment by Lee.
On “Thursday Throughput: Chat GPT Is Not Your Research Assistant”
Can we keep the teenage-Earth creationism out of the science threads?
"
An LLM predicts the most likely next token, given the previous n tokens of context (initially the prompt, and then the text it already generated) and based on what it saw in the training data. It's not that it chooses to make up fake citations or factoids. It's just that it gets to a point in the generated text where a citation seems likely to occur, and then it chains together a plausible-sounding citation given the context.
I've actually gotten ChatGPT to give valid citations in the past, e.g. here's the first paragraph when I ask it about the paper that claimed vaccines cause autism:
Buuurn!
At first I thought this was gibberish, but that actually was the title. I think the key is that it has to be a paper that's talked about a lot in contexts similar to the one you provide.
I've gotten valid citations for important papers in more niche topics, as well, so I think it's the context-relevant prominence of the paper that matters, not the absolute prominence.
On “Open Mic for the week of 6/12/2023”
We do tolerate a lot of mediocrity and good-enough performance. You're making it sound like everyone who's not in the top 5% of performers lives in poverty, and obviously that's not true. It really doesn't take that much effort to have a decent middle-class life, or even upper-middle-class.
This is such a strawman that I'm not even sure what it's standing in for. What, in concrete terms, are you objecting to here?
Obviously there are people who argue that the top spots should go to the top performers, but that doesn't mean that everyone else is condemned to a crappy life. Why would the top spots not go to the top performers? What other criteria would you use? Lottery? Hereditary sinecure?
"
Are you sure you know what a Luddite is?
On “Open Mic for the week of 6/5/2023”
One of the quotes they're being dragged for is "How about someone who's not a Jew?" in reference to a job candidate.
Here's the thing, though: A lot of the people who are rightly condemning this would be wrongly defending "How about someone who's not a white man?" Discrimination against white people, especially men, is not only tolerated, but actively encouraged. It's done out in the open, not in email between two trusted colleagues.
Supposedly this is justified because people with poor critical thinking skills think that overrepresentation can only be due to privilege and underrepresentation can only be due to oppression.
But you know who's much, much more overrepresented in law than white people? Jewish people! If law firms should discriminate against gentile white candidates to make the legal profession look more like America (they shouldn't), then they should discriminate even more harshly against Jewish candidates (they shouldn't do this, either).
I'm not saying that they're actually super woke, of course, but this does highlight the intellectual inconsistencies and blatant hypocrisy of the charlatans preaching the systemic racism mythology. The continued success of market-dominant minorites undermines their narrative, so they quietly sweep it under the rug rather than do the work to bring their ideology into conformance with reality.
"
In US dollar terms (probably more relevant than PPP) the ratio of Mexico's gdp per capita to the United States' has been falling for the past decade, and is not significantly higher than in was in the early 70s.
https://fred.stlouisfed.org/graph/?g=15Whj
"
Huh. I'm not really familiar with his work, but I had always assumed based on his reputation that he was a serious intellectual. But here he comes off as deeply unserious, and frankly a bit dim-witted.
On “The Debt Limit Deal Has Been Cut”
You're off by an order of magnitude. GDP per capita is about $75k; 4% of that is $3k.
Of course, this is only explicitly means-tested programs. As Michael Cain points out, Social Security and Medicare are also anti-poverty programs, but since they're not means-tested, they're not counted. State and local spending aren't counted, either.
Also, about a sixth of GDP is depreciation, i.e. money we have to spend to replace aging and obsolete capital just to maintain current levels of productivity. So the percentage of net domestic product that gets spent on anti-poverty programs is about 20% higher than the percent of GDP that gets spent.
"
For the last 40 years or so, there's been a back-and-forth on tax policy where Republicans cut taxes on all income levels, followed by Democrats raising taxes only on the rich. As a result, effective federal tax rates have fallen dramatically over time for the lower brackets, while effective federal tax rates on the top 1% have gone up and down with no discernible long-run trend. The bottom three quintiles are now paying virtually no federal taxes other than payroll taxes, which on average aren't even enough to fund their expected retirement benefits.
https://www.taxpolicycenter.org/statistics/historical-average-federal-tax-rates-all-households
The reason Republicans keep cutting taxes on the rich is that Democrats keep jacking them back up.
