Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Cash Aid to Poor Mothers Increases Brain Activity in Babies, Study Finds

nytimes.com

Post reply on HN
Print view View original

11–15 of 15 posts

Re: Cash Aid to Poor Mothers Increases Brain Activity in Babies, Study Finds

#11

Post by SamReidHughes »

cacioepepe wrote: post #10

In the supplement what's buried is that there were no effects on measured cognitive development in the overall sample. Meaning that in the subsample of parents who received cash disbursements and did EEG, they found differences in EEG, but that the outcome of importance -- cognitive performance -- did not differ. Also, as far as I can tell, the EEG measures did not relate to cognitive performance, which calls into qu…

Even for EEG the p-values were 0.02, 0.04, 0.07, and are not significant after applying their preregistered multiple comparison adjustment.

IMO a more interesting outcome to measure is the quantity of bad outcomes. Assuming that almost all poor mothers take care of their kids, and I'd think $350/month would help the worst-off kids of low income mothers a lot more than the average kid. It would also be nice to see how "low income" is defined for this study.

Re: Cash Aid to Poor Mothers Increases Brain Activity in Babies, Study Finds

#12

Post by beerandt »

SamReidHughes wrote: post #11
cacioepepe wrote: post #10

In the supplement what's buried is that there were no effects on measured cognitive development in the overall sample. Meaning that in the subsample of parents who received cash disbursements and did EEG, they found differences in EEG, but that the outcome of importance -- cognitive performance -- did not differ. Also, as far as I can tell, the EEG measures did not relate to cognitive performance, which calls into qu…

Even for EEG the p-values were 0.02, 0.04, 0.07, and are not significant after applying their preregistered multiple comparison adjustment. IMO a more interesting outcome to measure is the quantity of bad outcomes. Assuming that almost all poor mothers take care of their kids, and I'd think $350/month would help the worst-off kids of low income mothers a lot more than the average kid. It would also be nice to see how…

I think it's actually a pretty bad experiment, even as a concept. And because the same behavior is not applicable at scale, which is what these types experiments seem destined to promote as policy, whether intentionally or not.

It's attributing a gain in purchasing power to an increased absolute dollar amount, when the extreme top and bottom segments of market participants are affected more by their position relative to everyone else participating in that market.

In other words, as we've sort of seen with covid, the markets where everyone at the bottom gets aid in the same/ similar amounts are also the markets that see the greatest shortages and price increases, leaving those at the bottom in the same relative position as before, despite having more money.

It's a similar phenomenon to how anti price-gouging laws will often make shortages worse or result in higher demand or both. Or will manifest a black market of profiteering arbiters.

(There's also effects due to lack of overall supply and lack of elasticity or a slow market response time. Eg, in rental markets with the most protective eviction moratoriums, landlords were much more likely to require extreme vetting of new tenants as well as large deposits or even prepayment. All things that disproportionately hurt those who are at the bottom end of the market.)

>It would also be nice to see how "low income" is defined for this study.

Yeah, especially if it's based on a family size/status-adjusted percentile-type metric, which would make everything above even more pronounced and potentially inflationary.

Re: Cash Aid to Poor Mothers Increases Brain Activity in Babies, Study Finds

#13

Post by SamReidHughes »

beerandt wrote: post #12
SamReidHughes wrote: post #11

Earlier quoted context omitted.

Even for EEG the p-values were 0.02, 0.04, 0.07, and are not significant after applying their preregistered multiple comparison adjustment. IMO a more interesting outcome to measure is the quantity of bad outcomes. Assuming that almost all poor mothers take care of their kids, and I'd think $350/month would help the worst-off kids of low income mothers a lot more than the average kid. It would also be nice to see how…

I think it's actually a pretty bad experiment, even as a concept. And because the same behavior is not applicable at scale, which is what these types experiments seem destined to promote as policy, whether intentionally or not. It's attributing a gain in purchasing power to an increased absolute dollar amount, when the extreme top and bottom segments of market participants are affected more by their position relative…

$3K when a baby comes out is different than UBI, it wouldn't get passed through to landlords or to the extent it does it'd redistribute from poor people without a baby to the one with a new baby.

I think this experiment or concept could succeed if there weren't _already_ welfare programs, including charities, and if there was a large ultra poor population, _and_ the study focused on the ultra poor, not "low income," and it focused on more relevant child performance metrics instead of EEG, and especially if the outcome measured was how many kids die of starvation or sewer parasites or such, worst case outcomes that an extra $3K to the right person could avoid, instead of trying to catch a <.2 SD boost in mean outcome on top of a big diminishing returns curve.

Re: Cash Aid to Poor Mothers Increases Brain Activity in Babies, Study Finds

#15

Post by beerandt »

SamReidHughes wrote: post #13
beerandt wrote: post #12

Earlier quoted context omitted.

I think it's actually a pretty bad experiment, even as a concept. And because the same behavior is not applicable at scale, which is what these types experiments seem destined to promote as policy, whether intentionally or not. It's attributing a gain in purchasing power to an increased absolute dollar amount, when the extreme top and bottom segments of market participants are affected more by their position relative…

$3K when a baby comes out is different than UBI, it wouldn't get passed through to landlords or to the extent it does it'd redistribute from poor people without a baby to the one with a new baby. I think this experiment or concept could succeed if there weren't _already_ welfare programs, including charities, and if there was a large ultra poor population, _and_ the study focused on the ultra poor, not "low income,"…

I'm admittedly looking at it from a different angle, but I think it's still relevant. Even if you're focusing on performance metrics and I'm looking at evonomics.

The big 5 marketing target events: graduate, marry, buy house, get pregnant, retire. All events proven to change spending habits.

Having a baby often means moving into a bigger space.

The market isn't "housing", so it's not appropriate to consider differences in 3k between someone having a child vs someone not having a child.

It's "housing suitable for singles/couples expecting their first baby", etc.

So you are differentiating (sub)markets, but not necessarily the individuals within.

Similar to how brides planning a wedding aren't competing in a national market, but find themselves competing with a dozen or so other couples all essentially trying to plan the same wedding.

But with the extra money at childbirth, the more likely sink for the competing funds (in the US) is going to be daycare and then preschool.

That is, if it's not finding a bigger home that's more child-friendly. Which is similar to the wedding thing. Available properties near work, within budget sets a sub-market. Then it's just bidding for who gets the nicer home on the park with the better school.

You already see this with hud section 8, where apartment complexes especially set their lowest price at the maximum govt reimbursement rate. Similar situation in college towns where off campus apartments and neighborhoods fluctuate based on dorm rates and allowable student loan amounts.

And slightly nicer but older single family homes set their prices slightly to moderately higher, because as a sfh landlord, you don't need 50 renters, just 1.

But the situation is similar, where a large number of individuals in a particular submarket are all being provided with equal or at least very similar "allowances", and the market finds both monetary and non-monetary ways to adjust and differentiate individuals within.

Post reply on HN