The dynamic welfare habitus and its impact on Brazilian migration to Lisbon and Barcelona

Geo o um 120 (2021) 58–66 A ailable online 4 Feb ua y 2021 0016-7185/© 2021 The Au ho s. Published by Else ie L d. This is an open access a icle unde he CC BY license (h p://c ea i ecommons.o g/licenses/by/4.0/). The dynamic wel a e habi us and i s impac on B azilian mig a ion o Lisbon and Ba celona Dominique Joli e a , * , Sonia Pe ei a b a Uni e si y o Ams e dam, Facul y o Social and Beha iou al Sciences, P og amme G oup Poli ical Sociology: Powe , Place and Di e ence, Ne he lands b Ins i u e o Geog aphy and Spa ial Planning (IGOT), Cen e o Geog aphical S udies (CEG), Uni e sidade de Lisboa, Lisboa, Po ugal ARTICLE INFO Keywo ds: Mig a ion d i e s Sending coun y Social policy T ansna ional social p o ec ion Wel a e magne ABSTRACT Li le is known on how people’s way o hinking and doing a ound wel a e p o ision – wha we call he wel a e habi us – plays a ole in mig a ion and how such cul u al e e ences change o e he mig a ion p ocess. Th ough an empi ical case s udy on B azilian mig a ion o Sou he n Eu ope, his a icle explo es he dynamism o he wel a e habi us ocusing on h ee elemen s. Fi s , he wel a e- ela ed esou ce en i onmen in he coun ies o o igin and des ina ion. Second, he ole o he wel a e habi us in shaping mig a ion aspi a ions. Thi d, he ans o ma i e lea ning p ocess aking place du ing he li e cou se and he mig a ion expe ience. The a icle d aws on a li e a u e e iew, he analysis o seconda y quan i a i e da a and 24 in-dep h semi-s uc u ed quali a i e in e iews collec ed in 2016 in Lisbon and Ba celona among men and women bo n in B azil selec ed a di e en s ages o he li e cou se –young people wi hou child en, pa en s o school-aged child en and people aged 55 o mo e. D awing on p ac ice heo y he pape looks a he in e ac ion be ween ex e nal s uc u es, habi us and e e yday ac ions a ound wel a e p o ision in con ex s o mig a ion. Doing so, he pape con ibu es o a be e unde s anding o he e ec o wel a e p o isions in he coun y o o igin on mig a ion ocusing on he empo al pe spec i e. 1. In oduc ion Jo˜ ao was bo n in he B azilian s a e o Goi´ as. When he was 45 yea s old, he wo ied abou how he and his wi e Ma ia would cope wi h he loss o income once hey e i ed. A ha ime, hey lacked he job con- inui y hey needed o con ibu e egula ly o hei pension scheme in o de o e i e in good inancial condi ions a he age o 65. Being close o and aking ca e o amily membe s was an impo an alue o hem bu , when he possibili y a ose, his couple mig a ed o wo k in Spain and con ibu e wi h hei ea nings o hei B azilian pension scheme. Zuleida, om he s a e o Minas Ge ais, mig a ed o Po ugal o ca e o he son who had mig a ed ea lie . He son was going h ough a dep ession and she wen o Lisbon o assis him, complemen ing he “ o mal” suppo ha he was ecei ing om heal h p o essionals. She hough i would be a sho - e m pe iod ab oad. Once in Po ugal, she liked he coun y and he secu i y she el he e. She also began o app ecia e he heal hca e ha she had access o in he Po uguese public sys em. The e o e, she se led in Lisbon, whe e she wo ked o comple- men he e i emen pension she ecei ed in B azil. These wo cases highligh mul iple ways in which wel a e p o isions in o igin and des ina ion migh d i e mig a ion beyond he s ic ac ha social bene i s p o ided by he s a e migh be mo e gene ous in he des ina ion coun y – he so-called wel a e magne hypo hesis (Bo jas, 1999). Mo ing beyond his limi ed concep ualisa ion o wel a e p o- isions, a g owing body o li e a u e on ansna ional social p o ec ion has ocused on he di e se pools o esou ces ha mig an s and hei amily membe s d aw on globally in ansna ional ields o p o ec and p omo e hei well-being (Bilecen & Ba glowski, 2015; Fais e al., 2015; La leu & Lizin, 2016; Le i e al., 