The short- and medium-term distributional effects of the German minimum wage reform

Caliendo, Ma co; Fedo e s, Alexand a; P euss, Mal e; Sch öde , Ca s en; Wi b od , Linda A icle — Published Ve sion The sho - and medium- e m dis ibu ional e ec s o he Ge man minimum wage e o m Empi ical Economics P o ided in Coope a ion wi h: Sp inge Na u e Sugges ed Ci a ion: Caliendo, Ma co; Fedo e s, Alexand a; P euss, Mal e; Sch öde , Ca s en; Wi b od , Linda (2022) : The sho - and medium- e m dis ibu ional e ec s o he Ge man minimum wage e o m, Empi ical Economics, ISSN 1435-8921, Sp inge , Be lin, Heidelbe g, Vol. 64, Iss. 3, pp. 1149-1175, h ps://doi.o g/10.1007/s00181-022-02288-4 This Ve sion is a ailable a : h ps://hdl.handle.ne /10419/308177 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/ Empi ical Economics (2023) 64:1149–1175 h ps://doi.o g/10.1007/s00181-022-02288-4 The sho - and medium- e m dis ibu ional e ec s o he Ge man minimum wage e o m Ma co Caliendo1 ·Alexand a Fedo e s2 ·Mal e P euss3 · Ca s en Sch öde 4 ·Linda Wi b od 5 Recei ed: 13 July 2021 / Accep ed: 13 July 2022 / Published online: 5 Sep embe 2022 © The Au ho (s) 2022, co ec ed publica ion 2022 Abs ac This s udy quan i ies he dis ibu ional e ec s o he minimum wage in oduced in Ge - many in 2015. Using de ailed Socio-Economic Panel su ey da a, we assess changes in he hou ly wages, wo king hou s, and mon hly wages o employees who we e en i- led o be paid he minimum wage. We employ a di e ence-in-di e ences analysis, exploi ing egional a ia ion in he “bi e” o he minimum wage. A he bo om o he hou ly wage dis ibu ion, we documen wage g ow h o 9% in he sho e m and 21% in he medium e m. A he same ime, we ind a educ ion in wo king hou s, such ha he inc ease in hou ly wages does no lead o a subo iona e inc ease in mon hly wages. We conclude ha wo king hou s adjus men s play an impo an ole in he dis ibu ional e ec s o minimum wages. Keywo ds Minimum wage ·Wage dis ibu ion ·Hou ly wages ·Inequali y JEL Classi ica ion J31 ·J38 ·J22 We a e g a e ul o inancial suppo om he Leibniz Associa ion unde he esea ch p ojec “E alua ing he Minimum Wage In oduc ion in Ge many (EVA-MIN)—Inno a i e Knowledge T ans e and E idence-Based E alua ion.” The au ho s hank S e en Du lau , Ingo Ispho ding, Je ey Woold idge, as well as pa icipan s in he annual con e ences o ESPE (Glasgow, 2017), he Socie y o he S udy o Economic Inequali y (ECINEQ, New Yo k Ci y, 2017), EEA (Lisbon, 2017), Ve ein ü Socialpoli ik (Vienna, 2017), and EALE (S . Gallen, 2017) o hei eedback. The S a a code used in he p epa a ion o his pape is a ailable on eques om Alexand a Fedo e s in he SOEP Depa men a DIW Be lin, 10117 Be lin, Ge many. A p e ious e sion o his pape was ci cula ed as “The Sho -Te m Dis ibu ional E ec s o he Ge man Minimum Wage Re o m.” A supplemen a y appendix is a ailable online. This pape e lec s he pe sonal iews o he au ho s and no necessa ily hose o he Ge man Council o Economic Expe s. BAlexand a Fedo e s [email p o ec ed] Ex ended au ho in o ma ion a ailable on he las page o he a icle 123 1150 M. Caliendo e al. 1 In oduc ion As social policy ins umen , minimum wages aim o educe wage inequali y. By inc easing ea nings a he lowe end o he wage dis ibu ion, minimum wages ha e been shown o achie e his goal in he Uni ed S a es (DiNa do e al. 1996;Lee1999; Teulings 2003; Neuma k e al. 2004; Au o e al. 2016) and he Uni ed Kingdom (Dick- ens and Manning 2004a,b; Dol on e al. 2012). Howe e , mos o he co esponding li e a u e e e s o inc emen al changes in ede al minimum wage le els.1Agains his backd op, he p esen a icle examines a majo labo ma ke e o m in Ge many ha in oduced a high s a u o y na ional minimum wage in 2015. The e o m di ec ly a ec ed 10–15% o he employed popula ion, co esponding o be ween 4 and 6 mil- lion wo king people (Amlinge e al. 2016; B enke 2014; Falck e al. 2013; Kalina and Weinkop 2014; Lesch e al. 2014). In he ollowing, we p o ide an in-dep h analysis o he e o m’s dis ibu ional e ec s a he indi idual le el in he sho and medium e m: We es how much hou ly wages inc eased, who bene i ed om he inc ease in hou ly wages, and wha happened o wo king hou s and mon hly wages. Ou analysis is based on da a om he Ge man Socio-Economic Panel (SOEP). I allows us, on he one hand, o iden i y eligible wo ke s wi hin a ep esen a i e longi- udinal sample. On he o he hand, i includes de ailed in o ma ion abou employees’ ea nings and wo king hou s, which is i s majo ad an age compa ed o al e na i e da ase s o Ge many. In adminis a i e da a, o example, wo king hou s a e no a ailable h oughou he pe iod o in e es (Bossle and Schank 2020)so heyha e o be app oxima ed— o ins ance, by keeping wo king hou s ixed ac oss ime (Dus - mann e al. 2021) o by impu ing indi iduals’ wo king hou s using he a e age hou s o ull- and pa - ime employees (Ahl eld e al. 2018). The SOEP, in con as , allows us o cons uc and dis inguish wo hou ly wage concep s: an hou ly wage based on con- ac ual wo king hou s and an hou ly wage based on ac ual hou s wo ked. Whe eas con ac ual wo king hou s a e se ou in black and whi e in employmen con ac s, ac ual hou s e lec he e ec i e wo kload o he employed pe son— ha is, con ac- ual hou s plus paid and unpaid o e ime. Since he na ional minimum wage applies o o e ime as well, he ac ual hou ly wage concep is he policy a iable o in e es .2 Me hodologically, ou analysis uses a di e ence-in-di e ences (DiD) eg ession amewo k wi h con inuous ea men (Ca d 1992b). The DiD amewo k builds on egional a ia ion in ea men in ensi y, measu ed by he sha e o eligible employ- ees in 96 di e en egions ( Raumo dnungs egionen , ROR) who we e paid below he minimum wage in a p e- e o m pe iod. E en hough he s a u o y minimum wage is ixed a a uni o m le el ac oss all Ge man egions, he wage s uc u e exhibi s sig- ni ican egional di e ences. Ahl eld e al. (2018) and Caliendo e al. (2018)ha e used his a ia ion o iden i y wage and employmen e ec s on an agg ega e le el. We ake his app oach one s ep u he and apply i o indi iduals. This s ands in con as o he p e ious li e a u e ha iden i ies he e ec o he policy e o m on indi idual 1Changes in egional minimum wages, o ins ance in Sea le, Washing on (Uni ed S a es) by 16 and 18% in 2011 and 2015, espec i ely, a e no able excep ions (Ja dim e al. 2018). 