US employment inequality in the Great Recession and the COVID-19 pandemic

S e en M. Fazza i, Ella Needle A icle US employmen inequali y in he G ea Recession and he COVID-19 pandemic Eu opean Jou nal o Economics and Economic Policies: In e en ion (EJEEP) P o ided in Coope a ion wi h: Edwa d Elga Publishing Sugges ed Ci a ion: S e en M. Fazza i, Ella Needle (2021) : US employmen inequali y in he G ea Recession and he COVID-19 pandemic, Eu opean Jou nal o Economics and Economic Policies: In e en ion (EJEEP), ISSN 2052-7772, Edwa d Elga Publishing, Chel enham, Vol. 18, Iss. 2, pp. 223-239, h ps://doi.o g/10.4337/ejeep.2021.02.09 This Ve sion is a ailable a : h ps://hdl.handle.ne /10419/277512 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/ US employmen inequali y in he G ea Recession and he COVID-19 pandemic* S e en M. Fazza i** Washing on Uni e si y in S . Louis, MO, USA and FMM Fellow Ella Needle *** Washing on Uni e si y in S . Louis, MO, USA This a icle compa es inequali y in employmen ac oss demog aphic g oups in he G ea Recession and he COVID-19 pandemic. We de elop a measu e o cap u e bo h how much employmen declines du ing a ecession and he pe sis ence o employmen losses. Resul s show a signi ican shi o job loss om men in he G ea Recession o women in he COVID-19 lockdown. Whi e wo ke s a e be e han o he acial/e hnic g oups in bo h ecessions. Black and Hispanic women a e hi especially ha d in he COVID-19 pandemic. Wi h ou job-loss measu e, less- educa ed wo ke s had modes ly wo se ou comes in he G ea Recession. Howe e , du ing COVID-19, less-educa ed wo ke s su e much mo e se e e employmen consequences han mo e- educa ed g oups. We discuss long- e m e ec s o employmen inequali y and how hese indings a e ele an o deba es abou policy esponses. Keywo ds: employmen , unemploymen , inequali y, G ea Recession, COVID-19 JEL codes: E24, J21 1 INTRODUCTION Unde s anding he impac o ecessions on economic secu i y, especially employmen , mo i a es much o mac oeconomics. Mos analysis and commen a y desc ibe agg ega e e ec s, such as he peak- o- ough d op in employmen , he ise in he unemploymen a e, o he ime i akes o employmen o eco e back o i s p e- ecession le el. Bu as a en ion o ising economic inequali y has g own in ecen yea s, s udies o eces- sions ha e inc easingly conside ed connec ions be ween ecessions and inequali y. Mos ob iously, esea che s ha e looked mo e deeply in o how ecessions ha e unequal e ec s on di e en segmen s o socie ies. In addi ion, some s udies ha e explo ed how * An ea lie e sion o his pape was p esen ed a he 24 h FMM Con e ence, 29 Oc obe 2020. We hank Tom Fe guson and Jan Beh inge o help ul commen s. ** S e en M. Fazza i, Campus Box 1208, 1 B ookings D i e, S . Louis, MO 63130, USA; email: [email p o ec ed]. *** Email: [email p o ec ed]. Recei ed 24 Feb ua y 2021, accep ed 8 May 2021 Eu opean Jou nal o Economics and Economic Policies: In e en ion, Vol. 18 No. 2, 2021, pp. 223–239 © 2021 The Au ho Jou nal compila ion © 2021 Edwa d Elga Publishing L d The Lypia s, 15 Lansdown Road, Chel enham, Glos GL50 2JA, UK and The William P a House, 9 Dewey Cou , No hamp on MA 01060-3815, USA inequali y a ec s he condi ions ha lead o ecessions and he cou se o ecessions and eco e ies. 1 This a icle analyses inequali y in employmen ou comes ac oss social g oups du ing ecessions. We ake a compa a i e pe spec i e, s udying esul s om wo ecen and se e e US ecessions: he ‘G ea Recession’linked wi h he global inancial c isis beginning in la e 2007 and he ‘lockdown’ ecession caused by he COVID-19 pandemic. While com- pa ing jus wo ecessions in one coun y limi s he gene ali y o ou indings, hese a e signi ican , ecen down u ns in a la ge economy caused by e y di e en p ocesses. In addi ion, Hoynes e al. (2012: 28), in a de ailed s udy o employmen inequali y ac oss h ee decades o US da a, s a e di e ences in employmen cyclicali y ‘ac oss demog aphic g oups a e ema kably s able ac oss h ee decades,’sugges ing he G ea Recession is ep e- sen a i e o a longe his o y o employmen down u ns. Thus, compa ing he COVID-19 c isis o he G ea Recession p o ides an in e