2016; Mazzuca o, 2009). The ansna ional social p o ec ion amewo k cap u es b oade wel a e p o ision needs and esou ces in con ex s o mig a ion, aking in o ac- coun he wide ange o o mal and in o mal p i a e ac o s ha , aside om he s a e, p o ide wel a e in di e en domains such as heal h, educa ion, unemploymen , and e i emen (Bilecen & Ba glowski, 2015; Le i e al., 2016). Wi hin his esea ch agenda, some s udies ha e explo ed changes and di e ences in mig an s’ needs and p io i ies in wel a e p o isions depending on hei li e s age o hei ole in he household (Joli e , 2019; Se a Mingo & Mazzuca o, 2019), howe e mo e esea ch is needed o be e unde s and he cons an change and adap a ion in he * Co esponding au ho . E-mail add ess: [email p o ec ed] (D. Joli e ). Con en s lis s a ailable a ScienceDi ec Geo o um jou nal homepage: www.else ie .com/loca e/geo o um h ps://doi.o g/10.1016/j.geo o um.2021.01.017 Recei ed 13 Sep embe 2019; Recei ed in e ised o m 4 Oc obe 2020; Accep ed 15 Janua y 2021 Geo o um 120 (2021) 58–66 59 ways o hinking and doing a ound wel a e p o ision, wha we call he wel a e habi us, h oughou he mig a ion p ocess. The cases o Jo˜ ao and Zuleida illus a e ha mig a o y p ojec s a e dynamic and change acco ding o he con ex , indi idual and collec i e s a egies, and wi h he in o ma ion and knowledge ha mig an s accumula e o e ime and ac oss space (Boye , 2005: 49-53). As such, how does he wel a e habi us based on he con ex in he coun y o o igin shape concep ualiza ions o wel a e p o ision and how hese a ec mig a ion p ojec s? Beyond he limi ed s udies looking a he e ec o wel a e p o isions by he s a e in mig a ion and immobili y (Ku eko ´ a, 2013; Mahend a, 2014), he o igin coun y e ec s o b oade wel a e p o isions in con ex s o mig a ion emain o e looked. Changes in he o igin coun y e ec s o e he mig a ion p ocess a e also unexplo ed – Wha is he e ec o pe - cep ions o he wel a e egime o he o igin coun y in p e e ences and decisions a ound wel a e p o ision upon a i al and wha changes wi h he con ac wi h new eali ies? To answe hese ques ions, using a simila app oach o he one p oposed by Raghu am (2016) o unde s and he e hics o ca e, his a icle explo es he wel a e habi us and i s ans o ma ions o e ime and wi h he mig a ion p ocess. To do so, i analyses how concep uali- za ions o wel a e p o ision a e shaped by geog aphical and his o ical ac o s and he changing cha ac e o expec a ions and meanings a ound wel a e p o ision. I also explo es how his changing wel a e habi us shapes e e yday ac ions a ound wel a e p o ision. To eally g asp he e ec o wel a e p o isions in mig a ion we conside “ he en i e se o ins i u ional a angemen s, policies and p ac ices a ec ing wel a e ou comes and s a i ica ion e ec s in di e se social and cul u al con- ex s” (Gough, 2004: 26). Taking his b oade unde s anding o wel a e egimes, we explo e he ole ha he di e en geog aphies and di- mensions in wel a e p o ision play in mig a ion and examine changes o e ime. Ou main ocus lies on dis inc ions ha mig an s make be- ween wel a e esou ces p o ided by he s a e and hose p o ided by o mal p i a e ins i u ions such as p i a e schools and uni e si ies, o p i a e heal h insu ance companies. We also conside mig an s’ ways o hinking and doing when o mal ins i u ional wel a e esou ces a e pe cei ed as insu icien , absen , o inaccessible. We b ing in o he analysis he dynamic in e play be ween pas , p esen , and u u e in people’s conside a ions a ound wel a e p o ision a di e en s ages o he mig a ion p ocess. To do so, we d aw on Emi baye and Mische (1998, p. 963), who p oposed o concep ualise agency as a empo ally embedded p ocess o social engagemen . We pay pa icula a en ion o he ole o ex e nal s uc u es in he coun y o o igin in shaping he wel a e habi us and he lea ning p ocess ha akes place h ough he li