2Un o una ely, in oduc ion o a mo e p ecise ques ion on paid and unpaid ime in he SOEP in 2015 was accompanied by a s uc u al b eak in hese a iables which limi s he analysis o unpaid o e ime. In a p e ious e sion o his pape (see IZA DP No 11246), we discuss hese issues in mo e de ail. 123 The sho - and medium- e m dis ibu ional e ec s o he… 1151 wages, wo king hou s, o ea nings by exploi ing he indi iduals’ p e- e o m wage le els as a quasi-na u al expe imen (Bu auel e al. 2020a,b; Dus mann e al. 2021). He e, employees wi h hou ly wages below (abo e) he minimum wage le el se e as he ea men g oup (con ol g oup). Howe e , po en ial spillo e e ec s (Dickens and Manning 2004b) pu he s able uni ea men alues assump ion (SUTVA) in his ype o iden i ica ion a isk. By using he egional a ia ion in ROR, which a e sepa a ed by economic s uc u e and commu e lows (BBSR 2016), we a e able o inco po a e spillo e e ec s wi hou a ec ing ou iden i ica ion s a egy. This allows us o analyze he e o m’s e ec s on he whole wage dis ibu ion. The esul s o ou analysis can be summa ized as ollows. The desc ip i e e idence indica es ha , in he pos - e o m pe iod (2015–2018), hou ly wages in he bo om wage segmen inc eased a an abo e-a e age a e. Howe e , his inc ease did no ansla e in o highe mon hly wages. Ou DiD analysis con i ms hese esul s and documen s a nega i e e ec on wo king hou s ha p e en ed he minimum wage om a ec ing mon hly wages. We u he compa e ou indi idual-le el esul s o egional- le el esul s (Ahl eld e al. 2018) and conclude ha he posi i e e ec on mon hly wages in he egional-le el es ima ion does no s em om he di ec e ec on mon hly wages o he exposed indi iduals, bu a he om in ensi ied upwa d wage mobili y in he exposed egions. To es he alidi y o ou indings wi h espec o mis epo ing o mon hly wages and wo king hou s, we compa e he wage dis ibu ion in he SOEP wi h he S uc u e o Ea nings Su ey (SES), a epea ed c oss-sec ional su ey o employe s abou hei wo ke s’ wages. He e, we ind ha hey a e e y close. Ou indings complemen he exis ing li e a u e in h ee ways. Fi s , using egional a ia ion in wage le els, we apply a obus me hod o iden i y he dis ibu ional e ec s o he minimum wage e o m on hou ly wages. P e ious s udies using employees’ p e- e o m wage le els as a sou ce o a ia ion ha e been compelled o make s ong assump ions wi h espec o po en ial spillo e -e ec s (Bu auel e al. 2020a,b;Dus - mann e al. 2021). Ou me hod is no cons ained in his espec . We con i m he p e ious esul s and show ha spillo e e ec s o he minimum wage a e modes . Second, by looking a mon hly wages, hou ly wages and wo king hou s, we shed mo e ligh on he indings o Ahl eld e al. (2018), Caliendo e al. (2018), and Bossle and Schank (2020), who look a isola ed a iables only. We a e hus able o pain a comple e pic u e o he key policy a iables by analyzing wages, ea nings, and wo king hou s wi hin a uni ied amewo k. Thi d, we ex end p e ious indings on he dis ibu ional e ec s wi h espec o he ime ho izon. Looking a he e ec s om 2015 o 2018, we show ha he indings o Bu auel e al. (2020a,b) a e no mo ning-a e e ec s only: Changes in wo king hou s dampen he posi i e e ec o he minimum wage e o m on mon hly wages in bo h 2017 and 2018. The documen ed wo king hou s educ ions a e impo an o social policy: They imply ha he hou ly wage inc ease do no ansla e o a ull ex en in highe g oss mon hly wages. Due o income axes, social secu i y con ibu ions and cu s in ans e paymen s, he e ec on households’ disposable income is e en smalle (Backhaus and Mülle 2019; Dube 2019). Fu he mo e, he educ ion in wo king hou s migh be he eason, why he minimum wage did no a ec employmen in Ge many as p edic ed (e.g., Knabe e al. 2014). 123 1152 M. Caliendo e al. The pape p oceeds as ollows. Sec ion 2gi es an o e iew o he ins i u ional de ails and e iews he li e a u e on dis ibu ional e ec s o minimum wages. Sec ion 3in oduces ou da a sou ce and p o ides desc ip i e e idence. Sec ion 4in oduces he iden i ica ion s a egy, discusses i s alidi y, and p o ides esul s o hou ly wages, mon hly wages, and hou s wo ked. Sec ion 5p o ides a ious sensi i i y es s, and Sec ion 6concludes. 2 Minimum wage e o m 2.1 Ins i u ional backg ound On Janua y 1, 2015, a gene al s a u o y minimum wage o e8.50 pe hou became e ec i e in Ge many. I is codi ied in he Minimum Wage Law ( Mindes lohngese z , MiLoG). Be o e i s in oduc ion, he e we e only sec o -speci ic wage loo s se by collec i e ag eemen s. These had been in oduced s a ing in he 1990s in se e al sec o s, including cons uc ion and oo ing (in 1997), hai d essing (in 2013), and secu i y se ices (in 2011).3In addi ion, some ede al s a es ( Lände ) in oduced minimum wages in he amewo k o public p ocu emen law (see Sack and Sa e 2018). In he ollowing, we ocus on egula ions pe aining o he s a u o y minimum wage. Wi h he in oduc ion o he Minimum Wage Law, he Ge man Minimum Wage Commission also ecommended u u e adjus men s o he minimum wage le el. In ligh o he negligible employmen e ec s in he i s yea a e he e o m, he mini- mumwagewas aisedbye0.34pe hou e ec i e Janua y 1, 2017. Theminimumwage was