es ing case s udy o explo e inequali y in ecessions. This wo k also ex ends compa a i e analysis o employmen e ec s o eces- sions o a d ama ic economic c isis wi h his o ically unique ea u es. We begin in Sec ion 2 wi h an o e iew o he wo majo ecessions. We ocus ou analysis on employmen dynamics and accoun o he ac ha he se e i y o a ecession depends bo h on how much employmen declines and he pe sis ence o hose declines. In Sec ion 3, we p opose a simple measu e ha cap u es bo h o hese dimensions: job-mon hs los , de ined as he sum o mon hly employmen losses ela i e o he business-cycle peak added o e a gi en numbe o mon hs since he cycle peak. While he concep is simple, implemen ing i p esen s wo challenges in ou compa a i e con ex . Fi s , we need o accoun o di e en employmen ends ac oss ecessions and ac oss he demog aphic g oups we s udy. Second, we need o deal wi h he e y di e en dynamics o he G ea Recession ( ela i ely slow decline in employmen ollowed by e y slow eco e y) compa ed wi h he COVID-19 lockdown (d ama ically sha p all in employmen and quick pa ial bounceback). To ou knowledge, his a icle is he i s in esea ch on compa a i e labo ma ke ecessions o accoun o hese ac o s. In addi ion, we p esen a simple summa y s a is ic o quickly assess he ela i e impac o job-mon hs los o di e en demog aphic g oups ac oss he wo ecessions we s udy. We di ide a g oup’s sha e o job-mon hs los in each ecession by he g oup’s sha e in employmen a he cycle peak p io o he ecession. This s a is ic p o ides a s aigh o wa d summa y o a g oup’s ela i e ad an age (i he s a is ic is less han one) o a g oup’s disad an age (i he s a is ic exceeds one). Sec ion 4 p esen s ou main esul s based on da a om he US Cu en Popula ion Su ey (CPS). Ou esul s show: •A signi ican shi o he bu den o job losses om men in he G ea Recession o women in he COVID-19 lockdown. These indings a e consis en wi h o he s u- dies o employmen e ec s o he G ea Recession (and ea lie down u ns as well, see Hoynes e al. 2012). •Whi e wo ke s a e be e han Asian, Black, and Hispanic employees in bo h eces- sions. Black and Hispanic women a e hi especially ha d in he COVID-19 1. S udies ha analyse he unequal e ec s o ecessions include K uege e al. (2017) and Am o- min e al. (2018). Examples o esea ch ha conside how inequali y a ec s he mac oeconomic dynamics o ecession include Ba ba/Pi e i (2009), Cynamon/Fazza i (2015; 2016), and Aue bach e al. (2020). He e odox g ow h models in he Kaleckian adi ion ha e, o decades, emphasized he impac o he unc ional dis ibu ion o income be ween wages and p o i s, an economic cha ac e - is ic closely ela ed o inequali y. This esea ch is summa ized and in e p e ed in se e al ecen book- leng h ea men s including Hein (2014), La oie (2014), and Blecke /Se e ield (2019). 224 Eu opean Jou nal o Economics and Economic Policies: In e en ion, Vol. 18 No. 2 © 2021 The Au ho Jou nal compila ion © 2021 Edwa d Elga Publishing L d pandemic. Job-mon hs los a e 45 pe cen highe o Black women and 63 pe cen highe o Hispanic women han hei sha e in p e-c isis employmen would imply. •Young wo ke s su e disp opo iona e, and simila , job losses in bo h ecessions. Middle-aged wo ke s ha e been a ec ed less se e ely in he COVID-19 c isis han in he G ea Recession. Olde wo ke s ha e done much wo se in COVID-19 com- pa ed wi h he G ea Recession. •Job-mon hs los in he G ea Recession we e somewha skewed owa d wo ke s wi h lowe educa ion (a good p oxy o income), bu , when app op ia ely accoun - ing o ends, he inequali y ac oss educa ion g oups in he G ea Recession is ea- sonably modes . Howe e , less-educa ed wo ke s su e d ama ically mo e employmen loss due o COVID-19 han mo e-educa ed g oups wi h sha es o job-mon hs los 50 pe cen highe han he p e-c isis employmen sha e o wo ke s wi h a high-school diploma and 79 pe cen highe o wo ke s wi h less han a high- school educa ion. Th ee majo ac o s help explain hese esul s. Fi s , he e was a signi ican shi in he sec o al composi ion o job losses be ween he ecessions. While he G ea Recession (as well as ea lie down u ns) hi male-domina ed manu ac u ing and cons uc ion