How Interest Rates Redistribute Income

Fix, Blai Wo king Pape How In e es Ra es Redis ibu e Income P o ided in Coope a ion wi h: The Bichle & Ni zan A chi es Sugges ed Ci a ion: Fix, Blai (2023) : How In e es Ra es Redis ibu e Income, Economics om he Top Down, To on o, h ps://economics om he opdown.com/2023/04/16/how-in e es - a es- edis ibu e-income/ , h ps://bna chi es.yo ku.ca/779/ This Ve sion is a ailable a : h ps://hdl.handle.ne /10419/270866 S anda d-Nu zungsbedingungen: Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen Zwecken und zum P i a geb auch gespeiche und kopie we den. Sie dü en die Dokumen e nich ü ö en liche ode komme zielle Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich machen, e eiben ode ande wei ig nu zen. So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen (insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en, gel en abweichend on diesen Nu zungsbedingungen die in de do genann en Lizenz gewäh en Nu zungs ech e. Te ms o use: Documen s in EconS o may be sa ed and copied o you pe sonal and schola ly pu poses. You a e no o copy documen s o public o comme cial pu poses, o exhibi he documen s publicly, o make hem publicly a ailable on he in e ne , o o dis ibu e o o he wise use he documen s in public. I he documen s ha e been made a ailable unde an Open Con en Licence (especially C ea i e Commons Licences), you may exe cise u he usage igh s as speci ied in he indica ed licence. h ps://c ea i ecommons.o g/licenses/by/4.0/ How In e es Ra es Redis ibu e Income Blai Fix Ap il 16, 2023 [I]n e es a es a e, i s and o emos , a dis ibu ional a iable ha a ec s he income sha es o a ious social g oups. —Louis-Philippe Rochon & Ma k Se e ield (2007) When I ead abou mone a y policy, I ha e a ule o humb. E e y ime I see he ph ase in e es a e, I eplace i wi h he e m wage a e. Then I ask mysel whe he he discussion s ill makes sense. O en, i does no . The eason I make his subs i u ion is ha in concep ual e ms, he in e es a e and he wage a e a e simila : hey a e bo h a es o e u n. Wages a e he e u n on employmen . In e es a es a e he e u n on c edi . Now, he impo an hing abou a es o e u n is ha when we change hem, we a e oying wi h he dis ibu ion o income. Hike wages and we send mo e income o wo ke s. Hike he a e o in e es and we send mo e income o c edi o s. Su e, he speci ics o his edis ibu ion a e open o inqui y. Bu by de ini ion, a es o e u n a e ‘dis ibu ional a iables’ — hey de e mine how he income pie ge s di ied up. Back o my wo d subs i u ion. When i comes o wages, he issue o dis ibu- ion is ypically on and cen e . Tha ’s why alk o a minimum-wage hike p omp s businesses (and many economis s) o complain abou educed p o - i s. Bu when c edi o s hike he a e o in e es , alk o income dis ibu ion is cu iously absen . Ins ead, we ge a ba age o mac oeconomic ja gon — e ms like he ‘na u al a e o in e es ’ and he ‘non-accele a ing in la ion a e o unemploymen ’. Why he disc epancy? Blai Fix Economics om he Top Down One possibili y is ha economis s know some hing ha we don’ . Pe haps hey’ e looked a he e idence and concluded ha in e es a es ha e a ‘neu- al’ e ec on he dis ibu ion o income. Ano he possibili y is ha he mac oeconomic ja gon is mos ly a dis ac ion. In o he wo ds, like wages, he a e o in e es is a ‘dis ibu ional a iable’. Bu i ’s one ha mains eam economis s p e e o igno e. So which op ion is ue? In his pos , I le he e idence speak o i sel . Look- ing a c oss-coun y e idence, I ind ha in e es a es a e decidedly non- neu al. As in e es a es ise, h ee hings happen: 1. he in e es sha e o income inc eases; 2. he labo sha e o income dec eases; 3. income inequali y inc eases. In sho , he e idence sugges s ha in e es a es play a key ole in he game o class wa a e. And ha makes sense. In