g adually aised o e9.82 in Janua y 2022 (Mindes lohnkommission 2020a). As o Janua y 2015, almos all employed people in Ge many became eligible o he s a u o y g oss minimum wage o e8.50. Some sec o s wi h p e-exis ing sec o - speci ic minimum wages we e exemp o a ansi ional pe iod ending in Janua y 2017. Pe manen exemp ions apply o mino s (pe sons below he age o 18) as well as ainees and in e ns (e.g., s uden s o app en ices comple ing equi ed o elec i e in e nships o up o h ee mon hs). Unemployed people who ha e been egis e ed as such o a leas wel e mon hs may be employed below e8.50 o up o six mon hs. Howe e , om Be ge e al. (2016) show ha his exemp ion is a ely used. The exemp- ion o ainees and mino s, in con as , educes he numbe o eligible indi iduals subs an ially. In 2014, abou 5.5 million employed people ea ned less han e8.50 pe hou , 4 million (72 pe cen ) o whom we e eligible o he minimum wage (Des a is 2016). Th ee g oups o employed people we e o e - ep esen ed in his igu e: Eas Ge man esiden s, people in ma ginal employmen , and women. While he i s g oup esul s om he con inuing s uc u al di e ences be ween Eas and Wes Ge many, as e iden in di e en p ice le els, he hi d g oup esul s om he gende wage gap and he highe p opo ion o pa - ime wo k ca ied ou by women. The second g oup, people in ma ginal employmen , a e no subjec o social secu i y con ibu ions. Thus, 3Mos sec o -speci ic minimum wages we e highe han he s a u o y minimum wage and we e inc eased u he a e he minimum wage e o m (Amlinge e al. 2016). An o e iew o sec o -speci ic minimum wages is gi en in Sch öde (2014). 123 The sho - and medium- e m dis ibu ional e ec s o he… 1153 in many cases, hei g oss income is equal o hei ne income, meaning ha hei wages a e no en i ely compa able wi h egula wages. 2.2 Li e a u e on dis ibu ional e ec s Soon a e he in oduc ion o he minimum wage in 2015, agg ega ed da a began showing ha wageso low-wageg oups—suchasunskilledwo ke s,women,pa - ime employees in small i ms, and employees in Eas Ge many—we e g owing a abo e- a e age a es (Amlinge e al. 2016; Mindes lohnkommission 2016). Howe e , since mul iple ac o s besides he minimum wage may a ec wage g ow h, he obse ed changes a e no necessa ily a ibu able o he minimum wage e o m in a unicausal way. Fo his eason, se e al s udies implemen ed causal designs o iden i y he e ec o he Ge man policy e o m on wages, ea nings and inequali y. In he ollowing, we summa ize hese s udies, o ganized by hei ype o iden i ica ion, and gi e an o e iew o ac o s which can impede he wage e ec s o minimum wages. 2.3 Iden i ica ion using a ia ion be ween employees’wage le els In he absence o a ia ion be ween sec o s,4indi iduals’ wage le els can be used o sepa a ing a con ol and ea men g oup o employees and o implemen ing a quasi-expe imen al se ing. The wage g ow h o indi iduals wi h wages jus abo e he h eshold be o e he e o m is used as con ol g oup o hose below he h eshold, he di ec ly a ec ed. S ewa and Swa ield (2008)aswellasS ewa (2012)make use o his design o iden i y wage and employmen e ec s in he Uni ed Kingdom. Fo he Ge man case, Bu auel e al. (2020b) use he SOEP and apply a di e en- ial end-adjus ed di e ence-in-di e ences model ha compa es hou ly and mon hly wage g ow h o indi iduals below and jus abo e he minimum wage h eshold. They iden i y posi i e e ec s on hou ly and mon hly wages in he i s wo yea s a e he in oduc ion o he minimum wage. In a ollow-up s udy, Bachmann e al. (2020)use he same me hod and ind posi i e bu smalle wage e ec s o 2017. Howe e , he model comes wi h s ingen da a demands since changes in ends (second di e ences) a e analyzed. Sample a i ion is hus a po en ial h ea . Dus mann e al. (2021)also make use o a ia ion by indi idual wage le els and iden i y posi i e wage e ec s in he i s wo yea s a e he e o m. Howe e , hei da a—a combina ion o admin- is a i e da a and he Employee His o y ( Beschä ig enhis o ik ) da a p o ided by he Ins i u e o Employmen Resea ch—include in o ma ion on hou s wo ked up o 2014 only. A e ha poin , hey make he assump ion ha employe s’ wo king hou s a e ixed o e ime. Bu auel e al. (2020a) call his assump ion pa ially in o ques ion: 4The e ec s o sec o -speci ic minimum wages ha e no only been e alua ed by using he a ia ion o indi iduals wage le el (e.g., König and Mölle 2009) o by egions (e.g., om Be ge and F ings 2019). Since he g oup o a ec ed employed is p ecisely de ined, se e al s udies ha e used non-a ec ed bu compa able sec o s o model he coun e ac ual (Fi zenbe ge and Doe 2016). A s udy by F ings (2013) e alua es he e ec o minimum wages on pain e s and elec icians, using he anspo and communica ion indus y and he wholesale and e ail sec o s as con ol g oups. A e z e al. (2013) as well as G ego y and Zie ahn (2022) analyze he minimum wage in he oo ing sec o and use no -a ec ed sub-sec o s o he building indus y as coun e ac ual. In case o he s a u o y uni o m minimum wage, howe e , his app oach is no applicable. 