Welfare Reform Killed People
A 20th anniversary reminder that an authentic randomized control trial proved beyond all conventional statistical significance levels and all reasonable doubt that welfare reform killed people.
I will just note a very crude calculation. The experimental point estimate is that Florida’s welfare reform caused death rates to increase by 16%. This is a huge gigantic immense estimated effect. A silly assumption is that the increase was proportionally the same at all ages. If that were true, welfare reform would reduce life expectacy by a factor of 1.16 or about 11 years.
The available evidence suggests that welfare reform is like smoking a pack of cigarettes a day or like morbid obesity.
A crude estimate of the number of US deaths caused by welfare reform dwarfs US deaths caused by al Qaeda, plus the decision to invade Iraq, plus murder.
Defenders of welfare reform must critique the study to which I link or argue that human lives don’t matter. There is no third choice.
update: well that was quick. Warren notes in comments that my very crude calculation differs from a calculation in the paper by a factor of 14.67 or 0.917 Posners. I will try to figure out why.
I note that my calculation is very simple and based on a clearly stated assumption which I didn’t claim was reasonable.
Obviously a very different result holds if it is assumed that welfare reform has no effect on people of ages other than those in the study. The point is that a 16% increase in mortality is huge.
You might want to read the article you cite: “This amounts to nine months of life expectancy lost between the ages of thirty and seventy for people in FTP.” – See more at: http://angrybearblog.strategydemo.com/2013/06/welfare-reform-kills.html#sthash.36y2CR4K.dpuf
Nine months. Not eleven years.
Warren,
That’s shortened lifespan, not additional deaths. In the reform vs. control group, by 2011, there were ~additional 7 deaths/1,000 people. And Florida was supposedly one of the more ‘humane’ welfare states.
So thousands per year die from welfare reform that otherwise wouldn’t have.
Take money from welfare for the pentagon’s trough.
US will still be there if they ever get it working.
F-35: $1500B to assure the Taliban cannot strafe US infantry.
The pentagon spends north of $500B a year for war making stuff; specified for lots of bells and whistles to reduce casualties in wars against lightly armed evil doers.
Sadly the trillions spent never meet specified performance so we don’t know how many US KIA are result of weapons failure, faulty leadership, or insane strategy to make honored dead to justify more honored dead with huge money going to war profiteers.
We can see that war profiteers own congress, so more money goes their way!
No one cares about US citizens’ lives, unless they are worth a lot of money.
Need to stay with the A10 and lose the F35
Ok Warren I have to ask. If it’s nine DAYS, is that the right policy? Why is taking food and shelter from people that shortens their lives by any measure a good value for taxpayers?
Most of us believe that the point of welfare is to manage people’s transition back into the tax paying workforce whenever possible. AFAIK the data shows that for the majority of recipients that is actually what occurs. Including before slick willie’s horrible “reforms”.
Messrs. Mad Biologist and Socialist,
Disputing the numbers (the assertion of eleven years of shortened life span vs. the actual nine months) is not disagreeing on policy.
If one wants to make an intelligent argument, he must read all of the article one is citing and not make up numbers wildly out of line with the numbers further down in the article.
I said nothing at all about the policy.
I did not make up numbers. I performed a simple calculation which I admitted was crude. Surely you understand the calculation. IF the effect on mortality was proportionally the same 16% for all ages, then life expectancy would be reduced by a factor of 1/1.16.
The authors made other assumptions. However, the calculations must be based on assumptions. It is just not true that their calculation is the one objectively correct calculation. Nor is it true that I made up my number.
On the other hand, there is no ambiguity about the fact that reduction of life expectancy is not the same thing as ““life expectancy lost *between the ages of thirty and seventy for people in FTP*.”
Your assertion is plainly false. the figure 9 months is not comparable to the figure 11 years.
My calculation follows from the assumption I made (which I did not assert was reasonable or useful for any purpose but illustrating magnitudes).