e es , a e all, is a a e o e u n. And when i comes o di ying up he income pie, a es o e u n a e always ze o sum. How economis s lea ned o igno e he dis ibu ion o income Be o e we jump in o he income-dis ibu ion da a, i ’s wo h e iewing some his o y. Among he e odox economis s, he dis ibu ion o income is a ho - bu on opic. Bu among mains eam economis s, i emains a seconda y conce n. Why? To unde s and his apa hy, we need o e ace he ajec o y o economic hough . When he s udy o poli ical economy go olling in he 19 h cen u y, he dis ibu ion o income was on and cen e . Fo ea ly poli ical economis s, wha ma e ed mos was class-based income. Why did land owne s, capi alis s, and wo ke s ecei e hei espec i e cu s? So c ucial was his ques ion ha Da id Rica do deemed i he ‘p inciple p oblem in poli ical economy’. As he 19 h cen u y played ou , hinke s like Ka l Ma x and Hen y Geo ge ecognized ha dis ibu ing income in ol ed ob ious con lic . Fo example, i Alice he wo ke and Bob he capi alis bo h wan a 60% cu o he pie, only one o hem can ge hei way. In o he wo ds, he ze o-sum na u e o 2 Blai Fix Economics om he Top Down class-based income necessi a es class wa a e. O a he , i necessi a es class wa a e, supposing ha he a ious classes wan mo e han hei cu en ake. En e John Ba es Cla k. Seeing he ins abili y in ol ed in class s uggle, Cla k wan ed o show ha con lic was unnecessa y. And he ound an ingenious way o do i . Wi h a ew swi assump ions, Cla k ‘p o ed’ ha in a compe i i e ma ke , each agen ge s back he amoun o weal h which ha agen c ea ed. Thus was bo n he heo y o ‘ma ginal p oduc i i y’, which ells e e yone ha hey ea n wha hey dese e. The message? In ma ke economies, class s uggle doesn’ exis . O cou se, Cla k’s heo y is based on assump ions ha a e ob iously alse. (Fo example, you mus assume ha socie y p oduces a single commodi y.) So he eal ques ion is why economis s se led on a heo y ha was clea ly w ong. He e’s my ake. Fi s , Cla k’s heo y o ma ginal p oduc i i y old powe ul people wha hey wan ed o hea — namely, ha he dis ibu ion o income is ‘jus ’. Second, Cla k’s app oached gelled nicely wi h he eme ging obsession wi h economic g ow h. In he mid 20 h cen u y, economis s ound ha i hey ea ed all o socie y like a single i m, hey could use a p oduc ion unc ion o ‘explain’ economic g ow h. 1 Impo an ly, his unc ion assumed no only ha each class ea ned hei ‘ma ginal p oduc ’, bu ha hei income sha es we e cons an . In o he wo ds, when modeling economic g ow h, economis s could ea he dis ibu ion o income as a non-issue. And so hey did. O cou se, he la ge backd op is ha ollowing Wo ld Wa II, humani y mas- si ely inc eased i s consump ion o esou ces — a pa e n ha had no hing o do wi h economis s’ models and e e y hing o do wi h he exploi a ion o ossil uels. Bu like e e yone else, economis s came along o he ide. And so hey go obsessed wi h economic g ow h, and lea ned o igno e he dis- ibu ion o income. Figu e 1illus a es his ideological shi as i is w i en in English wo d equency. 1 Ac ually, he his o y o economic g ow h heo y is jus as emba assing as he his o y o Cla k’s heo y o ma ginal p oduc i i y. Like Cla k, neoclassical g ow h heo is s ealized ha o make hei heo y wo k, hey had o assume ha socie y p oduced a single com- modi y. Wo se, c i ics la e showed ha economis s’ a o i e p oduc ion unc ion — he Cobb-Douglas unc ion — was a hinly eiled e sion o an accoun ing iden i y used o de- ine he na ional accoun s. In o he wo ds, he unc ion ‘wo ked’ because i was a au ology. Fo mo e de ails, see ‘Economic G ow h Theo y ... Bah Humbug!’. 