123 1154 M. Caliendo e al. Using he same iden i ica ion s a egy as S ewa and Swa ield (2008) and using he SOEP da a, hey ind nega i e e ec s o he minimum wage e o m on wo king hou s in 2015. Howe e , he e ec s a e no longe signi ican in 2016. S udies o his kind ace he p oblem o po en ial spillo e e ec s—in his case, he possibili y ha employees abo e he minimum wage a e a ec ed by he policy e o m. These e ec s migh a ise, o ins ance, i employe s wan o main ain he wage s uc u e wi hin hei i m. I spillo e e ec s occu , he s able uni ea men alues assump ion (SUTVA) o he DiD app oach is a isk. Es ima o s will be biased and a causal in e p e a ion o he esul s will no be possible. The li e a u e on spillo e e ec s o he Uni ed S a es and he Uni ed Kingdom is inconclusi e (Dickens and Manning 2004a;S ewa 2012; Neuma k and Wasche 2008). Au o e al. (2016) a gues ha he na u e o spillo e s is no ully unde s ood and migh be a leas pa ly a ibu able o e o s in measu ing wages in su ey da a. Fo Ge many, e idence is mixed. Bossle and Schank (2020) as well as Dus mann e al. (2021) ind e idence o spillo e e ec s, while Bu auel e al. (2020b) do no . I hese spillo e e ec s did uly occu , he wage e ec s iden i ied by he li e a u e summa ized abo e a e unde es ima ed. Al e na i e iden i ica ion s a egies will be necessa y o iden i y he causal e ec o he minimum wage co ec ly. 2.3.1 Iden i ica ion using a ia ion be ween egions Regional he e ogenei ies can a ise om legisla i e di e ences in minimum wage se - ings and/o a ia ion in wage s uc u es. Then, a ia ion in he egional “bi e” o he minimum wage can be exploi ed o iden i ica ion (Ca d and K uege 1992;Ca d 1992b; Dube e al. 2010). Lee (1999) and Dol on e al. (2012) exploi egional he e o- genei ies o analyze he wage inequali y in he Uni ed S a es and he Uni ed Kingdom. Fo Ge many, Bossle and Schank (2020) use di e ences in wage le els ac oss he coun y o analyze changes in mon hly wages. Using a decomposi ion, hey show ha he minimum wage was esponsible o 40–60% o he obse ed educ ion in ea nings inequali y be ween 2015 and 2017. Fu he mo e, Ahl eld e al. (2018) and Caliendo e al. (2018) analyze egional wage g ow h as pa o hei analysis. As p oposed by Ca d (1992b), hey i s examine wage e ec s be o e analyzing employmen e ec s. Ahl eld e al. (2018) inds posi i e e ec s a he en h pe cen ile o he wage dis i- bu ion, Caliendo e al. (2018) a educ ion in he a e age ac ion o employed people ea ning less han he minimum wage. Howe e , nei he o he la e s udies look a wage e ec s in mo e de ail. 2.3.2 Nega i e employmen e ec s and non-compliance When minimum wages a e accompanied by job losses, o e all wage inequali y may decline as he popula ion o in e es sh inks a he lowe end o he wage dis ibu ion. Acco dingly, when assessing he e ec o minimum wages on inequali y, (po en ial) nega i e e ec s on disposable income mus be aken in o accoun (Backhaus and Mülle 2019; Neuma k e al. 2004). While he discussion on employmen e ec s is s ill ongoing (e.g., Manning 2021; Neuma k and Shi ley 2021), he exis ing li e a u e 123 The sho - and medium- e m dis ibu ional e ec s o he… 1155 on he Ge man case does no iden i y sho - un employmen losses on he ex ensi e ma gin (Caliendo e al. 2019; Mindes lohnkommission 2020b). Ano he po en ial ac o ha can p e en minimum wages om achie ing hei ull in endedimpac onwagesis non-compliance. I companies iola e he new egula ions, whe he unin en ionally o in en ionally— o ins ance, h ough a lack o imekeeping o he eplacemen o employee posi ions wi h alse sel -employmen – hou ly wages and, hus, mon hly wages will no inc ease as expec ed (B own 1999; Me cal 2008; Mindes lohnkommission 2020b). Using he SOEP, Bu auel e al. (2017) iden i y 1.8 million employees ea ning less han he s a u o y h eshold in 2016. In p inciple, he Ge man cus oms adminis a ion is esponsible o en o cing compliance wi h social secu i y and minimum wage egula ions. In case o iola ions, i can impose ines up o e500,000. Howe e , inspec ions a e ela i ely a e and gene ally ocus on high- isk sec o s such as cons uc ion (Mindes lohnkommission 2020b). 3 Da a and desc ip i e e idence 3.1 Socio-economic panel In he subsequen empi ical analysis, we in es iga e hou ly wages, wo king hou s, and mon hly wages be o e and a e he in oduc ion o he minimum wage e o m. Ou analysis elies on da a om he Ge man Socio-Economic Panel (SOEP), an ongo- ing ep esen a i e longi udinal panel su ey o a ound 30,000 indi iduals in 15,000 households pe yea (see Goebel e al. 2019). We use SOEP e sion 35 and conside indi idual-le el da a om 2012 o 2018 (see SOEP 35 2019). In he ollowing, we desc ibe he co e a iables and he composi ion o ou wo king sample. 3.2 Co e a iables o he empi ical analysis The SOEP ou inely allows o compu a ion o hou ly wages as he a io o g oss mon hly wages and weekly wo king hou s adjus ed by a e age weeks pe mon h and by he in o ma ion on o e ime compensa ion.5Hou s a e s a ed in he SOEP as ac ual and con ac ual weekly wo king hou s. While he la e a e he numbe o hou s de e mined in he employmen con ac , ac ual hou s include o e ime. Thus, hou ly wages can be cons uc ed wi h ei he measu e o wo king hou s, p oducing wo di e en wage concep s: ac ual and con ac ual hou ly wages. Each o hese wage concep s has i s own ad an ages. F om a legal pe spec i e, minimum wage egula ions a e binding o any numbe o wo king hou s, including o e ime. Thus, ac ual wages a e he a ge o minimum wage policy and he ocus o in e es , bu hey a e also he blind spo in adminis a i e da a. A guably, con ac ual wo king hou s a e less a ec ed by measu emen e o s. Fo he sake o b e i y, we p esen e idence based on ac ual hou ly wages. Resul s based on con ac ual hou ly wages p o ide a quali a i ely simila pic u e and a e a ailable om he au ho s upon eques . 5h ps://www.diw.de/de/diw_02.c.222729.de/ins umen e___ elda bei .h ml, las accessed on May 25, 2020. 123 1156 M. Caliendo e al. A key ad an age o he SOEP da a is ha hey con ain hou ly wages, hou s wo ked, and mon hly wages in addi ion o de ailed socio-demog aphic in o ma ion. Adminis- a i e da a usually canno p o ide he same densi y o in o ma ion on a egula basis, especially wi h espec o wo king hou s. Su ey da a a e, howe e , p one o imp eci- sion. I em non- esponse o ounded answe s may bias hou ly wage compu a ions. In Sec . 