e cou se as mig an s mo e om he coun y o o igin o coun ies o des ina ion, whe e hey become acquain ed wi h and begin o na iga e new wel a e egimes. The a icle d aws on p ac ice heo y applied o he s udy o mig a ion (O’Reilly, 2012) o look a he ans o ma ions o cul u al e e ences a ound wel a e p o ision o e he mig a ion p ocess. I ocuses on an explo a o y empi ical case s udy on B azilian mig a ion o Sou he n Eu ope h ough 24 in-dep h semi-s uc u ed quali a i e in e iews applying a li e his o y app oach. The in e iews we e collec ed in 2016 in Lisbon and Ba celona among men and women bo n in B azil selec ed a di e en s ages o he li e cou se –young people wi hou child en, pa en s o school-aged child en and people aged 55 o mo e. The a icle en iches he g owing li e a u e on ansna ional social p o ec ion ha ocuses on he p o ec ion and p o ision o social wel a e in con ex s o mig a ion bu ends o neglec he s udy o wel a e p o- ision as a d i e o mig a ion. Ou analysis also en iches he sca ce li e a u e on he ole o wel a e egimes in he coun ies o o igin in mig a ion om he mic o-le el pe spec i e. Finally, we con ibu e o he s udy o he li e cou se e ec s in wel a e and mig a ion looking a he ways in which li e e en s (including mig a ion) and s ages in he li e cou se in luence wel a e p ac ices in con ex s o mig a ion. Se en sec ions ollow his in oduc ion. The nex sec ion explains he heo e ical amewo k used o explo e he dynamism o he wel a e habi us o e he mig a ion expe ience. The second sec ion desc ibes he da a and me hods used o ou analysis. The hi d sec ion p esen s he con ex o B azilian mig a ion o Ba celona and Lisbon. The ou h sec ion explo es how oppo uni ies and cons ain s in a pa icula wel- a e egime shape he wel a e habi us. The i h sec ion del es in o di e en ways in which he wel a e habi us acqui ed in he coun y o o igin is aken in o accoun in ini ial mig a ion p ojec s. The las sec ion be o e he conclusion explo es mig an s’ decisions a ound wel a e p o- ision a e a i al. The sec ion highligh s ha , simila ly o ca e p ac- ices (Raghu am, 2016), wel a e p ac ices a ound wel a e p o ision a e si e-speci ic, complex and dynamic. 2. Ways o wa d o unde s and he e ec o wel a e in mig a ion To o e come he iden i ied gaps, ou s udy app oaches mig a ion as a p ocess shaped by he cons an ly changing in e play be ween s uc- u al ac o s and people’s agency and choice (Mo awska, 2009; O’Reilly, 2012; S ones, 2012) and explo es how mig a ion is a li e e en ha , h ough i s inhe en lea ning p ocess, a ec s and ans o ms he wel a e habi us o e ime. D awing on O’Reilly (2012), we look a he in e play be ween ex e nal s uc u es and habi us in con ex s o mig a ion. The ex e nal s uc u es unde s udy a e he public and p i a e wel a e e- sou ces ha people ha e access o in a pa icula wel a e egime. To s ay aligned o p e ious s udies on ansna ional social p o ec ion, we call hem he wel a e esou ce en i onmen . We limi ou analysis o he esou ce en i onmen in he domains o heal h, educa ion, unemploy- men , and e i emen and ake in o accoun he unequal access o p i a e and public wel a e esou ces o mo e o less a luen people. We also d aw on O’Reilly (2012) o concep ualise he wel a e habi us. I e e s o he ways o doing and hinking in e ms o wel a e p o ision ha a e acqui ed h ough pas collec i e and indi idual expe iences and ac ions. The wel a e habi us includes cul u al schemas, knowledge, and wo ld- iews (O’Reilly, 2012: 150-51). The wel a e habi us shapes con- cep ualisa ions and expec a ions o he wel a e egime. I has also an e ec on e e yday ac ions a ound wel a e p o ision. To Bou dieu’s idea ha changes in people’s habi us ake place a a a he slow pace, we p e e F iedmann’s a gumen ha mig a ion and he exposi ion o di e en expe iences and ideas can ans o m he habi us (F iedmann, 2005, pp. 302-303), some