sec o s especially ha d, he COVID-19 c isis had a much bigge impac on se ice and ca e jobs in which women comp ise a majo i y, some imes a la ge majo i y, o employmen . The se ice-sec o jobs los in he pandemic also paid low wages on a e age, explaining he massi e inequali y ac oss educa ion g oups. Second, childca e challenges ose d ama i- cally in he pandemic due o school and dayca e closings. Female employmen was a ec ed ela i ely mo e by expanding childca e needs. Thi d, al hough we do no ha e di ec da a on age esponses o he c isis, i seems likely ha he mo e se e e heal h impac o he pandemic on olde popula ions a ec ed employmen ou comes ac oss age g oups. While ou main esul s emphasize how ecessions magni y p e-exis ing inequali ies in he sho e m, Sec ion 5 conside s how he inequali ies caused by ecessions a ec longe - e m ends o inequali y and how inequali y ma e s o long- e m g ow h. Sec ion 6 con- cludes, wi h a b ie summa y and discussion o policy implica ions o ou indings. 2 TWO HISTORIC RECESSIONS As ea ly as he 1990s, mains eam mac oeconomis s decla ed ic o y in he wa agains deep ecessions. De eloped economies had en e ed a pe iod desc ibed as he ‘G ea Mode a ion’ (S ock/Wa son 2002). Con en ional economic wisdom came o belie e ha mone a y policy is he p ima y ool o mac oeconomic s abiliza ion and, when implemen ed wisely, should keep ecessions modes . Indeed, a e he deep and dis up i e ecession o he ea ly 1980s, he US expe ienced a qua e -cen u y wi h only wo mild ecessions. The global inancial c isis and subsequen G ea Recession sha e ed his benign iew. Fo a ew sca y mon hs in la e 2008 and ea ly 2009, job losses mimicked he collapse o he ea ly 1930s. Unemploymen soa ed. Despi e agg essi e mone a y policy, including a apid d op o sho - e m in e es a es o ze o, he ecession con inued and go e nmen s summoned iscal ools o a emp a escue. The wo s o he economic bleeding s opped by mid 2009, bu eco e y was slow. I con inually disappoin ed con en ional o ecas s ha p ojec ed a e u n o he US economy o i s p e-c isis end. Some mains eam econ- omis s, mos p ominen ly Law ence Summe s (2014; 2018), began o sound mo e he e o- dox. Summe s esu ec ed he concep o ‘secula s agna ion’due o insu icien agg ega e US employmen inequali y in he G ea Recession and he COVID-19 pandemic 225 © 2021 The Au ho Jou nal compila ion © 2021 Edwa d Elga Publishing L d demand g ow h, while poli ically mo i a ed conse a i es blamed excessi e egula ion and an icipa ed ax inc eases due o P esiden Obama’s policies. Bu all ag eed ha he 2008– 2009 collapse and subsequen eco e y showed his o ic weakness in he US economy. Al hough he eco e y beginning in 2009 was weak, i was pe sis en . Indeed, i would ha e su ely las ed longe had he public-heal h c isis c ea ed by COVID-19 no hi he economy like a sledgehamme in he sp ing o 2020. A e 113 mon hs o posi i e job g ow h, US pay oll employmen declined by mo e han 22 million jobs (14.5 pe cen ) in Ma ch and Ap il. Al hough he G ea Recession employmen decline was his o ically se e e i was no nea ly as apid (25 mon hs om he Janua y 2008 peak o he Feb ua y 2010 ough) and he o al job loss was 6.3 pe cen , less han hal o he peak- o- ough loss caused by he COVID-19 lockdown. Bu he US job ma ke su - p ised many o us wi h a ema kably quick bounceback in he summe o 2020. The economy added mo e han 12 million jobs om May h ough Oc obe (4.8 million in June alone). Bu as i us in ec ions ose in he all, as p edic ed by epidemiologis s, he job eco e y s alled. As o Decembe 2020, US pay oll employmen emained 6.5 pe cen below i s Feb ua y peak, abou he same as he job-loss pe cen age a he ough o he G ea Recession labo ma ke . These wo ecessions we e bo h se e e, bu wi h di e en dynamics. They p o ide a use ul labo a o y o explo ing how ecessions bege inequali ies and how hese inequali ies di e ac oss wo majo down u ns wi h ma kedly di e en sou ces and cha ac e is ics. I is use ul o no e, howe e , ha bo h ecessions we e he esul o demand-side e ec s ha domina ed any p oblems on he supply side. The dominance o demand in he G ea Recession is widely accep ed. This