"
This never happened. From 1972 to 2012, means-tested federal spending increased tenfold in inflation-adjusted dollars (population increased 50% during this time), and went from 1% of GDP to nearly 4%.
https://www.cbo.gov/sites/default/files/113th-congress-2013-2014/reports/43934-means-testedprogramsone-column0.pdf
I don't have more recent data on hand, but I guarantee you that it has increased further since then, possibly not as a percentage of GDP (as the economy was still in recovery in 2012), but certainly in terms of inflation-adjusted dollars per capita.
On “Open Mic for the week of 5/29/2023”
"Women and minorities hit hardest" is how the media say they care about an issue. When the media are behind an issue, government---especially Democratic government---is more likely to take action.
On “The Debt Limit Deal Has Been Cut”
I could be wrong, but that doesn't sound right to me. I know that you can't deduct hobby expenses from your income from other sources like your full-time job, but surely you can use them to offset income from the hobby itself?
On “Open Mic for the week of 5/22/2023”
Maryland actually does quite badly when adjusting for student demographics, or at least did in 2015. It looks like the Urban Institute hasn't updated this for more recent years:
https://www.urban.org/urban-wire/how-do-states-really-stack-2015-naep
"
Or, rather, we didn’t attempt to solve it until the pandemic, where we discovered that pretty small payments could make a _huge_ difference in childhood poverty
That's because poverty is defined by having income over a specific threshold for your family size. Apparently many families with children and incomes below the poverty line had incomes that were just barely below the poverty line. With the additional handouts, they were then just barely above the poverty line. It's not like it moved families from deep poverty to the middle class. There was no transformative effect here---it was just the mechanical effect in which giving people money causes their incomes to be higher when you count the money you gave them.
there’s almost no evidence the problem has anything to do with ‘culture’ instead of just ‘literally not having any money or a hope for a future in which they might have money’.
The idea that racial gaps in outcomes are entirely mediated by poverty is a dogma with no basis in fact. It just isn't true. Look at the research instead of just assuming whatever facts are most convenient for your ideology. On crime and test scores in particular, white people from families with poverty-level incomes perform about as well as black people from families with low six-figure incomes.
Controlling for income does narrow the gaps significantly, but that doesn't demonstrate that income is a major causal factor. Income is to at least some extent acting as a proxy for heritable and culturally transmissible traits.
Furthermore, the intergenerational elasticity of income just isn't high enough to support the poverty hypothesis. Rank-rank intergenerational elasticity of permanent earning income is about 0.4, which means that parents with earnings at the 10th percentile (40 percentiles below the median) have, on average, children with earnings at the 34th percentile, only 16 (40 * 0.4) points below the median.
And this includes the influence of hereditary and cultural factors. The true causal effect of parental income is even smaller.
What this means is that when poverty is purely exogenous, attributable entirely to external causes like discrimination, we should see aggressive regression towards the mean once the exogenous factors are alleviated, with convergence occurring within a generation or two. We saw this with Asians. We saw it with Jews. We don't see it with black Americans.
The evidence strongly points to endogenous causes for the persistence of black-white achievement gaps. The only
Wealth, by the way, has even lower intergenerational elasticity, net of wealth attributable to children's earnings. "Generational wealth" is a meme, not a serious hypothesis. Inheritance explains at most about 10-15% of the black-white gap in median net worth; in fact, the gap barely shrinks at all when comparing white households with no inheritance to black households that have already received one. In reality, it's almost entirely attributable to current-generation earnings, marriage rates, and age (the median black American is several years younger than the median white American).
"
No, it would not, and the fact that you remembered it as such indicates that it went right over your head.
What you want Hanania to be saying is that the police should just go round up random black people and throw them in jail. This allows you to dismiss it as racist, and avoid grappling with the actual substance of the point.
What Hanania was actually saying is that in order to reduce crime, we need more cops on the beat in high-crime areas, and we need to arrest and incarcerate the people committing those crimes. Because of racial gaps in criminal offending, this will necessarily mean policing predominantly black areas more heavily and disproportionately incarcerating black people. He's being blunt---maybe even a bit trollish---but this is a fact that needs to be confronted. We can't solve the crime problem if we can't accept the disparate impact (weirdly, lefties much more sanguine about the disparate impact of crime on black people; I guess those aren't the #BlackLives that #Matter).
No, as much as you might like it to be, the disparate impact is not the point
"
Richard Hanania has some screenshots of the Karens who demanded to speak to Lee's manager:
https://twitter.com/RichardHanania/status/1660295715791446016
*Comment archive for non-registered commenters assembled by email address as provided.