imes d as ically. As such, we conside ha he wel a e habi us is cons an ly changing and adap ing o he e ol ing oppo uni ies and cons ain s (O’Reilly, 2012) encoun e ed h oughou he mig a ion p ocess. Be o e mig a ion, a di e en ia ed indi idual wel a e habi us emana es om people’s li ed expe iences o he wel a e esou ce en i onmen in he o igin coun y ha may be dependen on people’s socio-economic backg ound. An example o wel a e habi us in B azil is he widesp ead p ac ice o ca ing o he elde ly wi hin he amily a he han using nu sing homes. This p e e ence migh be due o expec a ions and obliga ions o in o mal social p o ec ion (Fais , 2000; Fais e al., 2015; Sienkiewicz e al., 2015). Ano he explana ion could be he high cos s associa ed wi h ex e nal elde ly ca e se ices in B azil ha ende hem inaccessible o la ge segmen s o he popula ion. We also use he concep o ac ion o e e o conc e e ac ions and decisions a ound wel a e p o ision. Pa icula ac ions can ep esen a u ning poin in he ans o ma ions o he wel a e habi us. We bo ow om Emi baye and Mische (1998) he idea ha ac ion is in luenced by he dynamic cha ac e o agency, in o med by pas expe iences (wel a e habi us), in luenced by (pe cep ions o ) he p esen si ua ion in cons an change, and o ien ed owa ds he u u e (Fig. 1). The si ua ion in he p esen includes, on he one hand, he (pe cei ed) wel a e esou ce en i onmen in he coun ies o o igin and des ina ion. In ou empi ical example, his e e s o he o mal (s a e as well as p i a e) componen o wel a e p o isions in B azil, Po ugal and Spain, including he access o public wel a e p o isions h ough social secu i y ag eemen s be ween B azil and he wo Eu opean coun ies. I also encompasses, he in o mal esou ces a ailable o mig an s wi hin social ne wo ks, including mig an communi ies in Ba celona and D. Joli e and S. Pe ei a Geo o um 120 (2021) 58–66 60 Lisbon. The his o y o mig a ion ha links B azil and he wo Sou he n Eu opean ci ies is also likely o play a ole – he mo e o less es ablished social ne wo ks in Ba celona and Lisbon shape oppo uni ies and con- s ain s in he access o in o mal wel a e esou ces. On he o he hand, he p esen con ex is also shaped by li e cou se e en s such as he bi h o a child o eaching he age o e i emen . Finally, s ill d awing on Emi baye and Mische (1998), wel a e- ela ed ac ions a e o ien ed o- wa ds pe cep ions and iden i ica ions o wel a e esou ces ele an o he u u e in ela ion o mig an s’ e ol ing needs and aspi a ions. Such o ien a ions shape possible ajec o ies o ac ion. Paying mo e a en ion o he weigh o each o hese h ee empo al dimensions a di e en s ages o he mig a ion p ocess is use ul o cap u e o wha ex en ames o e e ence based on he access o wel a e esou ce en i onmen in di e en wel a e egimes shi o e ime. Wi h he iden i ica ion o di e en wel a e egimes o e he mig a ion p o- cess, mig an s migh ques ion hei wel a e habi us, and inco po a e, ep oduce and ans o m local p ac ices o he coun y o des ina ion – see Nowicka (2018, p. 825) on he ans o ma ion o acial discou ses wi h mig a ion. As such, h ough his ans o ma i e p ocess, o en media ed by social ne wo ks, mig an s econ igu e hei wel a e habi us. The nex sec ion desc ibes he da a we use in he analysis. 3. Da a and me hods The s udy emb aces he social cons uc i ism p inciple acco ding o which eali y is subjec i e, socially cons uc ed h ough he con inuous social in e ac ion wi h o he s. Such in e ac ions occu in he ou ine o he p esen e e yday li e bu conside ing p ojec ions owa ds he pas and he u u e (Be ge & Luckmann, 1991, pp. 43-47). We d aw on da a collec ed o he p ojec MobileWel a e, unded by he No ace Wel a e S a e Fu u es p og amme. We ocus on 24 semi- s uc u ed in e iews conduc ed wi h B azilian men and women li ing in Lisbon and Ba celona me opoli an a eas be ween Ap il and Sep embe 2016. The