Se­na­tor’s wel­fare state­ment doesn’t add up

The Washington Post Sunday - - POLITICS & THE NATION - kess­lerg@wash­ To read pre­vi­ous Fact Checker col­umns, go to wash­ing­ton­ factchecker.

“Con­verted to cash, we spend enough on fed­eral wel­fare to mail ev­ery house­hold liv­ing be­neath the poverty line a check for $60,000 each year.”

— Sen. Jeff Ses­sions (R-Ala.),

Bud­get Com­mit­tee hear­ing, Feb. 13

Ses­sions’s state­ment is de­rived from his cal­cu­la­tion that what he de­scribes as some 80 “wel­fare pro­grams” make up the sin­gle­largest item in the fed­eral bud­get — larger than So­cial Se­cu­rity, Medi­care or De­fense De­part­ment spend­ing. Build­ing on that cal­cu­la­tion, he has also touted a chart that claims that to­tal wel­fare spend­ing “equates” to $168 per day for ev­ery house­hold in poverty, com­pared with me­dian in­come of $137 a day per house­hold — a 20 per­cent gap.

Ses­sions comes up with his list of 80 wel­fare pro­grams from an Oc­to­ber 2012 report by the Con­gres­sional Re­search Ser­vice, ti­tled “Spend­ing for Fed­eral Ben­e­fits and Ser­vices for Peo­ple with Low In­come.” The report con­cluded that the fed­eral government spent $746 bil­lion on such pro­grams in fis­cal 2010; state con­tri­bu­tions boost the fig­ure to about $1 tril­lion.

The CRS, how­ever, in­cluded a num­ber of caveats, warn­ing that “there is no sin­gle la­bel that best de­scribes all pro­grams in­cluded in this report.”

Ses­sions, in his state­ment, ig­nores such caveats and sim­ply de­clares all such spend­ing as “wel­fare.” As his spokesman, Stephen Miller, put it: “The caveats are frankly pre­pos­ter­ous.”

That’s a mat­ter of opin­ion, and Ses­sions cer­tainly is try­ing to shake up tra­di­tional no­tions of what con­sti­tutes wel­fare. But do his cal­cu­la­tions add up?

The Facts

We had long dis­cus­sions with Ses­sions’s staff mem­bers about the $168 fig­ure, which they con­tend is math­e­mat­i­cally cor­rect and in­tended to il­lu­mi­nate the money now spent by the fed­eral government on low-in­come peo­ple. But we have some se­ri­ous prob­lems with both the nu­mer­a­tor and de­nom­i­na­tor in this cal­cu­la­tion.

First, health-care spend­ing, es­pe­cially Med­i­caid, makes up nearly 50 per­cent of the to­tal fig­ure. But a ma­jor­ity of Med­i­caid spend­ing goes to the el­derly and dis­abled, not fam­i­lies with chil­dren.

More­over, health-care spend­ing is dif­fer­ent from food stamps or the earned in­come tax credit in that such aid gen­er­ally does not add to a fam­ily’s in­come level; in­stead, such as­sis­tance helps pays for bills that are the di­rect re­sult of how sick or dis­abled a pa­tient is. (That’s why so much of Med­i­caid spend­ing is di­rected to the el­derly in the last years of life.)

“Med­i­caid is a fed­eral pro­gram that is in­tended to pro­vide health-care ser­vices to peo­ple who are poor or near­poor,” re­sponded a Ses­sions aide. “It also pro­vides health ben­e­fits to sick peo­ple, but those peo­ple must first meet cer­tain in­come cri­te­ria (and in some cases an as­set test) in or­der to qual­ify for the ben­e­fit.”

Still, while the chart com­pares what Ses­sions terms wel­fare spend­ing to me­dian in­come, the Cen­sus Bureau does not in­clude health ben­e­fits (such as em­ployer-pro­vided health care) in that cal­cu­la­tion, even though such ben­e­fits ac­count for half of the wel­fare side of the ledger. So, he’s really com­par­ing ap­ples and or­anges.

Fi­nally, Ses­sions adds up many means-tested pro­grams, which are aimed at peo­ple with low in­comes, but then di­vides the fig­ure by the num­ber of peo­ple un­der the poverty level — even though mil­lions of peo­ple above the poverty level re­ceive th­ese ben­e­fits. That also sig­nif­i­cantly gooses up the fig­ure for spend­ing per house­hold.

At first glance, many might as­sume that Ses­sions is say­ing this is how much money is given to each house­hold un­der the poverty line. Ses­sions’s staff fiercely dis­puted that, not­ing that the chart says “equates,” which they say in­di­cates it is not claim­ing that this money is spent only on peo­ple be­low the poverty line.

But that im­pres­sion is cer­tainly left, par­tic­u­larly given the way Ses­sions dis­cussed the fig­ure at the hear­ing: “We spend enough on fed­eral wel­fare to mail ev­ery house­hold liv­ing be­neath the poverty line a check for $60,000 each year.”

In tes­ti­mony be­fore the House Bud­get Com­mit­tee in 2012, Robert Rec­tor of the Her­itage Foun­da­tion said that sim­ply di­vid­ing the means-tested spend­ing by the num­ber of the poor “can be mis­lead­ing be­cause many per­sons with in­comes above the of­fi­cial poverty lev­els also re­ceive means-tested aid.” In an in­ter­view, he said the GOP staff of the Se­nate Bud­get Com­mit­tee does good work in this area, but it would be bet­ter to di­vide the to­tal by the num­ber of peo­ple — more than 100 mil­lion — re­ceiv­ing meanstested ben­e­fits.

The Con­gres­sional Bud­get Of­fice, in a report this month, had an even more nu­anced ap­proach, es­ti­mat­ing the av­er­age fed­eral spend­ing per house­hold in 2006 for the 10 largest meanstested pro­grams (worth about 75 per­cent of Ses­sions’s to­tal) by dif­fer­ent in­come quin­tiles. For the low­est quin­tile, the fig­ure is nearly $9,000, af­ter ad­just­ing to 2012 dol­lars.

In both cases, when a more dis­creet ap­proach was taken, the head­line num­ber shrinks.

Miller said our con­clu­sions are “dis­ap­point­ingly an­ti­in­tel­lec­tual, ap­peal­ing to sen­ti­ment over rea­son.” He de­fended Sess­sions’s cal­cu­la­tions as “hon­est, ac­cu­rate and, most im­por­tantly, a con­struc­tive step to­wards help­ing those in need.”

The Pinoc­chio Test

Ses­sions — who says he went to col­lege on Pell grants, which he lists as a wel­fare pro­gram — clearly wants to jar some of the con­ven­tional wis­dom about an­tipoverty pro­grams. Call­ing at­ten­tion to their cost and ques­tion­ing their ef­fec­tive­ness are cer­tainly wor­thy en­deav­ors for a law­maker.

But he runs into trou­ble with his chart and the short­hand de­scrip­tion he gave at the hear­ing. The com­par­i­son to medium in­come is spe­cious, given that it does not in­clude health-care ben­e­fits, while the cost per house­hold ap­pears in­flated, given that the pro­grams he lists cover a whole range of pro­grams that as­sist more than just those in poverty.

We wa­vered back and forth be­tween two and three Pinoc­chios, but the ap­ples-and-or­anges com­par­i­son to me­dian in­come tipped it to three.

The Fact Checker


Newspapers in English

Newspapers from USA

© PressReader. All rights reserved.