3 Blai Fix Economics om he Top Down Figu e 1: Lea ning o igno e he dis ibu ion o income Using he Google English co pus, his igu e cha s he equency o he ph ases ‘economic g ow h’ and ‘dis ibu ion o income’. Un il he 1940s, he wo ph ases we e equally popula . Bu du ing he pos -WWII boom, economis s lea ned o lo e g ow h and igno e dis ibu ion. Sou ces and me hods Looking o he coming cen u y, my guess is ha economis s will e en ually ekindle hei in e es in he dis ibu ion o income ... bu only because hey will be o ced o. As he ossil- uel e a wanes, economic g ow h will become a elic o he pas . And so economis s will be d agged back in o he business o s udying how he income pie is di ided. In his pos , we’ll ge ahead o he cu e. The na ional income pie Jumping in o he s udy o income dis ibu ion, oday we ha e excellen da a on how he income pie is di ied up. And ha ’s somewha i onic. 4 Blai Fix Economics om he Top Down You see, he income da a la gely comes om he na ional accoun s, which we e designed as ools o measu ing economic ‘p oduc ion’. He e’s he i ony. Because he na ional accoun s a e based on mone a y ansac ions, hey a e adubious measu e o ‘p oduc ion’. Bu hey a e a good measu e o income. 2 To s udy he dis ibu ion o income, we do he opposi e o wha mos economis s do. Ins ead o using he na ional accoun s o look a he size o he income pie, we measu e he pie’s composi ion. Typically, ha means spli ing he pie in o i e ca ego ies o income: 1. employee compensa ion 2. p o i 3. p op ie o income 4. en 5. in e es To gi e you a sense o he size o each income ca ego y, Figu e 2shows he US income spli in 2021. Looking a hese income ca ego ies, we na u ally wan o know whe e hey come om. Economis s ha e an answe . They claim ha each class ep esen s a ‘ ac o o p oduc ion’. This ja gon e lec s economis s’ lo e o equa ing income wi h ‘p oduc ion’. I ad ise you o igno e i . The eali y is ha he a ious classes o income a e legal cons uc s. In o he wo ds, he law de ines di e en o ms o p ope y igh s. And om hese igh s s em di e en ca ego ies o income. Fo example, people who own co po a ions by de ini ion ea n ‘p o i ’. People who own uninco po a ed businesses by de ini ion ea n ‘p op ie o income’. People who own physical and/o in ellec ual p ope y ea n ‘ en ’. People who own deb ea n ‘in e es ’. And people who own no hing ea n ‘wages’. A e hese di e en ypes o p ope y igh s a bi a y? Yes. Bu in some sense, ules a e always a bi a y. Wha ma e s is ha hese p ope y ules a e hea ily en o ced, which means ha hey dic a e he e ms o class wa a e. 2 Income measu es om na ional accoun s s ill ha e p oblems. The main one is ha because he na ional accoun s os ensibly measu e ‘p oduc ion’, hey exclude income ha is deemed ‘unp oduc i e’. Fo his eason, he na ional accoun s exclude capi al gains, which a e an impo an sou ce o p ope y income. 5 Blai Fix Economics om he Top Down Figu e 2: The US income pie in 2021 This cha shows he composi ion o US na ional income in 2021. Sou ces and me hods The in e es sha e o income Speaking o class wa a e, le ’s alk abou he a e o in e es . As Rochon and Se e ield pu i , in e es a es a e p ima ily a “dis ibu ional a iable ha a ec s he income sha es o a ious social g oups.” Tha ’s a sub le way o saying ha in e es a es a e a ool o waging class wa . On his ac , he e idence is qui e clea . Bu be o e we ge o he da a, le ’s ame he ba le. In he na ional accoun s, in e es income is abula ed as ‘ne in e es ’. (The ‘ne ’ pa means ha when adding up in e es paymen s, s a is ical agencies sub ac in e es ecei ed om in e es paid.) Re u ning o Figu e 2, we can see ha ne in e es is cu en ly he smalles slice o he US income pie. Bu don’ be misled by his ecen da a. When we look a US his o y, we ind ha he in e es sha e o income is ex emely ola ile. Figu e 3 ells he s o y. He e, he ed cu e shows he in e es sha e o US na ional income. O e he las 120 yea s, i a ied om a low o 1% (in 1946) o a high o nea ly 10% (in 1982). Wha