conclusion is less ob ious o he COVID-19 c isis, howe e . The lockdown ob iously discou aged spending on ac i i ies ha expose people o in ec ion (c uises, ai a el, es au an s, in-pe son en e ainmen , spo ing e en s, e c.). Bu i is also ob ious ha lockdowns had di ec supply impac , such as he shu down o au o manu ac u ing o se e al weeks in he sp ing o 2020. Supply also became a binding cons ain on p oduc ion o i ems o which demand su ged due o he c isis (pe sonal p o ec ion equipmen , bicycles, gym supplies, home o ice u ni u e, cleaning p oduc s, e c.). To ge a b oad sense o whe he supply-side o demand-side e ec s domina ed he ecession, we look a p ice da a. Figu es 1a and 1b show p ice-le el ends leading up o and h ough he G ea Recession and he COVID-19 c isis. In bo h c ises, in la ion declined. Fo he ‘co e’measu e ha excludes ola ile ood and ene gy p ices, in la ion was e ec i ely ze o in he G ea Recession, while headline pe sonal consump ion p ices ell signi ican ly. In he COVID-19 c isis, bo h headline and co e pe sonal consump ion p ices declined. 2 In addi ion, in e es a es plumme ed in bo h ecessions, inconsis en wi h wha would be expec ed i supply shocks we e he dominan engine o ecession. While one can iden i y some possible supply e ec s in he G ea Recession and la ge ce ain supply e ec s in he COVID-19 c isis, he e idence shows demand con ac ions domina ed, a inding ele an o bo h analy ical and policy pu poses. 3 MEASURING LABOR MARKET INEQUALITY To measu e inequali y in he job ma ke we analyse employmen o gende , ace, age, and educa ion le el wi h da a om he US CPS conduc ed mon hly by he Bu eau o Labo S a is ics. This is he same su ey used o de e mine he mon hly headline 2. P ice da a om he Consume P ice Index ha e simila pa e ns. 226 Eu opean Jou nal o Economics and Economic Policies: In e en ion, Vol. 18 No. 2 © 2021 The Au ho Jou nal compila ion © 2021 Edwa d Elga Publishing L d 94.0 92.0 90.0 88.0 86.0 84.0 82.0 80.0 PCE index PCE excluding ood and ene gy (a) G ea Recession (b) COVID-19 ecession Index Janua y, 2018 = 100 Jan 2006 Jul 2006 Jan 2007 Jul 2007 Jan 2008 Jul 2008 Jan 2009 Jul 2009 Jan 2010 Jul 2010 Jan 2011 Jul 2011 97.0 98.0 99.0 100.0 101.0 102.0 103.0 104.0 105.0 106.0 Jan 2017 Jul 2017 Jan 2018 Jul 2018 Jan 2019 Jul 2019 Jan 2020 Jul 2020 PCE index PCE excluding ood and ene gy Figu e 1 Pe sonal consump ion p ices US employmen inequali y in he G ea Recession and he COVID-19 pandemic 227 © 2021 The Au ho Jou nal compila ion © 2021 Edwa d Elga Publishing L d Index Janua y, 2018 = 100 unemploymen a e. 3 We ocus on employmen a he han unemploymen because indi iduals who s op wo king du ing a ecession will no be coun ed as unemployed i hey s op ac i ely sea ching o a new job. This kind o job loss could be pa icula ly impo an in he COVID-19 ecession when pa en s (disp opo iona ely mo he s) we e o ced o lea e wo k o ca e o child en who could no longe a end in-pe son childca e o school. Also, wo ke s in ha d-hi indus ies such as es au an s may ha e seen li le eason o ac i ely seek new jobs when he indus y almos comple ely shu down. As discussed ea lie , ou analysis needs o accoun o bo h he se e i y and pe sis ence o job losses. The di e en dynamics o he G ea Recession and he COVID-19 ecession con i m he impo ance o conside ing bo h dimensions o ecession employmen . The ema kably as and deep job losses in he ea ly mon hs o he COVID-19 ecession dwa - ed he job losses in any simila pe iod o he G ea Recession. And he ough o employ- men ela i e o he peak was much deepe in COVID-19. Tha said, he ini ial job eco e y was much as e in he summe o 2020 han he slow and pain ul job g ow h ollowing he G ea Recession. Figu e 2 p o ides an example o his issue. I shows he CPS da a o emale employmen om se e al yea s be o e he G ea Recession h ough he mos ecen da a on he COVID-19 c isis a ailable a he ime o w i ing (Decembe 2020). Bo h ecessions a e ob ious, bu he e is ambigui y abou which is wo se: sha p and e y deep (COVID-19) e sus mo e shallow bu mo e pe sis en (G ea Recession). The igu e also shows ha p e- ecession ends in employmen igu es a e ele an o