You do understand that a 16 % increase in mortality is huge don’t you ?
You do understand that years between 30 and 70 lost are not total years lost don’t you ?
I ask for information and expect answers.
Hmmmmm:
Lets update this a bit and bring it up to 2016. It is years when one looks at income and age. Lower income people die at a younger age and the differential has been growing. Here are the cliff notes at NYT http://www.nytimes.com/2016/02/13/health/disparity-in-life-spans-of-the-rich-and-the-poor-is-growing.html?_r=0 or you can read the Brookings study: http://www.brookings.edu/research/reports2/2016/02/life-expectancy-gaps-promise-social-security#recent/
“The poor are losing ground not only in income, but also in years of life, the most basic measure of well-being. In the early 1970s, a 60-year-old man in the top half of the earnings ladder could expect to live 1.2 years longer than a man of the same age in the bottom half, according to an analysis by the Social Security Administration. Fast-forward to 2001, and he could expect to live 5.8 years longer than his poorer counterpart.
New research released on Friday contains even more jarring numbers. Looking at the extreme ends of the income spectrum, economists at the Brookings Institution found that for men born in 1920, there was a six-year difference in life expectancy between the top 10 percent of earners and the bottom 10 percent. For men born in 1950, that difference had more than doubled, to 14 years.”
This is a little different than what Robert states; but, it lends support to the need for higher incomes which can be income distribution under healthcare, food stamps, housing support, retirement, etc.
Yes. I filed this article under “Why the whole country is way overdue for a raise (including the unemployed)”
Thanks Run, but I really have to stress that the claim that welfare reform killed is not based on indirect evidence. It is the result of a randomized controlled trial. The estimated effect was overwhelmingly significant.
Warren objects to a calculation I made (which followed given my clearly stated assumption and arithmetic). the basic assertion that welfare reform killed people is not at all related to the debate between Warren and me.
I think it is an established fact. I might add that googling I have found the abstract of another paper which looked at Connecicut and found further (by itself not significant) evidence that welfare reform killed people
(notice that a point estimate identical to the Florida point estimate would be insignificant).
http://www.ncbi.nlm.nih.gov/pubmed/23678929
Am J Public Health. 2014 Mar;104(3):534-8. doi: 10.2105/AJPH.2012.301072. Epub 2013 May 16.
Impact of welfare reform on mortality: an evaluation of the Connecticut jobs first program, a randomized controlled trial.
Wilde ET1, Rosen Z, Couch K, Muennig PA.
Author information
Abstract
OBJECTIVES:
We examined whether Jobs First, a multicenter randomized trial of a welfare reform program conducted in Connecticut, demonstrated increases in employment, income, and health insurance relative to traditional welfare (Aid to Families with Dependent Children). We also investigated if higher earnings and employment improved mortality of the participants.
METHODS:
We revisited the Jobs First randomized trial, successfully linking 4612 participant identifiers to 15 years of prospective mortality follow-up data through 2010, producing 240 deaths. The analysis was powered to detect a 20% change in mortality hazards.
RESULTS:
Significant employment and income benefits were realized among Jobs First recipients relative to traditional welfare recipients, particularly for the most disadvantaged groups. However, although none of these reached statistical significance, all participants in Jobs First (overall, across centers, and all subgroups) experienced higher mortality hazards than traditional welfare recipients.”
The claim that welfare reform killed people is on a more solid scientific footing than the claim that gravity waves have been detected. it should not be controversial.
Robert:
I know my answer was a distraction to your calculations and I apologize.
Having watched my dad wither from working in construction breathing dust and finally retire at 62.5 years of age. He could no longer work and lived to 78. At 67, I am talking to a company in Korea to work for and represent them in the US where they are just starting out. I am the upper 10 percenter and there is a huge difference between my dad and I.
I agree with your words.
I have to repeat that the Florida program included extraordinary support for participants. At the time, I thought it was a good program.
Connecticut is obviously not the sort of state in which one would expect to find stingy welfare reform or effects as bad as the average state.