explains his a ia ion? Well, he ough and peak o he in e es sha e o income co espond o some impo an 6 Blai Fix Economics om he Top Down Figu e 3: In he Uni ed S a es, he in e es sha e o income ose and ell wi h he yield on bonds The ed cu e shows how he in e es sha e o US na ional income a ied o e he las cen u y. The blue cu e shows he co esponding change in bond yields, which p oxy he a e o in e es . E iden ly c edi o s’ cu o he pie is de e mined la gely by he a e o in e es . Sou ces and me hods geopoli ical e en s ( he end o WWII in 1945 and he s ag la ion c isis o he ea ly 1980s, espec i ely). Bu in mo e abs ac e ms, he in e es sha e o income is dic a ed by some hing simple : he a e o e u n on c edi . In Figu e 3, he blue cu e es i ies o he connec ion be ween he in e es sha e o income and he in e es a e. This cu e shows he long- e m his o y o US bond yields, which ise and all wi h he in e es sha e o income. And bond yields, i you’ e un amilia , a e a way o measu ing he e u n on c edi — o he wise known as he a e o in e es . 7 Blai Fix Economics om he Top Down Summa izing he e idence, i appea s ha Rochon and Se e ield a e co ec o call in e es a es a ‘dis ibu ion a iable’. We can hink o he a e o in e es as a dial o se ing he in e es sha e o income. Tha said, i ’s unwise o d aw sweeping conclusions om s udying a single coun y. So be o e we decla e in e es a es an income-sha e dial, le ’s widen ou ne . To do ha , we’d ideally look a da a o a huge sample o coun ies. Bu he p oblem wi h his app oach is ha income-sha e da a is no widely epo ed. (This da a sca ci y s ems om he ac ha economis s mos ly ca e abou he size o he income pie, no i s composi ion.) To da e, I’ e been able calcula e in e es -sha e da a o a hand ul o OECD coun ies co e ing he las wo decades. This sample is no g ea . Bu i ’s enough o a consis ency check. On ha on , Figu e 4shows ha he OECD da a i s wi h he pa e n ound in he Uni ed S a es. Ac oss hese coun ies, he in e es sha e o income ises and alls wi h he a e o in e es . 8 Blai Fix Economics om he Top Down A i s glance, he op 1% sha e has li le o do wi h he class-based income ha I’ e discussed so a . A e all, he legal ca ego y o you income ells us no hing abou i s size (la ge o small). And since income size and income class a e seemingly un ela ed, he e’s no eason o suspec ha in e es a es a ec income inequali y. The ca ch is ha income size and income class a e ela ed. And because hey’ e ela ed, he a e o in e es does a ec income inequali y. Le ’s see how i wo ks. The in e es - o-wage a io When i comes o in e es income, he old adage is co ec : i akes money o make money. And so i ’s a good be ha i you ea n mo e income, you’ll ha e mo e money o in es , hence you’ll ea n mo e in e es . To lesh his hinking ou , le ’s look a he ela ion be ween income size and income composi ion. Speci ically, I’m going o calcula e some hing ha I call he in e es - o-wage income a io. This a io akes he income ha an indi idual ea ns om in e es and di ides i by he income ha hey ea n om wages/sala ies: in e es - o-wage a io =in e es income wage income Fo mos people, he in e es - o-wage a io is small, since hei wages dwa hei in e es income. Bu o a ew indi iduals, in e es is a signi ican sou ce o money. As i u ns ou , hese indi iduals also happen o be op ea ne s. Figu e 7shows he pa e n in he Uni ed S a es. He e, he ho izon al axis anks Ame icans by hei income pe cen ile. The e ical axis plo s hei in e es - o-wage a io. You can see ha as we app oach he uppe c us o ea ne s, in e es income explodes. Now in di ec e ms, Figu e 7 ells us abou in e es income. Bu indi ec ly, i ells us abou he dis ibu ion o c edi . You see, as a i s app oxima ion, he a e o in e es doesn’ a y much be ween in es o s. And so he end in Figu e 7mus be d