assessing he se e i y o job losses. In he example o emale employmen in Figu e 2, i is clea ha p e- ecession job igu es had a s ong upwa d end. Recession e ec s designed o explo e inequali y should be measu ed ele an o hese ends ha we ind di e ac oss spli s o he da a. 4 We p opose a simple measu e o employmen dis up ion ha combines bo h he dep h o job losses and hei pe sis ence: job-mon hs los . This s a is ic simply adds he di e ence be ween ac ual employmen and he p e- ecession end o a gi en numbe o mon hs ollowing he p e- ecession peak. E ec i ely, job-mon hs los is he a ea be ween ac ual employmen and he benchma k end on a g aph such as Figu e 2, up o a speci ied ho izon. To implemen his calcula ion, we need o choose an ending ho izon o each eces- sion. The choice is simple o he COVID-19 c isis. As o his w i ing, we ha e en eces- sion mon hs o da a since he Feb ua y 2020 peak ( ha is, Ma ch h ough Decembe ) and we use all o hese mon hs. The choice is less clea o he G ea Recession. We could use jus en mon hs, bu since he G ea Recession played ou much mo e slowly, his choice would g ea ly unde s a e bo h he dep h and pe sis ence o job losses in ha c isis. 5 Fo 3. To al employmen numbe s, such as he ones we men ioned in Sec ion 2, a e usually aken om he pay oll employmen epo s, which a e less ola ile han he CPS household su ey. How- e e , pay oll employmen da a a e no s a i ied by demog aphic g oups. The CPS da a coun employmen o all indi iduals aged 16 and abo e, excep o he educa ion spli s which coun indi- iduals aged 25 and up. 4. We use a simple linea end es ima ed o 36 mon hs p io o he peak employmen ha p e- cedes each ecession. The end is calib a ed so ha ac ual employmen equals end employmen a he p e- ecession peak. O cou se, he e a e good easons o ques ion whe he simple linea ends cap u e ‘no mal’employmen dynamics ha could easonably ha e been expec ed o con inue in he absence o a ecession. Fo example, ising a e age age o he labo o ce likely slows end g ow h. Such e ec s, howe e , play ou slowly o e ime. Fo ou sho - un analysis, non-linea i ies in he end likely play a mino ole. 5. Job-mon hs los ela i e o end in he i s en mon hs o he G ea Recession we e jus 13 pe cen o he losses in he i s en mon hs o he COVID-19 ecession. Bu as one ex ends he 228 Eu opean Jou nal o Economics and Economic Policies: In e en ion, Vol. 18 No. 2 © 2021 The Au ho Jou nal compila ion © 2021 Edwa d Elga Publishing L d ou p ima y s a is ics, we use a 36-mon h ho izon o he G ea Recession, bu we con- side he obus ness o esul s wi h sho e and longe ho izons as discussed in he nex sec ion. Ano he iming issue a ises om de ining he s a o he G ea Recession employ- men c isis o di e en demog aphic g oups. The CPS o al employmen se ies peaks in No embe o 2007 be o e declining in o he ecession (consis en wi h he con en ional de ini ion o he s a o he ecession in Decembe 2007). We use his s a da e o he G ea Recession o he s a is ics p esen ed in he ables o ollow. Howe e , o some demog aphic g oups, employmen be o e he G ea Recession peaks a ew mon hs ea lie o la e han he agg ega e peak. Ou main esul s a e no much a ec ed by adjus ing he G ea Recession peaks o di e en s a da es o each demog aphic g oup. Fo he COVID-19 c isis, he peak is clea ly in Feb ua y 2020 o all g oups. 4 COMPARISONS OF EMPLOYMENT INEQUALITY ACROSS RECESSIONS Table 1 p esen s he i s se o ou p ima y empi ical esul s o illus a e how gende inequali y shi ed be ween he G ea Recession and he COVID-19 c isis. A he peak o employmen be o e he ecession, men held 53.6 pe cen o o al CPS jobs, bu du ing 55 000 60 000 65 000 70 000 75 000 80 000 Female employmen (millions) Women em p loyed P e-G ea Recession T end P e-COVID-19 end Figu e 2 Female employmen in he G ea Recession and COVID-19 c isis ho izon, hese igu es change d ama ically because he deepes job losses in he G ea Recession a e 12 o 25 mon hs a e he beginning o he down u n and he e is i ually no eco e y ela i e o end. We made some simple p ojec ions o he eco e y p o ile o COVID-19, an icipa ing as g ow h when he accine becomes widesp ead in he summe o 2021. Wi h hese p