ec ui men o pa icipan s occu ed h ough pu posi e snowballing and he esul ing sample (Tables 1 and 2) included di e si ied p o iles in e ms o sex, educa ion le el, age and s age in he li e cycle. We conside ed only mig an s bo n in B azil who had spen a yea o mo e in a leas one Eu opean coun y. The in- e iews we e conduc ed in Po uguese and Spanish and ook be ween one and wo hou s. Taking a li e his o y app oach he in e iews explo ed o wha ex en and how do pe cep ions and use o wel a e esou ces shape mig a ion decisions o e ime. The in e iew ques ions we e s uc u ed o ace back he li e and mig a ion his o y o he pa icipan s since hei childhood. The i s pa o he in e iew included ques ions on he in- di idual and ex e nal con ex s be o e mig a ion. The second pa asked abou he si ua ion o he pa icipan s a e mig a ion and wha had changed o e ime in e ms o hei adminis a i e mig a ion s a us, employmen si ua ion, amily li e, con ac s wi h he coun y o o igin and changes in he educa ional, occupa ional and inancial ajec o y. Ques ions on social p o ec ion a angemen s in he coun y o esidence ollowed. The in e iew also asked o pa icipan s o compa e be ween he wel a e esou ce en i onmen in he coun y o o igin and in he coun y o des ina ion. The inal pa o he in e iew asked abou pa icipan s’ u u e aspi a ions, challenges and unce ain ies. The da a we e ansla ed in o English and analysed wi h N i o 11. Fo ou analysis we combined he hema ic and na a i e ap- p oaches. Using a codebook, we s a ed wi h he hema ic analysis and iden i ied place and ime speci ic pa e ns in he needs, access and use o wel a e p o isions. The codebook included codes o pa icipan s’ wel- a e esou ces in di e en domains (such as heal h ca e, educa ion, o e i emen ), he coun ies ha he pa icipan s e e ed o, pa icula li e e en s such as e i emen , he bi h o a child o unemploymen and pa icipan ’s mig a ion expe ience. Applying mul iple coding, we iden i ied changes o e ime in he pe cep ions and use o he wel a e esou ce en i onmen . Wi h he na a i e analysis app oach we iden i- ied commonali ies among pa icipan s in how hey concep ualised he wel a e egimes in B azil and how his concep ualisa ion changed o e ime, in luenced hei ac ions a ound wel a e p o ision and subse- quen ly shaped hei wel a e habi us o e he mig a ion p ocess. To complemen he quali a i e analysis o he p ima y da a we conduc ed a li e a u e e iew and analysed seconda y quan i a i e da a on he wel a e egimes in B azil, Po ugal and Spain and on B azilian mig a ion o Ba celona and Lisbon. I is beyond he scope o his s udy o explo e wha d i es B azilian mig a ion o Lisbon and Ba celona, ansna ional wel a e a angemen s o he access o wel a e esou ces o mig an s wi h di e en adminis- a i e s a uses. Ou choice o Lisbon and Ba celona is based on wo c i e ia. Fi s ly, he sample selec ion is based on he simila ele ance o non-s a e-p o ided wel a e esou ces -especially he amily- in hese wo Medi e anean wel a e egimes. Secondly, and mo e impo an ly, he sampling is based on he di e ence in mig a ion his o ies. Compa ing Lisbon and Ba celona, he longe mig a ion adi ion o B azilian mig a ion o Lisbon is likely o a ec mig an ’s wel a e esou ce Fig. 1. Tempo al dimensions unde analysis. Sou ce: Own elabo a ion d awing on Emi baye and Mische (1998). D. Joli e and S. Pe ei a Geo o um 120 (2021) 58–66 61 en i onmen . The sample is no ep esen a i e o he B azilian popula- ion in he wo Eu opean ci ies. Ins ead, he pu posi e sample aimed o collec da a on a di e se ange o needs and expe iences o wel a e p o ision aking in o accoun he mig a ion expe ience (be ween 2 and 33 yea s) and he ime in he cu en coun y o esidence (be ween a ew mon hs and 33 yea s). P elimina y conclusions ha e eme ged om his explo a o y s udy wi h a limi ed sample size. These insigh s could be u he explo ed and