i en by he owne ship o c edi . The e o e, we can conclude ha op ea ne s own a mo e c edi han he es o us. 15 Blai Fix Economics om he Top Down Figu e 7: The in e es - o-wage a io in he Uni ed S a es This igu e measu es he composi ion o Ame icans’ income as a unc ion o hei income pe cen ile. The ho izon al axis shows income pe cen ile, anked by o al income. The e ical axis shows he in e es - o-wage a io — he a io be ween annual in e es income and annual wage/sala y income. The s eps in he blue line indica e di e en pe cen ile bins. The line indica es he median in e es - o-wage a io ac oss he yea s 1918 o 2015. The shaded egion indica es he middle 50% o da a. Sou ces and me hods Wi h his c edi dis ibu ion in mind, le ’s hink abou wha happens when we aise he a e o in e es . Doing so sends money o people who own mo e c edi . And he people who own mo e c edi also happen o be op ea ne s. And so wha do in e es - a e hikes do? They send money o he ich. (Ac ually, he ma h is a bi mo e complica ed han I’m making ou . See he appendix o de ails.) To summa ize, he e’s good eason o suspec ha highe in e es a es migh wo sen inequali y. So le ’s see i hey do. 16 Blai Fix Economics om he Top Down Figu e 8: Highe in e es a es a e associa ed wi h g ea e income inequali y This igu e looks a he in e na ional ela ion be ween he lending in e es a e (ho izon al axis) and he op 1% sha e o income ( e ical axis). The da ase co e s 148 coun ies, obse ed o e he yea s 1954 o 201. To illus a e he end, I’ e binned he da a by in e es a e. Each poin shows he midpoin o an in e es - a e bin. Blue line shows he median op 1% sha e o income wi hin each bin. The shaded egion shows he middle 50% o da a. No e ha each bin con ains a leas 10 coun y-yea obse a ions. Sou ces and me hods In e es a es and he op 1% sha e o income Wi h inequali y in mind, le ’s look a Figu e 8. He e, I’ e gone o he Wo ld Inequali y Da abase and downloaded all o hei da a o he op 1% sha e o income wi hin coun ies. Then I’ e me ged his da a wi h in e es - a e ime se ies, and plo ed he esul . As you can see, he pa e n is ai ly ob ious. Ac oss coun ies, highe in e es a es a e associa ed wi h g ea e income inequali y. 17 Blai Fix Economics om he Top Down Now, he ca ea he e is ha he inequali y pa e n is non-linea . As in e es a es inc ease, he op 1% sha e g ows — bu only o a poin . Why? Well, i we assume ha changing in e es a es ope a e on a ixed dis ibu ion o wages and c edi , hen his ype o non-linea pa e n is exac ly wha we expec . (Fo de ails, see he appendix.) Speci ics aside, i ’s clea ha income inequali y is ela ed o he a e o in- e es . So again, Rochon and Se e ield a e co ec o label in e es a es a ‘dis ibu ional a iable’. The a e age sabo age All in all, he e idence shou s a us ha he a e o in e es a ec s he dis i- bu ion o income. O cou se i ’s nice o ha e his empi ical con i ma ion. Bu hen again, he ‘dis ibu ional’ na u e o in e es a es was ne e in doub . As a a e o e u n, in e es a es a e by de ini ion a ‘dis ibu ional a iable’. O in mo e incendia y e ms, in e es a es a e a weapon o class wa a e. Now, his language migh sound hype bolic, bu I hink i ’s accu a e. Con a y o wha neoclassical economis s claim, he e a e no neu al ma ke o ces ha alloca e income in p opo ion o p oduc i i y. Ins ead he e a e only ideas and he powe o implemen hem. In o he wo ds, people ha e ideas abou wha hei income should be (and also wha o he people’s income should be). And hey ha e he powe (o lack he eo ) o make hese ideas a eali y. Tha ’s i . So iewed h ough he lens o powe , a es o e u n a e, by de ini ion, ou - comes o social con lic . S ill, his an age poin doesn’ ge us e y a in unde s anding eal-wo ld ou comes. I ’s like saying ha animal beha io is an ou come o e olu ion. I ’s ue. Bu i ’s only