ojec ions, we es ima e ul ima e job-mon hs los o a 36-mon h ho izon in he COVID-19 c isis will be 25 pe cen less han he 36-mon h job-mon hs los igu e o he G ea Recession (all calcula ions a e ela i e o p e- ecession ends). US employmen inequali y in he G ea Recession and he COVID-19 pandemic 229 © 2021 The Au ho Jou nal compila ion © 2021 Edwa d Elga Publishing L d Jan 00 Jan 01 Jan 02 Jan 03 Jan04 Jan 05 Jan 06 Jan 07 Jan 08 Jan 09 Jan 10 Jan 11 Jan 12 Jan 13 Jan 14 Jan 15 Jan 16 Jan 17 Jan 18 Jan 19 Jan 20 he 36 mon hs ollowing he peak, cumula i e job-mon hs los o men accoun ed o 64.0 pe cen o o al job-mon hs los du ing his pe iod. Ou key s a is ic is he a io o hese wo sha es, he job-mon hs los sha e o he sha e o jobs held be o e he ecession began. I his a io is one, cumula i e job losses o e he ecession ho izon would be equal o he sha e o he g oup’s p e- ecession jobs implying an equal impac . Howe e , o men in he G ea Recession ou a io s a is ic is 1.19 (equal o 64.0 pe cen / 53.6 pe cen ), indica ing men did subs an ially wo se du ing he ecession han hei sha e o employ- men would sugges . In con as he a io s a is ic o women o 0.78 implies women did be e han implied by hei sha e o he jobs alone (e en hough women did lose jobs du ing he ecession). 6 Ano he way o in e p e hese s a is ics is ha job-mon hs los o men we e 19 pe cen highe in he G ea Recession han hei sha e in p e- eces- sion would imply, while job-mon hs los o women we e 22 pe cen less han expec ed gi en hei sha e o he employed popula ion a he beginning o he G ea Recession. Men aking a g ea e ela i e hi o employmen in he G ea Recession han women is no su p ising since he c isis hi manu ac u ing and esiden ial cons uc ion employmen pa icula ly ha d, bo h sec o s in which men cons i u e a much la ge sha e o employ- men han women. 7 Bu hese inequali ies e e se in he i s en mon hs o he COVID-19 c isis. In his pe iod, women su e ed a disp opo iona e sha e o job-mon hs los while men did a li le be e han hei p e- ecession employmen sha e would sugges . Two easons jump ou o his signi ican shi in impac . Fi s , COVID-19 hi se ice jobs pa icula ly ha d in indus ies such as es au an s, a el, and (pe haps su p isingly) heal hca e, all sec o s in which women hold a la ge sha e o he jobs. 8 Second, women Table 1 Gende inequali y G ea Recession (36-mon h ho izon) Men Women P e- ecession peak sha e o jobs 53.6% 46.4% Sha e o job-mon hs los 64.0% 36.0% Ra io o loss sha e o ini ial sha e 1.19 0.78 COVID-19 c isis (13-mon h ho izon) P e- ecession peak sha e o jobs 52.8% 47.2% Sha e o job-mon hs los 48.6% 51.4% Ra io o loss sha e o ini ial sha e 0.92 1.09 6. Al hough we conside en mon hs oo sho a ho izon o ully assess he G ea Recession, hese igu es do no change meaning ully e en i he ho izon o he G ea Recession calcula ions is sho - ened o en mon hs. 7. Hoynes e al. (2012) ind la ge e ec s on bo h he unemploymen and employmen a es on men compa ed o women in he G ea Recession and ea lie ecessions. Pissa ides (2013) inds simi- la esul s o OECD coun ies and links he bigge e ec o men speci ically o ela i ely la ge declines in he cons uc ion and manu ac u ing sec o s. 8. Da a om he es ablishmen pay oll employmen su ey s ongly con i m sec o al di e ences be ween he G ea Recession and he COVID-19 ecession. Peak- o- ough, manu ac u ing employ- men and cons uc ion employmen ell 17 pe cen and 26 pe cen espec i ely in he G ea Reces- sion. In he COVID-19 ecession he co esponding losses a e 11 pe cen and 15 pe cen . (Acco ding o he US Census Ame ican Communi y Su ey om 2019, 71 pe cen o manu ac u - ing employees and 90 pe cen o cons uc ion employees we e male.) In con as , leisu e and hos- pi ali y jobs, including accommoda ion and ood se ice, declined jus 5 pe cen in he G ea Recession while heal h and educa ion employmen ose by 5 pe cen . In he COVID-19 230 Eu opean Jou nal o Economics and Economic Policies: In e en ion, Vol. 18 No. 2 © 2021 The Au ho Jou nal compila ion © 2021 Edwa d Elga Publishing L d unce ain y o hose amilies mos se e ely a ec ed by he c isis. A mo e gene ous sys em wi h a la ge ole o he ede al go e nmen is needed. 