War­ren’s wage claim about the bot­tom 90 per­cent is based on com­pre­hen­sive, yet lim­ited, data

The Washington Post Sunday - - FROM PAGE ONE - The Fact Checker glenn.kessler@wash­post.com

“From 1935 to 1980, we’re com­ing out of the Great De­pres­sion. What do we do? We do two things. We put a cop on the beat on Wall Street, and we started in­vest­ing in our fu­ture to­gether. . . . And we build Amer­ica’s great mid­dle class: the 90 per­cent, ev­ery­body out­side the top 10 per­cent. The 90 per­cent get 70 per­cent of all wage growth in this coun­try. . . . Trickle­down eco­nom­ics hits in the 1980s, and you just watch this re­verse. So from 1980 to 2012, that’s the lat­est year for which we have data, the 90 per­cent, ev­ery­body out­side the 10 per­cent, do you know how much they got? They got zero per­cent of in­come growth.”

—Sen. El­iz­a­beth War­ren (DMass.), re­marks in var­i­ous venues.

War­ren has re­peated a ver­sion of th­ese re­marks in a va­ri­ety of high-pro­file set­tings, in­clud­ing on the “Late Show with Stephen Col­bert.” Read­ers have asked whether her frame of ref­er­ence is cor­rect. Did the bot­tom 90 per­cent of in­come earn­ers in the United States get 70 per­cent of the wage growth from 1935 to 1980 — and then zero per­cent of the growth af­ter 1980?

This is a very in­ter­est­ing but highly tech­ni­cal is­sue. War­ren pitches it in a fairly par­ti­san man­ner — every­thing turned south af­ter “trickle-down eco­nom­ics” (i.e., Pres­i­dent Ron­ald Rea­gan) was in­tro­duced. She also of­fers some­what sim­plis­tic rea­sons (such as the cre­ation of the Se­cu­ri­ties and Ex­change Com­mis­sion) for the shift in in­come growth, though her staff says she was not in­tend­ing to ex­clude other fac­tors.

Where do th­ese num­bers come from?

The Facts

War­ren de­rived her num­bers from data col­lected by two well­known econ­o­mists, Em­manuel Saez of the Univer­sity of Cal­i­for­nia at Berke­ley and Thomas Piketty of the Paris School of Eco­nom­ics. Most econ­o­mists agree that the two men have as­sem­bled a com­pre­hen­sive data set that helps ex­am­ine the dis­tri­bu­tion of mar­ket in­come over time.

Here are the fac­tors cap­tured in the data: wages, net self­em­ploy­ment in­come re­ported on in­come tax re­turns, tax­able in­ter­est and div­i­dends, rental in­come, and so forth.

But there are lim­i­ta­tions. So­cial Se­cu­rity and un­em­ploy­ment ben­e­fits are not counted. Nei­ther are gov­ern­ment trans­fer pay­ments such as food stamps and vet­er­ans ben­e­fits. The data also does not in­clude non­cash items such as em­ployer-pro­vided health in­sur­ance. Last, the im­pact of in­come and pay­roll taxes are not part of the cal­cu­la­tions, even though the rich pay most of the taxes — and some lower-in­come work­ers re­ceive pay­ments such as the earned-in­come tax credit that off­set all of their fed­eral pay­roll and in­come taxes.

In other words, the num­bers show part of the in­come pic­ture— es­sen­tially the wages. They do not nec­es­sar­ily show every­thing, be it food stamps or com­pany-pro­vided health in­sur­ance, that Amer­i­cans re­ceive to help them pay for what they con­sume.

“The in­come items Piketty and Saez omit tend to equal­ize in­comes of rich and poor,” said Gary Burt­less, an econ­o­mist at the Brook­ings In­sti­tu­tion who spe­cial­izes in in­come dis­tri­bu­tion. “More im­por­tantly, an in­creas­ing per­cent­age of the gross in­comes re­ceived by Amer­i­cans is ex­cluded from the in­come mea­sure tab­u­lated by Piketty and Saez.”

In essence, War­ren is just fo­cus­ing on the in­come that is re­ported on a tax re­turn — not all of the items that help fam­i­lies pay for what they con­sume.

There is a bit of an irony here. By fo­cus­ing on wage in­come, War­ren is not count­ing the im­pact of pro­grams of­ten cel­e­brated by Democrats as suc­cess sto­ries. In­deed, the of­fi­cial poverty rate also does not count the im­pact of those pro­grams, which is why the Cen­sus Bureau in re­cent years has de­vel­oped an al­ter­na­tive ver­sion that does in­clude it.

