A signaling theory of class x politics interaction

The me­dia, most re­cently The Economist and Scien­tific Amer­i­can, have been pub­li­ciz­ing a sur­pris­ing statis­ti­cal find­ing: in the cur­rent eco­nomic cli­mate, when more Amer­i­cans than ever are poor, sup­port for poli­cies that re­dis­tribute wealth to the poor are at their low­est lev­els ever. This new-found an­tipa­thy to­wards aid to the poor con­cen­trates in peo­ple who are near but not yet on the low­est rung of the so­cial lad­der. The Economist adds some re­lated statis­tics: those who earn slightly more than the min­i­mum wage are most against rais­ing the min­i­mum wage, and sup­port for welfare in an area de­creases as the per­centage of welfare re­cip­i­ents in the area rises.

Both ar­ti­cles ex­plain the para­dox­i­cal find­ings by ap­peal­ing to some­thing called “last place aver­sion”, an ob­served ten­dency for peo­ple to over­value not be­ing in last place. For ex­am­ple, in lab­o­ra­tory ex­per­i­ments where ev­ery­one gets ran­domly de­ter­mined amounts of money, most peo­ple are will­ing to help those with less money than them­selves gain cash—ex­cept the per­son with the sec­ond to low­est amount of money, who tends to try to thwart the per­son in last place even if it means en­rich­ing those who already have the most.

”Last place aver­sion” is in­ter­est­ing, and cer­tainly de­serves at least a foot­note in the cat­a­logue of cog­ni­tive bi­ases and heuris­tics, but I find it an un­satis­fy­ing ex­pla­na­tion for the ob­ser­va­tions about US at­ti­tudes to­ward wealth re­dis­tri­bu­tion. For one thing, the en­tire point of last place aver­sion is that it only af­fects those in last place, but in a mas­sive coun­try like the United States, ev­ery­one can find some­one worse off than them­selves (with one ex­cep­tion). For an­other, re­dis­tribu­tive poli­cies usu­ally stop short of mak­ing those who need gov­ern­ment hand­outs wealthier than those who do not; sub­si­diz­ing more home­less shelters doesn’t risk giv­ing the home­less a nicer house than your own. Fi­nally, many of the poli­cies peo­ple op­pose, like tax­ing the rich, don’t di­rectly trans­late to helping those in last place.

I pro­pose a differ­ent mechanism, one based on … wait for it … sig­nal­ing.

In a pre­vi­ous post, I dis­cussed multi-level sig­nal­ing and counter-sig­nal­ing, where each level tries to differ­en­ti­ate it­self from the level be­neath it. For ex­am­ple, the nou­veau riche differ­en­ti­ate them­selves from the mid­dle class by buy­ing os­ten­ta­tious bling, and the no­bil­ity (who are at no risk of be­ing mis­taken for the mid­dle class) differ­en­ti­ate them­selves from the nou­veau riche by not buy­ing os­ten­ta­tious bling.

The very poor have one strong in­cen­tive to sup­port re­dis­tri­bu­tion of wealth: they need the money. They also have a sec­ond, sub­tler in­cen­tive: most re­dis­tribu­tive poli­cies come pack­aged with a philos­o­phy that the poor are not per­son­ally re­spon­si­ble for the poverty, but are at least par­tially the vic­tims of the rest of so­ciety. There­fore, these poli­cies in­flate both their pock­et­book and their ego.

The lower mid­dle class gain what sta­tus they have by not be­ing the very poor; effec­tive sta­tus sig­nal­ing for a lower mid­dle class per­son is that which proves that she is cer­tainly not poor. One effec­tive method is to hold opinions con­trary to those of the poor: that re­dis­tri­bu­tion of wealth is evil and that the poor de­serve their poverty. This ide­ol­ogy cel­e­brates the su­pe­ri­or­ity of the lower mid­dle class over the poor by em­pha­siz­ing the biggest differ­ence be­tween the lower mid­dle class and the very poor: self-re­li­ance. By as­sert­ing this ide­ol­ogy, a lower mid­dle class per­son can prove her lower mid­dle class sta­tus.

The up­per mid­dle class gain what sta­tus they have by not be­ing the lower mid­dle class; effec­tive sta­tus sig­nal­ing for an up­per mid­dle class per­son is that which proves that she is cer­tainly not lower mid­dle class. One effec­tive way is to hold opinions con­trary to those of the lower mid­dle class: that re­ally the poor and lower mid­dle class are the same sort of peo­ple, but some of them got lucky and some of them got un­lucky. The only peo­ple who can com­fortably say “Deep down there’s re­ally no differ­ence be­tween my­self and a poor per­son” are peo­ple con­fi­dent that no one will ac­tu­ally mis­take them for a poor per­son af­ter they say this.

As a thought ex­per­i­ment, imag­ine your re­ac­tions to the fol­low­ing figures:

1. A bearded griz­zled man in ripped jeans, smelling slightly of al­co­hol, rant­ing about how the gov­ern­ment needs to give more free benefits to the poor.

2. A bearded griz­zled man in ripped jeans, smelling slightly of al­co­hol, rant­ing about how the poor are lazy and he worked hard to get where he is to­day.

3. A well-dressed, stylish man in a busi­ness suit, rant­ing about how the gov­ern­ment needs to give more free benefits to the poor.

4. A well-dressed, stylish man in a busi­ness suit, rant­ing about how the poor are lazy and he worked hard to get where he is to­day.

My gut re­ac­tions are (1, lazy guy who wants free money) (2, hon­or­able work­ing class salt-of-the-earth) (3, com­pas­sion­ate guy with good in­ten­tions) (4, in­sen­si­tive guy who doesn’t re­al­ize his priv­ilege). If these are rel­a­tively com­mon re­ac­tions, these would suffice to ex­plain the sig­nal­ing pat­terns in these de­mo­graph­ics.

If this were true, it would ex­plain the un­usual trends cited in the first para­graph. An area where welfare be­came more com­mon would see sup­port for welfare drop, as it be­came more and more nec­es­sary for peo­ple to sig­nal that they them­selves were not welfare re­cip­i­ents. Sup­port for min­i­mum wage would be low­est among peo­ple who earn just slightly more than min­i­mum wage, and who need to sig­nal that they are not min­i­mum wage earn­ers. And since up­per mid­dle class peo­ple tend to fa­vor re­dis­tri­bu­tion as a sta­tus sig­nal and lower mid­dle class peo­ple tend to op­pose it, a re­ces­sion that drives more peo­ple into the lower mid­dle class would cause a drop in sup­port for re­dis­tribu­tive poli­cies.