One thing to note is that this was initially going to be a "History of welfare reform through AFDC and TANF and how it has led us to the CTC" piece.
It had to be written substantially earlier this year when that didn't happen π«
Theme #1: The hollow intellectual arguments for welfare reform in the 1990s.
A lot of the case for welfare reform was based on Charles Murray's "Losing Ground". But I think the book's arguments are very weak.
Murray claimed to be basing his critique of AFDC on economic analysis, but many of the cases he argued do not add up.
Some of the imagined examples he came up with to show that people on welfare would not work actually showed the opposite - that a person was better off working than receiving welfare.
Murray's mistake - a common one - is confusing income effects and substitution effects. The anti-work effects of AFDC were primarily due to the sharp phase-out, not because AFDC made it possible to get by without working.
Theme #2: The weak empirics of supposed welfare reform "successes".
Many people in the late 1990s crowed about TANF's success in decreasing the number of people receiving welfare, claiming that this indicated the success of the program.
But this misses that the main *mechanism* by which TANF worked was kicking people off of welfare.
You can also look at employment - but it's hard to see any large effects from TANF when you compare the US and Canada.
One explanation is that the 1990s were just a good business cycle, not that welfare reform increased work.
Theme #3: TANF is ineffective because it implicitly assumes the government is able to costlessly and accurately collect complicated information about a households member, income, etc.
Successfully operating a program like AFDC or TANF means collecting real-time information about households.
As James Scott points out in "Seeing Like A State" government is NOT GOOD AT THIS SORT OF THING.
The CTC is a much better approach.
Unlike AFDC, it doesn't have a steep phase-out that functions as an implicit marginal tax.
Unlike TANF, it requires limited data to administer (though that could still be improved!)
I have a new piece up at @NiskanenCenter, which digs into the data used in a recent @asdomash and @LHSummers essay, and suggests an alternative interpretation:
This is a good example of the precise problem @JerusalemDemsas was discussing.
The article (nydailynews.com/new-york/manhaβ¦) notes that more than 7,000 evictions were filed in March, but doesn't note that this is 40% of the pre-pandemic average.
βThis is the first study to show that money, in and of itself, has a causal impact on brain development." - @KimberlyGNoble
You can hook babies who got a UBI up to an Electroencephalogram, and the babies whose family got a $333/month UBI have a measurably different profile than the ones who got $20/month.
This is one of those policy changes that it feels like everyone is making into a bigger deal than it actually is.
It's a pilot program that is 1.) Limited to 3,000 people 2.) Only allows teenagers to drive a truck alone after driving 280 hours with an adult passenger.
There are 3.5 million truck drivers - another 3,000 isn't going to impact the supply chain much (nor drive wages down).
Especially with the requirement to have an adult passenger!
Lots of criticism seems focused on teenagers being more likely to get into car accidents than adults.
But this seems like something that is mostly a side effect of "teens doing fun stuff in cars their friends" moreso than "an unconditional increased probability of accidents".