But a fuller set of data goes back only to the early 1960s, mak­ing the lengthy com­par­i­son high­lighted by War­ren all but im­pos­si­ble. But the in­come di­vide high­lighted by War­ren shrinks when the miss­ing data is part of the equa­tion.

The Con­gres­sional Bud­get Of­fice in 2014 pro­duced a re­port that fo­cused on the af­ter-tax re­source flow avail­able to a fam­ily to pay for its con­sump­tion. Fam­i­lies in the top 1 per­cent saw their af­ter-tax in­come triple from 1970 to 2011, but other groups also saw a siz­able im­prove­ment in house­hold in­comes.

An­other method, used by Cor­nell Univer­sity pro­fes­sor Richard V. Burkhauser with col­leagues, looks at “post-tax, post-trans­fer, size-ad­justed house­hold in­come in­clud­ing the ex-ante value of in-kind health in­sur­ance ben­e­fits.” In par­tic­u­lar, his ap­proach in­cludes cap­i­tal gains, which has been crit­i­cized by some econ­o­mists be­cause the volatil­ity of the stock and hous­ing mar­kets means the re­sults can change sig­nif­i­cantly de­pend­ing on the base year cho­sen.

In any case, his re­search reaches the con­tro­ver­sial con­clu­sion that once gov­ern­ment taxes and trans­fers are in­cluded and you ac­count for changes in house­hold size, the rich did get richer, but so did ev­ery­one else.

“The great ad­van­tage of tax record data is that it goes back to 1913,” Burkhauser said. “That is what is so spe­cial about the Saez num­bers. But they can eas­ily be mis­used by those who con­fuse the mar­ket in­come to tax units with the full af­ter-tax in­come of house­holds. El­iz­a­beth War­ren is one of those peo­ple.”

Burkhauser noted that the in­tro­duc­tion of So­cial Se­cu­rity in the 1930s meant that many older peo­ple stopped work­ing — and thus neg­a­tively af­fected the “mar­ket in­come” of older peo­ple.

Saez ac­knowl­edged the lim­i­ta­tions of his data and said that he and Piketty are try­ing to im­prove it. “Both pre-tax and post-tax are in­ter­est­ing pre­cisely to as­sess how gov­ern­ment pro­grams help,” he said. “We are go­ing to pro­duce both in the (hope­fully) near fu­ture.”

But Saez also said he agreed with the broad thrust of War­ren’s state­ment that the top 10 per­cent is sud­denly do­ing bet­ter.

Lacey Rose, a spokes­woman for War­ren, re­sponded that mar­ket in­come “paints a bet­ter pic­ture of the plight of mid­dle­class fam­i­lies than look­ing at an in­come mea­sure that in­cludes trans­fer pay­ments.” Rose noted that trans­fers are illiq­uid and not fun­gi­ble —hous­ing vouch­ers can­not be used to fix a car—and that re­fund­able tax cred­its such as the earned-in­come tax credit are er­rat­i­cally timed.

“The low in­come is a fact; trans­fer pro­grams are a pol­icy re­sponse to that fact,” she said. “They don’t negate the ex­is­tence of the fact.”

The Pinoc­chio Test

We dis­like leav­ing read­ers with an eco­nomic mud­dle. But this is a clas­sic case in which the out­come de­pends on the data se­lected. Should you count gov­ern­ment trans­fers or ben­e­fits such as health in­sur­ance? Do you count fam­i­lies or house­holds? Do you mea­sure be­fore or af­ter taxes?

Each of those choices can change the out­come in sig­nif­i­cant ways — and rea­son­able peo­ple can ar­gue about the im­por­tance of those choices. If one in­cludes health in­sur­ance, oth­ers might ar­gue that it’s silly to say a per­son’s wealth in­cludes pay­ments to med­i­cal pro­fes­sion­als.

Broadly speak­ing, most econ­o­mists would agree that the in­come growth has in­creas­ingly ben­e­fited the top 10 per­cent. War­ren’s rea­sons for that — and her spe­cific num­bers — are cer­tainly de­bat­able. She would do bet­ter to make clear to lis­ten­ers that she is cit­ing a par­tic­u­lar set of data that has cer­tain lim­i­ta­tions. Af­ter all, there is no cer­ti­tude in eco­nom­ics.

We can’t award her state­ment a Gep­petto Check­mark. But nei­ther does it quite ven­ture into Pinoc­chi­oland. So this state­ment will re­ceive no rat­ing.

Newspapers in English

Newspapers from USA

© PressReader. All rights reserved.