5, we discuss po en ial sou ces o imp ecise measu emen . In “Online Appendix C”, we p o ide an ex e nal alida ion o he in o ma ion on wages and wo king hou s om he SOEP da a using he S uc u e o Ea nings Su ey (SES) and he Ea nings Su ey (ES), which a e based on companies’ pay oll accoun ing eco ds and one o he mos widely used sou ces o quan i y he labo ea nings dis ibu ion o he Fede al S a is ical O ice. The mos impo an compa ison o ou pu poses is be ween SOEP 2014 and SES 2014, a yea in which bo h su eys we e ep esen a i e. In 2015, he ES was un on a olun a y basis, which esul ed in a subs an ially lowe esponse a es om he pa icipa ing i ms and hei po en ially high selec i i y.6In sum, wages and hou s om SES 2014 and SOEP 2014 a e ela i ely simila , and we do no see any s uc u al di e ences ha could in alida e ou es ima es. The second epo o he Minimum Wage Commission sha es his assessmen (pa ag aph 83 Mindes lohnkom- mission 2018). A he same ime, he dis ibu ions o he same a iables in 2015–2016 in SOEP and ES di e o a g ea e ex en , which may s em om he selec i i y issues o he la e . 3.3 Wo king sample In ou analysis, we ocus on hose employees in Ge many who a e eligible o he minimum wage. Hence, we exclude indi iduals belonging o he g oups and sec o s ha we e ini iallyexemp ed om he e o m.7Fu he ,ou da acon ainonlyemployees aged 18 o olde o whom we ha e wage in o ma ion.8In o de o p e en ou lie s in hou ly wages om biasing ou esul s, we winso ize he da a by se ing he op and bo om 1 pe cen o hou ly wages o he alue o he i s and 99 h pe cen iles, espec i ely. Table 1p esen s he losses in obse a ions o each imposed es ic ion and he esul ing sample sizes o ou wo king sample. Ou empi ical design elies on he p e- e o m le els o egional exposu e o he minimum wage om 2013, which is why ou sample con ains only esponden s o whom we can iden i y hei egional loca ion in 2013. We u he es ic ou sample o ha e only alid in o ma ion on basic socio-demog aphics (gende , age, Ge man ci izenship, p esence o child en in he household below he age o 16, and ma i al s a us). The DiD analysis elies on he wagedis ibu ionso Ge many’splanning egions o in e he egion-speci ic ea men in ensi y. Ge many is subdi ided in o 96 planning egions. The ea men in ensi y is 6Gi en ha adminis a i e da a om he Fede al Employmen O ice do no ea u e in o ma ion on wo king hou s, i is cu en ly no possible o compa e hei dis ibu ion o hou ly wages wi h he SOEP. Mo eo e , hese da a a e no ye a ailable o he b oad esea ch communi y o he pos -2015 pe iod. 7See “Online Appendix B” o he cons uc ion o he co esponding es ic ions. 8The main analysis elies on he wage in o ma ion a ailable di ec ly om he esponden s. In a obus ness check, we addi ionally employ impu ed alues o he missing wages. 123 The sho - and medium- e m dis ibu ional e ec s o he… 1163 Fig. 3 Regional in ensi y o ea men , SOEP 2013. No es: SOEP 35 (Sample 1), au ho s’ calcula ions 4.3 E ec on hou ly wages Table 2p o ides hecoe icien so he ea men e ec ,δ, om he eg essionEqua ion (1).14 As discussed abo e, we epo one coe icien o he i s pos - e o m yea (sho e m, 2015) and one o he h ee subsequen yea s (medium e m, 2016–2018). The uppe panel o Table 2 epo s he esul s o es ima ions o hou ly wages, i s on a e age and hen sepa a ely o indi iduals in speci ic quin iles, ha is, Q1 o Q5. All eg essions include socio-demog aphic con ols as well as ixed e ec s a he le el o indi iduals, su ey yea s, and egional uni s (ROR). The esul s show ha he es ima es o δa he mean a e insigni ican in 2015 bu posi i e in he medium e m, wi h a e age g ow h in 2016–2018 being abou 4%. Focusing on he subg oup a whom he e o m was aimed, employees wi h low wages, we now u n o eg essions by quin iles o he egional wage dis ibu ion. Fo he i s quin ile, we ind highly signi ican and posi i e ea men e ec s o abou 9% in 14 Tables wi h de ails on all es ima ed coe icien s a e a ailable upon eques . 123 1164 M. Caliendo e al. (a) (b) Fig. 4 E olu ion o hou ly wages by yea and egional ea men in ensi y . No es: SOEP 35, au ho s’ calcula ions. Whiske s deno e 95% boo s ap con idence in e als (200 epe i ions) 123 The sho - and medium- e m dis ibu ional e ec s o he… 1165 2015, which inc eases o 21 pe cen in 2016–2018.15 In 2016–2018, we also documen signi ican wage g ow h o 4.8% in he second quin ile. The e a e wo possible easons o he wage g ow h in he second quin ile. Fi s , in some egions, he minimum wage e o m a ec ed no only he i s bu also he second wage quin ile (see Fig.3). Second, he minimum wage e o m may also induce g ow h abo e he di ec ly a ec ed wage segmen ; his phenomenon is known as “spillo e e ec s.” Taken oge he , a e age wages wi hin he bo om quin ile ha e been g owing as e in highly a ec ed a eas, wi h an indica ion o a spillo e e ec on he second quin ile o he egional wage dis ibu ion.16 4.4 E ec on mon hly wages and hou s wo ked The abo e DiD amewo k also allows o es ima ing he e ec o he minimum wage e o m on mon hly wages and hou s wo ked. In he ollowing, we e-es ima e Eq. (1), eplacing he le -hand a iable wi h he loga i hm o mon hly g oss wages and log ac ual hou s wo ked. The middle and he lowes panels o Table 2summa ize he DiD esul s. Fo ac ual hou s wo ked, we ind a nega i e ea men e ec o he bo om quin ile o he hou ly wage dis ibu ion o abou 6% in 2015 and 10% in 2016–2018. This means ha he e o m educed wo king hou s mo e among low-paid employees in highly ea ed egions. Taking hese esul s oge he —an inc ease in hou ly wages and a dec ease in wo king hou s—we ind a posi i e bu s a is ically insigni ican esul o log mon hly wages in he lowes wage segmen s. We conclude ha he dec ease in wo king hou s plays a subs an ial ole in he obse a ion ha he inc ease in hou ly wages does no ansla e in o an equip opo iona e inc ease in mon hly wages (Bu auel e al. 2020a,b). Al hough his wage segmen is no di ec ly a ec ed by he minimum wage e o m, he e ec o he educ ion in wo king hou s is also obse ed in he ou h quin ile o he wage dis ibu ion in bo h 2015 and 2016–2018. Fo mon hly wages, we obse e a nega i e e ec in he ou h quin ile only in 2015. One plausible explana ion o he nega i e ea men e ec is ha employe s educe hou s wo ked in an a emp o educe labo cos s a he in ensi e ma gin. Ano he possible explana ion is ha employees lowe hei own labo supply in esponse o he e o m in o de o no exceed ea nings limi s ha secu e access o speci ic social secu i y o ax ad an ages. The po en ial unde lying mechanism o spillo e s o minimum wage egula ions on high-skilled wo ke s is desc ibed in G ego y and Zie ahn (2022). 