16 Fede al e o s a e also needed o imp o e e icien access o unemploymen assis ance; s o ies o huge backlogs and bo le- necks a ound unemploymen insu ance ha e been ampan in he COVID-19 c isis. The inequali ies induced by mac oeconomic luc ua ions also e eal he g ea need o ins i u- ional e o ms no speci ically associa ed wi h ecessions. A p ime example is suppo o childca e and amily lea e. I is clea ha gende inequali ies in employmen a e o a la ge ex en he esul o he need o ca e o child en who could no longe a end school o dayca e in pe son du ing he pandemic (see Collins 2020; Collins e al. 2021). Shi ing ou ocus o he mac oeconomic implica ions o inequali y, he mos ob ious poin is ha magni ica ion o inequali y du ing and a e he ecession will likely educe consump ion, lowe agg ega e demand, and slow he g ow h o he economy. This e ec would be con ined la gely o he sho un o ex book mac oeconomics, bu he e odox heo y and e idence iden i ies easons ha low demand can cons ain ou pu , employ- men , and li ing s anda ds ‘beyond he sho un,’as ecognized by he ecen p ominence o he ‘secula s agna ion’concep . 17 These conside a ions show how he inequali y e ec s o he ecession inc ease he need o e ec i e mac oeconomic policy. The connec ion be ween ecessions, inequali y, and mac oeconomic policy has ano he , somewha mo e sub le, implica ion. Ou esul s demons a e how indi iduals and amilies in lowe socioeconomic ci cums ances su e mo e se e e e ec s om he ecession. The goal o social equi y, he e o e, implies hey should ecei e disp opo iona e elie om policies designed o s imula e he agg ega e economy. This poin ela es o he deba e in he US abou he size and dis ibu ion o ‘s imulus checks’sen o Ame ican households. These paymen s a e a ixed amoun pe pe son up o a ce ain income le el, a e which he paymen s phase ou . Some economis s oppose his kind o demand s imulus because much o i will be sa ed a he han consumed, leading o less impac on demand. The c i icism has some ele ance, especially in he mids o a pandemic when people canno o choose no o spend on many disc e iona y ac i i ies o p o ec hei heal h. Bu lump-sum paymen s o a cons an amoun ac oss indi iduals ha e a much big- ge p opo ional impac on lowe income households. A disp opo iona e sha e o hese households a e a ec ed di ec ly by he pandemic and in c i ical need o assis ance. Fo lowe -income households ha may ha e dodged income losses, he lump-sum paymen will s ill be a welcome, and signi ican , inancial boos . I hey do no spend he unds, hey can pay down deb o build some ese e o u u e inancial shocks. These paymen s, a leas du ing he pandemic, a e a s ep owa d he impo an social goal o g ea e eco- nomic equali y. REFERENCES Alon, T., Doepke, M., Olms ead-Rumsey, J., Te il , M. (2020): This ime i ’s di e en : he ole o women’s employmen in a pandemic ecession, NBER Wo king Pape 27660, Augus . Am omin, G., De Na di, M., Schulze, K. (2018): Inequali y and ecessions, in: Chicago Fed Le e , (392), 1–5, URL: h ps://www.chicago ed.o g/~/media/publica ions/chicago- ed-le e /2018/ c l392-pd /pd . 16. The US ede al go e nmen is also much be e able o accommoda e he iscal needs o eces- sion elie since ede al budge s, unlike mos s a es, can un signi ican de ici s du ing ecessions. 17. Fo a b ie and ecen pe spec i e, see S ansbu y/Summe s (2019). Fazza i (2020) adds con ex o his poin . US employmen inequali y in he G ea Recession and he COVID-19 pandemic 237 © 2021 The Au ho Jou nal compila ion © 2021 Edwa d Elga Publishing L d Aue bach, A.J., Go odnichenko, Y., Mu phy, D. (2020): Inequali y, iscal policy and COVID19 es ic ions in a demand-de e mined economy, NBER Wo king Pape 27366. Ba ba, A., Pi e i, M. (2009): Rising household deb : i s causes and mac oeconomic implica ions: a long pe iod analysis, in: Camb idge Jou nal o Economics, 33(1), 113–137. Ba ik, A.W., Cullen, Z.B., Glaese , E.L., Luca, M., S an on, C.T. (2020): Wha jobs a e being done a home du ing he COVID-19 c isis? E idence om i m-le el su eys, NBER Wo king Pape 27422, June. Bell, D.N.F., Blanch lowe , D.G. (2011): Young people and he G ea Recession, in: Ox o d