con i med wi h mo e ambi ious s udies. The in- e iews allowed us o unde s and he in e wined links be ween his- o ical and geog aphical ac o s shaping concep ions o wel a e egimes in con ex s o mig a ion and highligh he ole ha he coun y o o igin plays in mig an s’ ames o e e ence o e he mig a ion p ocess. Ou esul s lead us o ques ion he na ow concep ualisa ion o wel a e p o ision in he exis ing li e a u e on wel a e as a d i e o mig a ion ha mainly looks a wel a e bene i s and social policies in he coun ies o o igin and des ina ion. Be o e p esen ing he esul s, he nex sec ion in oduces he con ex s unde s udy. 4. Con ex B azil is a coun y wi h s ong socio-economic inequali ies ha a e isible in i s wel a e egimes. In heo y, B azil belongs o he clus e o coun ies ou side he OECD wi h he closes wel a e ou comes in e ms o li e expec ancy and li e acy o hose in mo e de eloped wel a e s a es (Sha kh & Gough, 2010: 35). In B azil, he e is public uni e sal social p o ec ion unded by wo ke s’ con ibu ions and social assis ance in- dependen om con ibu ions 1 . The mo e a luen usually con ac complemen a y p i a e p o ec ion schemes. The B azilian cons i u ion o 1988 clea ly s a es in a icle 6 ha he igh s pe aining o social wel a e a e undamen al social igh s ha include: he igh o educa- ion, heal h, wo k, housing, leisu e, secu i y, social secu i y, p o ec ion in ma e ni y and childhood, and assis ance o hose wi hou p o ec ion (Nolasco, 2012). Acco ding o Nolasco he s a e app oach is o p omo e equali y by adop ing measu es ha imp o e li ing condi ions o hose wi h ewe esou ces. In p ac ice, howe e , dispa i ies o income a e much highe han he a e age le el in he OECD coun ies (OECD, 2015) and wel a e is no expe ienced in he same way by di e en ly posi ioned indi iduals in B azil’s socio-economic s uc u es. Na a i es o B azilian mig an s collec ed in his esea ch and o he s such as he THEMIS p ojec 2 , clea ly e lec how in B azil, people’s wel a e habi us is s ongly in luenced by hese unequal s uc u es. The con ex in he wo Sou h Eu opean ci ies and coun ies in e ms o wel a e and mig a ion policy owa ds B azilian immig an s is di e - en ia ed. In he dis ic o Lisbon, in 2015, esided legally (bu wi hou Po uguese na ionali y) 36,438 B azilian na ionals, o whom he ma- jo i y we e women (22,318) (da a om he Aliens and Bo de s Se ice 3 ). They cons i u e he mos nume ous o eign g oup in he coun y and in Table 1 Sampling ma ix. Men Women Age Li e s age Family si ua ion Ba celona Lisbon Ba celona Lisbon 18–35 (T ansi ion o) i s yea s o wo king li e No child en 2 2 2 2 25–54 (Planned) amily o ma ion/pa en hood Plans o ha ing child en, o a leas one child (up o 14 yea s) 2 2 2 2 55+(T ansi ion o) e i emen Di e se si ua ions 2 2 2 2 Sou ce: Own elabo a ion. Table 2 O e iew o pa icipan s. Place o esidence Pseudonym Yea o bi h Gende Ci izenship Le el o educa ion (comple ed) Fi s a i al o cu en coun y o esidence Ba celona 1959 FEMALE BR SECONDARY 2006 Jo˜ ao 1964 MALE BR SECONDARY 2008 1961 FEMALE BR, IT POSTGRADUATE 1983 1963 FEMALE BR UNDERGRADUATE 2003 1991 FEMALE BR SECONDARY 2016 Ra aela 1984 FEMALE BR, ES UNDERGRADUATE 2006 Bea iz 1978 FEMALE BR, IT POSTGRADUATE 2006 1983 MALE BR POSTGRADUATE 2006 1983 MALE BR, ES SECONDARY 2002 Julia 1983 FEMALE BR UNDERGRADUATE 2007 Robe o 1982 MALE BR, ES POSTGRADUATE 2006 1971 MALE BR, ES SECONDARY 2006 Lisbon 1982 MALE BR SECONDARY 2008 1977 MALE BR, PT SECONDARY 2000 Zuleida 1959 FEMALE BR, PT SECONDARY 2005 1951 FEMALE BR, PT BASIC LITERACY 2006 1958 FEMALE BR, PT SECONDARY 2007 Ma iana 1966 FEMALE BR, PT POSTGRADUATE 2007 Vinicius 1981 MALE BR UNDERGRADUATE 2014 Lua 1972 FEMALE BR UNDERGRADUATE 2013 1979 FEMALE BR, PT SECONDARY 1999 Ca la 1982 FEMALE BR, PT UNDERGRADUATE 2000 1987 FEMALE BR UNDERGRADUATE 2012 1950 FEMALE BR PRIMARY 2008 Sou ce: Own elabo a ion based on he MobileWel a e da abase. Coun ies o ci izenship: B azil (BR), I aly (IT), Po ugal (PT), Spain (ES). 