4.5 Placebo eg essions This sec ion add esses he alidi y o he common end assump ion o hou ly wages, mon hly wages, and wo king hou s. The g aphical ep esen a ions in Sec . 4.2 p o ide 15 A coe icien o 0.085 means ha , in a egion wi h he a e age ea men in ensi y (no malized o be 1.0), wages in he i s quin ile g ew by 8.5%. 16 We also un he es ima ion wi h he ea men in ensi y de ined by he con ac ual hou ly wage and ob ained quali a i ely simila esul s. These a e a ailable om he au ho s upon eques . 123 1166 M. Caliendo e al. Table 2 E ec on hou ly wages, g oss mon hly wages and wo king hou s, 2015–2018 e sus 2014 Pos -Re o m Placebo 2015 2016–2018 2014 Coe p- alue Coe p- alue Coe p- alue Hou ly wages Mean 0.012 0.259 0.036*** 0.003 0.006 0.643 By quin iles Q1 0.085** 0.044 0.206*** 0.000 −0.005 0.936 Q2 0.009 0.689 0.048* 0.069 0.017 0.589 Q3 0.003 0.857 0.006 0.774 −0.004 0.837 Q4 −0.021 0.197 0.008 0.627 −0.001 0.967 Q5 −0.002 0.903 −0.006 0.769 −0.009 0.716 Mon hly wages Mean −0.008 0.496 0.008 0.582 −0.005 0.703 By quin iles Q1 0.015 0.720 0.091 0.116 0.000 0.994 Q2 −0.005 0.862 0.041 0.221 −0.018 0.547 Q3 0.004 0.867 −0.016 0.529 −0.006 0.750 Q4 −0.046*** 0.006 −0.024 0.264 −0.006 0.749 Q5 −0.008 0.543 −0.007 0.738 −0.022 0.495 Ac ual hou s Mean −0.013 0.161 −0.023** 0.044 −0.009 0.435 By quin iles Q1 −0.056* 0.096 −0.099** 0.032 0.004 0.924 Q2 −0.007 0.767 0.002 0.933 −0.025 0.445 Q3 0.000 0.979 −0.022 0.260 0.006 0.783 Q4 −0.025* 0.053 −0.041** 0.016 −0.007 0.694 Q5 0.004 0.805 0.008 0.715 −0.010 0.681 SOEP 35 (Sample 2), own calcula ions. The ables con ains esul s o wo eg ession speci ica ions: Fi s , one DiD es ima ion o he pos - e o m pe iod, wi h sepa a e es ima ion o he i s pos - e o m yea 2015 and he h ee subsequen yea s 2016– 2018. Second, a placebo es ima ion o he p e- e o m pe iod. All speci ica ions con ain indi idual, yea and egional ixed e ec s. All speci ica ions also include con ols o age, Ge man ci izenship, p esence o child en aged below 16 in he household, ma i al s a us, wo-pe iods lagged egional GDP. S anda d e o s a e obus . Signi icance le els: *p<0.1, **p<0.05, ***p<0.01 suppo i e desc ip i e e idence, bu do no ep oduce he DiD eg ession design in de ail. He e, we explici ly check he common end assump ion by means o placebo eg essions. The minimum wage was in oduced on Janua y 1, 2015, and, o his eason, he ea men e ec s abo e a e de i ed by compa ing he wage dis ibu ions om 2014 and 2015–2018 using egional bi es om 2013. In he placebo eg essions, we es ima e a model wi h lagged a iables, ha is, wi h wage dis ibu ions om 2013 123 The sho - and medium- e m dis ibu ional e ec s o he… 1167 and 2014 and he egional bi e om 2012. I he common end assump ion holds, we would expec o ind no ea men e ec in his placebo eg ession. The eg ession equa ion akes he o m, log (yi )=δP2014 ×Bi e2012 +μXi +e +αi+ν +i ,(2) wi h yi deno ing one o he h ee a o emen ioned dependen a iables and ∈ (2013,2014). In sum, he placebo es ima ions (see he wo igh columns o Table 2) gi e e idence suppo ing he pa allel end assump ion, hus lending c edibili y o ou iden i ica ion s a egy. The esul s sugges ha no sys ema ic an icipa ion o he e o m ook place. Mos impo an ly, we de ec nei he g ow h in hou ly wages no a dec ease in wo king hou s ha co ela es wi h he iden i ying a ia ion in ea men exposu e p io o he e o m. 4.6 Dynamic ea men In o de o illus a e how he e o m’s e ec un olds du ing he pos - e o m pe iod and o jus i y ou esul s di e en ia ing be ween 2015 and he subsequen pos - e o m yea s, we adjus he Eq. (1) o include yea dummies and espec i e in e ac ion e ms o he yea s 2015–2018. log(wi )= 2018  y=2015 δyPy×Bi e2013 +μXi +αi+ν +e +i , ∈(2014,2018). (3) As in he main speci ica ion, we conduc sepa a e es ima ions by quin iles o he egional dis ibu ion o hou ly wages in 2013. Figu e 5p esen s he coe icien s δy, y∈(2015,2018) o he i s and second quin iles, whe e he main e ec o he e o m is de ec ed, and o he h ee dependen a iables—log hou ly wage, log mon hly wages, and log wo king hou s. The igu e co obo a es he main esul s o he bo om quin ile, he wage segmen ha is he ocus o he e o m: In all pos - e o m yea s, we obse e an inc ease in hou ly wages, a dec ease in wo king hou s, and unchanging mon hly wages. Fo he second quin ile, he e ec s on all h ee dependen a iables slowly un old and a e he highes in 2018. 