Re iew o Economic Policy, 27(2), 241–267. Blecke , R., Se e ield, M. (2019): He e dox Mac oeconomics: Models o Demand, Dis ibu ion and G ow h, Chel enham, UK and No hamp on, MA: Edwa d Elga Publishing. Bui, T.T.M., Bu on, P., Piccio i, E.G. (2020): Ea ly e idence on he impac o COVID-19 and he ecession on olde wo ke s, NBER Wo king Pape 27448, June. Cheng, W., Ca lin, P., Ca oll, J., Gup a, S., Rojas, F.L., Mon eno o, L., Nguyen, T.D., Schmu e, I.M., Sc i ne , O., Kosali, S.I., Wing, C., Weinbe g, B. (2020): Back o business and ( e) employing wo ke s? Labo ma ke ac i i y du ing s a e COVID-19 eopenings, NBER Wo king Pape 27419, June. Collins, C. (2020): Who o blame and how o sol e i : mo he s’pe cep ions o wo k– amily con lic ac oss wes e n policy egimes, in: Jou nal o Ma iage and Family, 82(3), 849–874. Collins, C., Ruppanne , L., Landi a , L.C., Sca bo ough, W. (2021): The gende ed consequences o a weak in as uc u e o ca e: school eopening plans and pa en s’employmen du ing he COVID-19 pandemic, in: Gende & Socie y, 35(2), 180–193. Co ing on, M., Ken , A. (2020): The ‘She-Cession’pe sis s, especially o women o colo , On he Economy Blog, Fede al Rese e Bank o S . Louis, 24 Decembe , URL: h ps://www.s louis ed. o g/on- he-economy/2020/decembe /she-cession-pe sis s-women-o -colo . Cynamon, B.Z., Fazza i, S.M. (2015): Rising inequali y and s agna ion in he US economy, in: Eu - opean Jou nal o Economics and Economic Policy, 12(2), 170–182. Cynamon, B.Z., Fazza i, S.M. (2016): Inequali y, he G ea Recession and slow eco e y, in: Cam- b idge Jou nal o Economics, 40(2), 373–399. Fazza i, S.M. (2020): Was Keynesian economics e e dead? I so, has i been esu ec ed?, in: Re iew o Keynesian Economics, 8(1), 46–60. Fazza i, S.M., Fe i, P., Va ia o, A. (2020): Demand-led g ow h and accommoda ing supply, in: Camb idge Jou nal o Economics, 44(3), 583–605. Gould, E. (2020): Olde wo ke s can’ wo k om home and a e a a highe isk o COVID-19, Economic Policy Ins i u e Wo king Economics Blog, 31 Ma ch, URL: h ps://www.epi.o g/ blog/olde -wo ke s-can -wo k- om-home-and-a -high- isk- o -co id-19/. Hein, E. (2014): Dis ibu ion and G ow h A e Keynes, Chel enham, UK and No hamp on, MA: Edwa d Elga Publishing. Hoynes, H., Mille , D.L., Schalle , J. (2012): Who su e s du ing ecessions?, in: Jou nal o Economic Pe spec i es, 26(3), 27–48. I ons, J. (2009): Economic sca ing: he long- e m impac s o he ecession, Economic Policy Ins i- u e B ie ing Pape No 243, Sep embe . K uege , D., Mi man, K., Pe i, F. (2017): On he dis ibu ion o wel a e losses o la ge ecessions, in: Hono é, B., Pakes, A., Piazzesi, M., Samuelson, L. (eds), Ad ances in Economics and Econo- me ics: Volume 2, Camb idge, UK: Camb idge Uni e si y P ess, 143–184. La oie, M. (2014): Pos Keynesian Economics: New Founda ions, Chel enham, UK and No hamp- on, MA: Edwa d Elga Publishing. Lundbe g, S. (1985): The added wo ke e ec , in: Jou nal o Labo Economics, 3(1), 11–37. Pissa ides, C.A. (2013): Unemploymen in he G ea Recession, in: Economica, 80, 385–403. Rampell, C. (2021): Opinion: mo e co id elie is u gen –Janua y’s jobs epo shows whe e he need is g ea es , in: The Washing on Pos , 5 Feb ua y, URL: h p://web.a chi e.o g/web/ 20210313041546i _/h ps://www.washing onpos .com/opinions/2021/02/05/janua y-jobs- epo -co id- elie -need/. S ansbu y, A., Summe s, L.H. (2019): Whi he cen al banking?, in: P ojec Syndica e, 23 Augus . 238 Eu opean Jou nal o Economics and Economic Policies: In e en ion, Vol. 18 No. 2 © 2021 The Au ho Jou nal compila ion © 2021 Edwa d Elga Publishing L d S ock, J., Wa son, M. (2002): Has he business cycle changed and why?, in: NBER Mac oeconomics Annual, 17, 159–218. Summe s, L.H. (2014): U.S. economic p ospec s: secula s agna ion, hys e esis, and he ze o lowe bound, in: Business Economics, 49(2), 65–73. Summe s, L.H. (2018): The h ea o secula s agna ion has no gone away, Blog Pos , URL: h p:// la ysumme s.com/2018/05/06/ he- h ea -o -secula -s agna ion-has-no -gone-away. US employmen inequali y in he G ea Recession and he COVID-19 pandemic 239 © 2021 The Au ho Jou nal compila ion © 2021 Edwa d Elga Publishing L d