1 h p://www.planal o.go .b /cci il_03/dec e o/d3048.h m, accessed in Janua y 2017. 2 Fo mo e in o ma ion: h ps://www.imi.ox.ac.uk/comple ed-p ojec s/ he mis. 3 Fo mo e in o ma ion: h ps://se s a .se .p /dis i os.aspx accessed in Janua y 2017. D. Joli e and S. Pe ei a Geo o um 120 (2021) 58–66 62 he Lisbon dis ic . Since he la e 1990s, he numbe o mig an s om B azil g ew e y s ongly only coming o a hal ollowing he e ec s o he 2008 global economic c isis and he in oduc ion o he aus e i y p og amme in Po ugal in 2011. Gi en he his o ical ela ionship be- ween he wo coun ies, apa om he Po uguese immig a ion policy owa ds hi d coun y na ionals, wo legisla i e measu es a ec B azil- ians in pa icula . Fi s , since 2000, he e is he “S a us o Equali y” be ween Po uguese and B azilian ci izens. The bene icia ies o such s a us (B azilians in Po ugal o Po uguese in B azil) enjoy he same igh s and obliga ions as he na ionals –Resolu ion o he Assembleia da República 83/2000. Second, he e is a bila e al Social Secu i y Ag ee- men ha gua an ees equal igh s and ans e ing social secu i y en i- lemen s be ween he wo coun ies, o example, upon e i emen . The e is also an ag eemen ega ding mu ual access o he na ional heal hca e sys em called PB4. This documen g an s access o heal hca e in Po ugal o B azilian empo a y mig an s and ou is s, as well as o Po uguese ci izens in he same ci cums ances in B azil. In he municipali y o Ba celona, by Janua y 2016, 5659 B azilian na ionals we e eco ded in he ci y egis e (Pad on Municipal) inde- penden ly o hei esidence s a us in Spain; 55.8 pe cen o hem we e women – da a on o eign popula ion acco ding o hei ci izenship. B azilian na ionals a e a ela i ely small g oup in Spain; in Ba celona, hey ep esen ed only wo pe cen o he o eign na ionals. Despi e hese ela i ely small numbe s, Spain has become in ecen yea s a main des ina ion in Eu ope o B azilian mig an s (OECD, 2017). In 1996, only 5387 B azilian na ionals we e egis e ed in Spain; by 2008, he igu e had isen o 116,548. A e he ou b eak o he global economic c isis, he eco ds o B azilians in Spain p og essi ely dec eased a na- ional le el (73,863 in 2015). In Ba celona, a e d opping o he lowes le el in 2015 (5451), igu es ha e inc eased o each 7107 mig an s in Janua y 2019 (INE, 2020). Spain and B azil signed in 1991 an ag ee- men on Social Secu i y ha came in o e ec in Decembe 1995 and was amended in Oc obe 2012. This ag eemen s ipula es (wi h some ex- cep ions) ha people who wo ked in any o he wo coun ies and hei amily membe s a e subjec o he social secu i y sys em o he coun y whe e hey wo k. The pe iods o con ibu ion in bo h coun ies can be summed up o con ibu o y bene i s. These bene i s can be alloca ed independen ly o he place o esidence in he wo coun ies. Since 2010, hese ans e abili y igh s apply o social bene i s ela ed o in alidi y, e i emen , dea h and su i o bene i s, and o occupa ional acciden and illness allowances (BOE, 1996, 2012). Howe e , by he ime o ou s udy o eign ci izens wi hou a alid esidence pe mi had limi ed o no access o wel a e s a e p o isions. Fu he mo e, many B azilian-bo n wo k in o mally in he domes ic sec o p o iding ca e o cleaning se - ices, which excludes hem om he o mal social p o ec ion mechanisms. 