4.7 Uncondi ional quan ile eg ession In o de o unde line he di e ence be ween he pe spec i e o indi idual-le el wage de elopmen due o he minimum wage e o m and he e o m’s e ec on he o e all dis ibu ion o wages, we adap he empi ical app oach o Dube (2019) and conduc addi ional es ima ions in he spi i o Eq. (1) by he means o an uncondi ional quan ile eg ession (Fi po e al. 2009) o 2015–2018. Table 3documen s he e o m’s e ec on he dis ibu ion o hou ly wages (uppe panel) and mon hly wages (lowe panel) on he 123 1168 M. Caliendo e al. Fig. 5 Dynamic ea men e ec . No es: SOEP 35, au ho s’ calcula ions 123 The sho - and medium- e m dis ibu ional e ec s o he… 1169 Table 3 Uncondi ional quan ile eg ession (2014–2018) Q10 Q20 Q30 Q40 Q50 Q60 Q70 Q80 Log hou ly wages 0.157*** 0.079*** 0.035 −0.005 −0.014 −0.017 −0.009 0.010 (0.032) (0.028) (0.025) (0.019) (0.019) (0.019) (0.017) (0.020) Log mon hly wages −0.213 0.061 0.048 0.032 0.005 −0.015 −0.015 0.003 (0.131) (0.046) (0.036) (0.031) (0.023) (0.021) (0.022) (0.023) SOEP 35, au ho s’ calcula ions. Signi icance le els: *p<0.1, **p<0.05, ***p<0.01 Speci ica ionsinclude con ols o egionala e ages o gende ,age, Ge manci izenship, p esenceo child en aged below 16 in he household, ma i al s a us, wo-pe iod lagged egional GDP, and egional ixed e ec s quan iles o he espec i e dis ibu ions. We de ec an e ec o he e o m o 16 pe cen on he 10 h pe cen ile and 8 pe cen on he 20 h pe cen ile o hou ly wages, which is gene ally in line wi h ou indi idual-le el es ima es. Fo mon hly wages, we do no ind a s a is ically signi ican e ec . In gene al, his inding co obo a es he p e ious obse a ion ha he minimum wage e o m is inducing g ow h in hou ly wages o low-wage ea ne s, bu no imp o ing hei mon hly wages (see Fig. 2). This e idence unde lines ha he e ec o minimum wages on mon hly wages is highly dependen on he numbe o hou s ha he a ec ed people wo k (Bossle and Schank 2020). Compa ing hese esul s o he indi idual-le el, we can conclude ha he posi i e e ec on mon hly wages (sugges ed by Fig. 2a) is d i en by minimum-wage wo ke s wi h longe wo king hou s and who a e less likely o expe ience a educ ion o hei wo king ime. 5 Robus ness checks Due o i s panel cha ac e and he weal h o in o ma ion i con ains on socio- demog aphics and job cha ac e is ics, he SOEP o e s se e al ad an ages o e o he da ase s. Howe e , he e may be measu emen e o in epo ed mon hly wages and wo king hou s. Fo ou es ima ion design and esul s, a egionally equip opo iona e bias is innocuous. I he e o is andomly dis ibu ed, a ia ion inc eases, which solely diminishes he s anda d de ia ion o he coe icien s. Measu emen e o is p oblem- a ic i i is sys ema ic and co ela ed wi h he ea men in ensi y. In o de o e alua e whe he hese and o he conce ns a ec ou esul s, we pe o m a se ies o obus ness checks. 5.1 Pooled model wi h and wi hou indi idual ixed e ec s Fo compa abili y o ou esul s wi h o he s udies, we e-es ima e he DiD model wi hou di e en ia ing be ween sho - and medium- e m coe icien s. The uppe panel o Table A.3 p o ides coe icien s on he pooled e ec o 2015–2018 wi h indi idual- le el ixed e ec s (le ) and wi hou hem ( igh ). I documen s ha he FE es ima es a e highe in he i s quin ile, which poin s a he exis ing nega i e co ela ion be ween 123 1170 M. Caliendo e al. wage g ow h and ime-in a ian cha ac e is ics o he eligible employees in his wage segmen . A he same ime, we do no ind subs an ial s uc u al di e ences be ween he OLS and he FE es ima ions. 5.2 I em non- esponse No all esponden s p o ide answe s o all SOEP ques ions. I non- esponse is sys em- a ic, his c ea es ano he elemen o unce ain y.17 Fo conce ns like his, some missing alues a e s a is ically impu ed. This is also he case o mon hly g oss wages bu no o wo king hou s. Because impu a ions a e accompanied by addi ional unce ain y, ollowing Au o e al. (2016), we disca d obse a ions wi h impu ed mon hly wages in ou main esul s. In o de o check whe he inclusion o he impu ed alues a ec s he esul s, we edo he DiD es ima ion o hou lywages,in eg a ing hep e iouslydisca ded impu ed wages. The second panel o Table A.3 summa izes he esul s. The able econ i ms ou p e ious e idence. The e o m has a small e ec on he mean wage: he s onges posi i e e ec in he i s quin ile, a posi i e e ec in he second quin ile, and no e ec in he uppe quin iles. 5.3 Inclusion on non-exposed economic sec o s As p esen ed in Table 1, ou wo king sample does no con ain he economic sec o s ha we e empo a ily exemp ed om he minimum wage egula ion. These sec o s had o adjus hei sec o -speci ic minimum wage le els o he na ional h eshold wi hin wo yea s, which makes he wage adjus men o all employed people a ele an ques ion. The hi d panel o Table A.3 shows he es ima ion coe icien s based on he sample including all economic sec o s—bo h hose exposed and hose no exposed o he minimum wage egula ion. Fo his ex ended sample, we ind p onounced wage g ow h in hou ly wages in he i s and second quin iles. 5.4 Al e na i e bi e measu e The analysis p esen ed he e elies on a bi e measu e cons uc ed om he SOEP da a. In some egions, numbe s o obse a ions in SOEP a e small, calling he alidi y o he de i ed bi e in o ques ion. This is a h ea o ou iden i ica ion s a egy, pa icula ly i he measu emen e o is sys ema ic. Fo his eason, we e- un ou analysis using a bi e cons uc ed om la ge-scale company da a, he S uc u e o Ea nings Su ey (SES).18 Un o una ely, he da a a e a ailable only o 2014, bu no o ou p e e ed pe iod, 2013, meaning ha an icipa o y e ec s may al eady ha e aken e ec on he bi e. Howe e , Caliendo e al. (2018) show no an icipa ion e ec s on wages. The ou h panel o Table A.3 eplica es he main analyses o hou ly wages, excep ha ins ead 17 Fo he s uc u e o i em non- esponse on income and weal h ques ions in SOEP, see F ick and G abka (2005); F ick e al. (2007). 18 Sou ce: FDZ de S a is ischen Äm e des Bundes und de Lände , Ve diens s uk u e hebung, 2014. 