he s a ing poin o a scien i ic explana ion. On ha on , how should we s udy he class s uggle in ol ed in in e es income? Fo Ma x, he answe was ha in e es is abou in e -capi alis com- pe i ion. In e es paymen s, Ma x a gued, ans e o he ‘money-capi alis ’ some o he p o i s ecei ed by he ‘indus ial capi alis ’. Al hough I admi e he simplici y o his app oach, i implies ha we can cleanly di ide be ween ‘indus ial’ and ‘money’ capi alis s. And by ex ension, Ma x’s iew implies ha p o i is ea ned om ‘p oduc i e’ ac i i ies, whe eas in e es s ems om he unp oduc i e owne ship o money. 18 Blai Fix Economics om he Top Down Faced wi h his Ma xis di ision, capi al-as-powe heo is s Jona han Ni zan and Shimshon Bichle hink ha i misses he poin . Looking a capi al, hey a gue ha all o i is unp oduc i e. Tha ’s because capi al is no hing bu he quan i ica ion o p ope y igh s. And p ope y igh s, in u n, a e inhe en ly nega i e; hey a e an ins i u ional ac o exclusion. So in ha sense, p o i and in e es bo h s em om en o ced exclusion — wha Ni zan and Bichle call s a egic sabo age. Viewed his way, p o i and in e es ep esen di e en ac ics o in lic ing sabo age: [I]n e es on deb ep esen s a e age sabo age, while p o i on equi y deno es di e en ial sabo age. (Ni zan and Bichle , 2009; emphasis added) Unpacking his claim, he idea is ha when you pu chase equi y in a speci ic company, you a e in es ing in ha company’s abili y o wield p ope y igh s o you bene i . And because you hope ha his company will bea he a e age a e o e u n, you a e in es ing in ‘di e en ial’ sabo age. Wi h deb , howe e , you’ e no in es ing in any speci ic se o p ope y igh s. Ins ead, you’ e buying access o an a e age e u n on all p ope y igh s — he a e age sabo age. Clea ly his hinking is a wo ld away om mains eam mac oeconomics, which iews he a e o in e es as a a iable o b inging ‘ inancial ma - ke s in o equilib ium’ (G ego y Mankiw’s wo ds). 6 Bu we should expec as much. The co e o neoclassical economics has always been o pape o e class con lic . By using language like ‘sabo age’, Ni zan and Bichle emphasize he con lic in ol ed in se ing a es o e u n. Now whe he he a e o in e es ep esen s he ‘a e age sabo age’ is some hing ha we can deba e. Bu one hing seems clea : when i comes o dis ibu ing income, in e es a es a e no ‘neu al’. 6 The e’s a deligh ul i ony o Mankiw’s in e es - a e desc ip ion, ound in he 7 h edi ion o his ome Mac oeconomics. A e claiming ha in e es a es equilib a e inancial ma ke s, he moun s a case s udy which he calls ‘wa and in e es a es’. In i , he shows ha in he UK, in e es - a e spikes co ela e wi h wa -d i en spikes in mili a y spending (measu ed as a po ion o GDP). I akes a ue neoclassical economis s o desc ibe wa in e ms o inancial equilib ium. 19 Blai Fix Economics om he Top Down Suppo his blog Economics om he Top Down is whe e I sha e my ideas o how o c ea e a be e economics. I you liked his pos , conside becoming a pa on. You’ll help me con inue my esea ch, and con inue o sha e i wi h eade s like you. Deb and he in e es sha e o income Assuming ha o al deb and o al income g ow oge he , he e’s why he in e es sha e o income should be la gely de e mined by he a e o in e es . Le be he annual a e o in e es . Le D be he amoun o ou s anding deb wi hin a coun y. The in e es ea ned on his deb will be ·D . Le ing Y be na ional income, he in e es sha e o na ional income is: in e es sha e o income = ·D Y Now, i o al deb is oughly he same as na ional income (D≈Y) , i ollows ha he in e es sha e o income is de e mined solely by he a e o in e es , : in e es sha e o income = ·D Y ≈ ·Y Y= O cou se, his equi alence is an app oxima ion. In he eal-wo ld, deb le els a e a ely iden ical o na ional income — much o he chag in o deb hawks. In e es a es and labo -sha e clawbacks He e’s a simple model o how highe in e es a es cause he labo sha e o income o decline. 