Welfare “reform” is a euphemism for welfare reduction. which can only have a net negative effect on that proportion of our society that rely upon it.
It was a redistribution of tax revenue from a large group poor people.to a small group of capital owners in the defense industry.
It was a windfall profit to employer’s by increasing the supply of low wage labor relative to demand for it and thereby suppressed wages at the lower end of the wage scales.
Why therefore wouldn’t it be expected to have a net negative effect on the poor to benefit the wealthy. Reductions in longevity is just one of the net negatives on the poor. Arguing over how many years longevity are lost is tantamount to arguing over when people should be put to death to maximize productive use of taxes. How many more prison and jail cells were filled? How many more homeless were created?
When we let the laissez-faire capitalists run the country what else can we expect under the euphemism “reform”?.
hey Run thanks for your informative and insightful comments. I apologize for giving you the impression you had some reason to apologize.
Now please please don’t apologize for causing me to apologize for causing you to apologize.
heh, heh, I just enjoy reading your posts and glad you are still a part of AB.
Robert:
There is a huge difference in how I am ending up in life, including my 2 brothers and a sister as compared. Where my dad never finished grade school and mom graduated high school, three of us have advanced degrees and one a BA. He was always proud of what we had accomplished. What could have been, did indeed take place for the 4 of us. My Masters courses were $300 – $350 a course in the late seventies into the early eighties as compared to ~$4,800 now. Multiple it by 18 courses and it is way out of the ball park in cost. It fries my butt when people such as Delisle, Akers, and Chingos of New America Foundation and Brooking say interest rates for loans do not matter. Cost needs to be controlled and interest rates kept low in education. It is not the Profs or the Adjuncts driving costs. I am on a roll here.
Somewhere and some how we need to get control of the educational process.
Never finished by going to U of C for my doctorate as my profs wanted me too. I had a family going and I was on the 3rd child. My wife wanted me home. Don’t regret it and mine have done well.
Thank you for your kind comments. I enjoy reading your material.
Mr. Waldmann,
Did you read the article, or just the abstract?
The Jobs First program had a huge positive impact on the most disadvantaged participants:
“The most socioeconomically disadvantaged
group experienced large gains. For instance, the
most disadvantaged participants benefited
from a 15% point increase in employment in
the first 2 years of the program, relative to 4%
points for the most advantaged group—that is,
those in the MDRC sample who were not
long-term welfare recipients, who had a general
equivalency diploma or high school diploma,
and had recent previous work experience. This
amounted to a full 79% increase in employment.
This group did not benefit from increases
in health insurance coverage.”
The abstract rightly point out that, “all participants in Jobs First (overall, across centers, and all subgroups) experienced higher mortality hazards than traditional welfare recipients,” it also makes clear that, “none of these reached statistical significance.”
For the most disadvantaged, and for the next-lowest group, the Hazard Ratio was 1.01, with a 95% confidence interval extending from 0.63 to 1.63.
Sounds like the program worked pretty well for them.
Where it really fell apart is for those with more than two children. They had a Hazard Ratio of 1.54 (although again not statistically significant, with a CI from 0.68 to 3.49).
So what is good for the goose may not be so good for the gander.
(The researchers also posit that the increased hazard my involve traffic accidents, since they are now getting to work. Getting to day care may add to the hazard of those with more children.)
“[Welfare reform] was a redistribution of tax revenue from a large group poor people.to a small group of capital owners in the defense industry.”
What time period are you talking about?
Defense spending has declined from about 50% of federal spending in the early 1960’s to 30% in the 1980’s, and now down to 20%.
Welfare spending has done exactly the opposite, going from 20% in the early 1960’s to 50% today.