123 The sho - and medium- e m dis ibu ional e ec s o he… 1171 o he SOEP-based bi e o 2013 we use he SES-based bi e o 2014. In gene al, he esul s using he SES-based bi e con i m he p esence o signi ican wage g ow h in he i s quin ile. Howe e , he magni ude o he coe icien is lowe , and no e ec on he second quin ile is de ec ed. 5.5 Regional-le el agg ega ion Ou main esul s ely on indi idual-le el da a combined wi h he egional-le el bi e in o ma ion. We belie e he indi idual-le el measu emen o he ou come a iable is app op ia e as i enables us o con ol o egional di e ences in employee cha - ac e is ics while also conside ing wi hin-subjec changes o e ime in cha ac e is ics ha a e ele an o he ou come. In o de o compa e hese mixed-le el esul s o a pu e egional-le el es ima ion (as conduc ed by Ca d 1992a), we e-es ima e Eq. 1 a he le el o egional uni s. The dependen a iable is mean wages in a pa icula yea measu ed in he o e all egional dis ibu ion o wi hin egion-speci ic quin iles. Resul s a e consis en wi h he indi idual-le el es ima ion: The e is a sizable e ec on wages in he i s quin ile (see he lowes panel o Table A.3). Howe e , he magni ude o he e ec is lowe and he e is no signi ican spillo e e ec on he second quin ile. 6 Conclusion This pape assesses he sho - and medium- e m e ec s o Ge many’s minimum wage e o m on he dis ibu ion o hou ly wages, mon hly wages, and wo king hou s. In Janua y 2015, Ge many in oduced a high s a u o y minimum wage wi h only a ew legal exemp ions. The new minimum wage was se a e8.50, which exceeded he hou ly wages o mo e han 10% o all eligible employees in 2014. Two yea s la e , he minimum wage le el was inc eased o e8.84. We analyze he implica ions o his majo labo ma ke in e en ion, which is o in e es bo h o he Ge man con ex and o o he de eloped coun ies. Wi h espec o he e o m’s main goal, which was o inc ease hou ly wages a he bo om o he wage dis ibu ion, ou empi ical da a sugges ha i was e ec i e. In he low-wage segmen , he desc ip i e analyses show an accele a ion o wage g ow h s a - ing in he i s yea ollowing he e o m’s implemen a ion. A di e ence-in-di e ences analysis elying on he egional a ia ion in ea men in ensi y as a sou ce o iden i- ica ion co obo a es his e idence. We ind sizable posi i e ea men e ec s o he bo om quin ile o he egion-speci ic wage dis ibu ions: In he i s pos - e o m yea (2015), he a e age wage g ow h in he i s quin ile was abou 9% and inc eased o 21% in subsequen yea s, 2016–2018. Howe e , mos o he a ec ed employees simul aneously expe ienced a educ ion in wo king hou s. Fo his eason, we canno iden i y a s a is ically signi ican imp o emen in mon hly wages o minimum-wage employees. This esul echoes indings om s udies ca ied ou in he Uni ed S a es (Neuma k e al. 2004), bu is somewha con a y o indings om s udies using es ima- ion designs based on egional-le el dis ibu ions o hou ly wages and mon hly wages, which show an inc ease a he low end o bo h dis ibu ions (Ahl eld e al. 2018). Ou 123 1172 M. Caliendo e al. esul s imply ha he e ec o he minimum wage e o m on exposed indi iduals may be di e en om he e ec on exposed egions. Da a agg ega ion a he egional le el dis ega ds he indi idual loca ion o exposed popula ions in he dis ibu ion o mon hly wages and he change in hese esul s should be seen as complemen a y in his espec . Ou analysis p o ides e idence ha —because o downwa d adjus men s in wo king hou s—an hou ly minimum wage does no necessa ily imp o e mon hly wages o employees wi h low pay. I he policy goal o a minimum wage is he p e en ion o po e y, ou esul s imply ha no only he in e ac ion wi h he social secu i y sys em and i s means- es ed income suppo mus be conside ed (Mülle and S eine 2011). Changes in wo king ime can also con adic wi h he poli ical in en ions. Simul aneously, his s esses he impo ance o he in ensi e ma gin when employmen e ec s a e analyzed: Employe s seem o use changes in wo king hou s as a means o cope wi h inc easing labo cos s wi hou educing jobs. This migh explain why e ec s on he ex ensi e ma gin ha e no been iden i ied in Ge many so a . Supplemen a y In o ma ion The online e sion con ains supplemen a y ma e ial a ailable a h ps://doi. o g/10.1007/s00181-022-02288-4. Funding Open Access unding enabled and o ganized by P ojek DEAL. Decla a ion Con lic o in e es This esea ch was unded by he Leibniz Associa ion, g an numbe SAW-2015-SOEP- 5. Open Access Thisa icleislicensedunde aC ea i eCommonsA ibu ion4.0In e na ionalLicense, which pe mi s use, sha ing, adap a ion, dis ibu ion and ep oduc ion in any medium o o ma , as long as you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons licence, and indica e i changes we e made. The images o o he hi d pa y ma e ial in his a icle a e included in he a icle’s C ea i e Commons licence, unless indica ed o he wise in a c edi line o he ma e ial. I ma e ial is no included in he a icle’s C ea i e Commons licence and you in ended use is no pe mi ed by s a u o y egula ion o exceeds he pe mi ed use, you will need o ob ain pe mission di ec ly om he copy igh holde . To iew a copy o his licence, isi h p://c ea i ecommons.o g/licenses/by/4.0/. Re e ences Ahl eld GM, Ro h D, Seidel T (2018) The egional e ec s o Ge many’s na ional minimum wage. Econ Le 172:127–130 Amlinge M, Bispinck R, Schul en T (2016) Ein Jah Mindes lohn in Deu schland—E ah ungen und Pe spek i en. WSI-Repo 28 A e zB,A n z M,G ego y T(2013)The minimumwagea ec s hemall:e idence onemploymen spillo e s in he oo ing sec o . Ge man Econ Re 14(3):282–315 Au o DH, Manning A, Smi h CL (2016) The con ibu ion o he minimum wage o US wage inequali y o e h ee decades: a eassessmen . Am Econ J Appl Econ 8(1):58–99 Bachmann R, Bonin H, Boockmann B, Demi G, Felde R, Ispho ding I, Kalwei R, Laub N, Vonnahme C, Zimpelmann C (2020) Auswi kungen des gese zlichen Mindes lohns au Löhne und A bei szei en, S udie im Au ag de Mindes lohnkommission Backhaus T, Mülle K-U (2019) Does he Ge man minimum wage help low income households? E idence om obse ed ou comes and he simula ion o po en ial e ec s, DIW Discussion Pape s, p 1805 123