20 Blai Fix Economics om he Top Down We s a wi h he obse a ion (plo ed in Figu e 4) ha highe in e es a es a e associa ed wi h a g ea e in e es sha e o income. Fi ing a linea e- g ession o he end, we can say ha o e e y pe cen age poin inc ease in he a e o in e es , he in e es sha e o income inc eases by abou 0.7 pe cen age poin s. Nex , we assume ha a g owing in e es sha e o income is di ec ly clawed ou o labo income. In ha case, o e e y pe cen age poin inc ease in he a e o in e es , he labo sha e o income will dec ease by 0.7 pe cen age poin s: labo sha e o income =c−0.7 ×( a e o in e es ) Yes, his model is almos nai ely simple. And ye i i s he in e na ional end ai ly well. Figu e 6-mod uns he numbe s. O cou se, he ca ea is ha I’ e chosen he pa ame e c o i he da a. Bu he main poin is ha he slope o he model is on pa wi h he empi ical end. Now his clawback model has ob ious limi a ions. Fo example, i he in e es a e exceeds 80%, he model p edic s ha he labo sha e o income becomes nega i e — an impossibili y. S ill, i ’s ascina ing ha such a simple model says any hing abou he labo sha e o income. How in e es - a e hikes edis ibu e income Looking a Figu e 7( he ela ion be ween income pe cen ile and he in e es - o-wage a io), I hink i ’s in ui i e ha aising in e es a es sends income o he ich. Bu since in ui ion some imes misleads, le ’s do some o mal ma h. To di e in o he ma hema ics o income edis ibu ion, we’ll s a wi h a common misconcep ion. I Alice owns mo e c edi han Bob, i seems ob ious ha she will p e e en ially bene i om an in e es - a e hike. I ha ’s you in ui ion, know ha i is w ong. You see, when i comes o income inequali y, i is income a ios ha ma e . And hese a ios, in u n, a e no a ec ed by changes in he a e o in e es . Fo example, suppose ha Alice owns 10 imes mo e c edi han Bob. As long as Bob and Alice ea n he same a e o in e es , Alice will ecei e 10 imes as much in e es . In o he wo ds, (uni o m) a e hikes don’ edis ibu e in e es income. 21 Blai Fix Economics om he Top Down Figu e 9: A clawback model o he labo sha e o income The blue line eplo s ( om Figu e 6) he c oss-coun y end be ween in e es a es and he labo sha e o income. The ed line shows a simple model o his pa e n. Using he eg ession om Figu e 4, I assume ha he in e es sha e o income is p opo ional o he a e o in e es . Nex , I assume ha his in e es income ge s clawed ou o labo income. F om he e, i ollows ha he labo sha e o income will decline linea ly wi h ising in e es a es. Sou ces and me hods And now you a e con used. You see, I jus claimed ha in e es - a e hikes send money o he ich. Bu now I’m now a guing ha hey don’ edis ibu e in e es income. Wha gi es? To uncon use ou sel es, we need o b ing o he o ms o income in o he equa ion. To do ha , le ’s imagine a wo ld in which he e a e wo ypes o income: wages and in e es . Nex , imagine ha we hike he a e o in e - es while keeping wages unchanged. Wha happens o he dis ibu ion o income? Ob iously c edi o s ge a aise ela i e o wo ke s. 22 Blai Fix Economics om he Top Down Tha ’s nice and simple. Bu he e’s he p oblem: in he eal wo ld, he e’s no di iding line be ween ‘wo ke s’ and ‘c edi o s’. In o he wo ds, almos e e yone ea ns some combina ion o wage and in e es income. And so when i comes o he dis ibu ion e ec o in e es - a e hikes, we’ e back o being con used. To uncon use ou sel es o a second ime, we need o do some ma h. We’ll s a by w i ing an equa ion ha desc ibes indi idual income. In ou imagi- na y wo ld, each pe son’s income consis s o hei annual wage (w) plus any in e es (i) hey ea n on c edi in es men s: income =w+i In e es income, in u n, is se by he size o he c edi in es men (K) and he a e o in e es ( ) . In o he wo ds, i=K· . Pu ing ha in o ou income equa ion, we ge : income =w+K· Nex , we wan o know how income will change i we hike he a e o in e es bu lea e wages he same. Le ’s suppose ha he a e o in e es ises om 1 o 2. In ha case, income g ow h is: income g ow h =w+K· 2 w+K· 1 One las s ep. In mos cases, we don’ ca e abou he absolu e alue o you wage o you c edi in es men . Wha ma e s is he a io o hese wo quan- i ies — he c edi - o-wage a io, K w . So le ’s ew i e ou income-g ow h equa- ion in e ms o his a io. Fac o ing ou w in bo h he nume a o and he denomina o gi es: income g ow h =1+K w· 2 1+K w· 1 This equa ion ells us how you income g ows in esponse o an in e es - a e hike. The la ge you c edi - o-wage a io, he mo e you bene i . 23 Blai Fix Economics om he Top Down Figu e 10: How in e es - a e hikes edis ibu e income This igu e illus a es how he income a ec o in e es - a e hikes depends on he amoun o c edi ha you own. To ha end, he ho izon al plo s he c edi - o-wage a io — he a io o an indi idual’s c edi owne ship o hei annual wage. (The la ge his a io, he mo e you a e a ‘c edi o ’. The smalle his a io, he mo e you a e a ‘wo ke ’.) The e ical axis shows how income g ows in esponse o an in e es - a e hike (while wages a e kep cons an ). Each colo ed cu e shows he e ec o a di e en a e hike, which s a s om a baseline in e es a e o 3%. To ge a sense o how his equa ion wo ks, le ’s look a Figu e 10. He e, he ho izon al axis shows he c edi - o-wage a io, which basically indica es he deg ee o which you a e a wo ke (le ) o a c edi o ( igh ). The e ical axis hen shows how income g ows in esponse o a gi en a e hike. Ob iously, he la ge he a e hike, he bigge he pay aise o c edi o s. Bu mo e impo an ly, he colo ed cu es show how he deg ee o bene i changes con inuously as a unc ion o you class s a us. 24 Blai Fix Economics om he Top Down Now, ax eco ds come wi h some p oblems — he simples being ha axes a e ypically epo ed by households, no indi iduals. Bu he o he ‘p oblem’ is ha ax eco ds a e no consis en wi h he accoun ing de ini ions used by he na ional accoun s. I’ e used sca e quo es he e, because I don’ hink his inconsis ency is a p oblem. I any hing, ax eco ds o e he bes way o accoun o income. You see, income is undamen ally a legal concep . And so wha be e way o measu e i han by using he eco ds gene a ed by he ax code — he legal sys em ha codi ies and classi ies income. Bu o be e o wo se, Pike y and collabo a o s ha e decided o abandon he ax app oach, and ins ead, make hei me hods consis en wi h he na- ional accoun s. (They go he ball olling in his 2016 pape .) In one sense, he mo e is a good idea, since i ’s nice o ha e accoun ing echniques ha a e consis en ac oss da abases. Bu in ano he sense, I hink he change is a mis ake. You see, he na ional accoun s we e ne e designed o gi e a comp ehensi e measu e o income. They we e designed o measu e p oduc ion. Tha ’s why capi al gains — which a e ob iously income — a e no pa o he na ional accoun s. They we e deemed ‘unp oduc i e’. A any a e, he legacy ax da a s ill si s in he Wo ld Inequali y Da abase (i you know whe e o ind i ). Bu i ’s no being upda ed. And i ’s a ailable o only a hand ul o ich coun ies. Figu e 8 •Fo in e es a e sou ces, see he sou ces o Figu e 6. • Da a o he op 1% o income is om he Wo ld Inequali y Da abase, using he ollowing income sha e se ies: sp inc992j, sp inc992i, sp - inc992 , sp inc999i, sp inc996i Fu he eading Ni zan, J., & Bichle , S. (2009). Capi al as powe : A s udy o o de and c eo de . New Yo k: Rou ledge. Rochon, L.-P., & Se e ield, M. (2007). In e es a es, income dis ibu ion, and mone a y policy dominance: Pos Keynesians and he “ ai a e” o in e es . Jou nal o Pos Keynesian Economics,30(1), 13–42. 31