While Warren seems to have disappeared, I’ll respond anyway. From the article:
“There were seventy-five deaths among the 1,611 participants in the experimental group in both locations in Escambia County and sixty seven
deaths among the 1,613 participants in the control group. After clustering on location and adding the control measures described above, we found that participants in the experimental group had a 16 percent higher mortality rate than members of the control group (hazard ratio: 1.16; 95% confidence interval: 1.14, 1.19; p < 0:01). This amounts to nine months of life expectancy lost between the ages of thirty and seventy for people in http://FTP."
This was a three-year trial (after which everyone received 'bad' welfare), and it resulted in seven additional deaths/1,000 recipients (another group had higher mortality). Given that millions of people are welfare recipients and that there is turnover in the welfare population (i.e., it's not the same millions for the last twenty years), it's not hyperbole to assume these policies result in thousands of needless deaths/year (which is what I wrote).
Hi Mike…welcome to comments and thanks for the links.
I was not disputing your bit, Mike. That’s why I addressed my comment to Mr. Waldmann. Not all welfare reform schemes are the same, and they will have different results.
I am not going to comment on the debate here between various parties about the article in question, nor do I have any idea how many have died due to welfare “reform,” although I have no doubt many have.
Rather I wanted to point out how useless our current TANF welfare system is when things get bad. This happened, of course, during the Great Recession of recent years. How much support did TANF provide to those large numbers of people thrown into poverty? Essentially zero. There was only a barely noticeable increase in TANF spending. It did nearly nothing. Most of the burden fell on such programs as SNAP and so on, food stamps, with, of course, various politicians getting all worked up and bent out about the increase in spending on those programs.
Warren,
In fact the reduction in welfare benefits was severe, dropping by nearly 10% since Clinton’s Welfare reduction began through 2006 (in constant 2006 dollars). It’s now even lower than it’s been since the early 1960’s.
In fact the percentage of recipients as a percent of the total population in 2005/06 was 1.6% while it was 5% in 1971/72 and peaked at 5.4% in 1993.
In 2006 there were less than half as many recipients in absolute numbers than there were in 1971 (10,050, v 4,550)
Ref:US Dept of Health and Human Services, 2008 Annual Report to Congress, Table TANF 2 in Appendix A-9 at https://aspe.hhs.gov/sites/default/files/pdf/53021/report.pdf
I’m referring to Welfare… you’re citing total social spending on Medicare, SS, and Welfare.
You’re right, Longtooth. Looks like the “other services” are squeezing things pretty hard in TANF: https://www.cbo.gov/publication/49887
Medicaid has gone up considerably, though.
Warren, your link has more recent information, and is appreciated. Thank-you. It shows an even far greater relative reduction in Welfare Payments than the link I had available.
As your link shows in Figure 5 (Recurring Cash Assistance in TANF & AFDC), both the number of recipients has been dramatically reduced (-61% since POWORA (Clinton’s Welfare reduction act), but so have the payments made to those that still receive it.. currently (2013) down by -24%.. In rough terms then the composite benefits (reduction in number of recipients x reduction in payments to them) is on the order of 30% of what the were when the act was implemented. That’s a 70% reduction!!!
% Difference relative to PRWORA Date (3rd column)
Average Monthly Benefit
(2013 dollars; right axis)”
1994 540 7.5%
1995 529 5.2%
1996 516 2.6%
1997 490 -2.6%
1998 486 -3.3%
1999 480 -4.5%
2000 456 -9.2%
2001 448 -10.8%
2002 448 -10.9%
2003 436 -13.3%
2004 435 -13.6%
2005 435 -13.5%
2006 424 -15.6%
2007 421 -16.4%
2008 413 -17.8%
2009 419 -16.7%
2010 415 -17.4%
2011 402 -20.1%
2012 382 -24.0%
It appears that the original article has been retracted. There is an Errata published and I quote: “Below the authors present the results of additional analysis they conducted to avoid the problems noted above for cluster-robust variance estimators with a small number of clusters. In the new analysis, the authors did not find a significant difference in mortality between the experimental and control groups.”
Welcome to AB. To prevent spammers and what not, we approve the first post.