5. How he wel a e esou ce en i onmen shapes he wel a e habi us in B azil This sec ion di es in o he wel a e esou ce en i onmen in B azil. I explo es how inequali ies in wel a e p o ision in B azil shape he wel- a e habi us. I also shows how he wel a e habi us is embedded in pa icula socio-economic s uc u es. In B azil, people’s socio-economic backg ound has an e ec on access o and expe iences a ound wel a e p o ision. This is pa icula ly obse ed in he ields o educa ion and heal h. In educa ion, he e is a clea dis inc ion in p ima y and seconda y school be ween public and p i a e educa ion. S a e unded and ee public schools end o be pe cei ed o poo quali y and only o hose who canno a o d a p i a e educa ion. P i a e schools a e, a his le el, conside ed o bes quali y and a p i ilege o he mos a luen classes. S a is ical da a con i ms his di ision. Child en om lowe socio- economic backg ounds a e mo e likely o be en olled in public schools. Acco ding o he 2011 B azilian Na ional Household Su ey (PNAD), he median mon hly household pe capi a income in amilies wi h child en en olled in public schools was 36.6 pe cen o ha o amilies wi h child en in p i a e schools. The pe cen age o child en aged 7–14 en olled in public schools declines as he household pe capi a income inc eases (Bu sz yn, 2016: 5). On he con a y, a he le el o highe educa ion, ee public uni e si ies a e conside ed he mos p es igious. Howe e , access o public uni e si ies is ex emely selec i e and s uden s coming om p i a e seconda y schools a e be e p epa ed o ul il he selec ion c i e ia o access o hem. Du ing he wo ke s’ pa y (PT) go e nmen (2002–2016), a s a egy o igh unequal access o highe educa ion based on income was in oduced and a quo a sys em was implemen ed in 2012: 50 pe cen o places in public uni e si ies ha e o be alloca ed o s uden s coming om public schools. This pe - cen age is spli be ween s uden s o low income and o highe income households (below/equal o and abo e one minimum wage and a hal espec i ely) and akes in o accoun he ‘ acial’ composi ion in he s a e. These di isions and changes in he quali y o educa ion a e e lec ed in people’s e ol ing wel a e habi us such as in he case o Ma iana om Bahía, in Lisbon: “We ha e always s udied in public schools. Only my younges sis e , who was bo n when I was 15, s udied in p i a e schools. My pa en s p e e ed o ha e he in a p i a e school because he public educa- ion was no as good as be o e. (…). Then I joined a p i a e uni e si y because i ’s mo e di icul o join a public han a p i a e uni e si y, so all o us s udied in p i a e ins i u ions”. Simila di isions be ween public and p i a e ac o s o wel a e p o- ision a e obse ed in heal hca e. The in e iewees explici ly e e o conce ns associa ed wi h he quali y o public heal hca e in B azil and he impo ance o ha ing access o a p i a e insu ance. This is he case o Robe o om Belo Ho izon e in Ba celona: “The public heal h se ice he e is eally bad. E en o eme gencies you may end up wai ing o 3 days be o e being a ended. O when you need an appoin men wi h a specialis doc o you may ge one in 6 mon hs’ ime. So, you can’ ely on i . I you can a o d i , you go o he p i a e se ice”. The public heal hca e sys em es ablished in B azil by he 1988 Cons i u ion is uni e sal, and i is assumed ha i is a s a e esponsibili y o gua an ee uni e sal and equal access o heal hca e. Howe e , Soa es San os e al. (2008: 1432) a gue ha in 1988, he p i a e heal hca e sec o was al eady obus , wi h hospi al in as uc u es being p edom- inan ly p i a e. P i a e insu ances we e and a e s ill widesp ead oo. Soa es San os e al. (2008: 1435) no ed ha p i a e hospi als o e mo e complex medical equipmen and se ices in compa ison o he public sys em. Thus, in p ac ice, socio-economic inequali ies and he poo quali y o he public heal hca e sys em lead o a di e en ia ed wel a e habi us, wi h well-o indi iduals seeking p i a e heal hca e (Ne i & Soa es, 2002: 78). Da a om Suplemen o Saúde da Pesquisa Nacional po Amos a de Domicílios (IBGE, 1998) analysed by Ne i and Soa es (2002: 82) e eals ha , among he poo es 10 pe cen , only 2.8 pe cen ha e some kind o p i a e heal h insu ance. The p opo ion goes up o 74 pe cen among he weal hies 10 pe cen . Di e ences a e also obse ed in he use o public and p i a e heal hca e se ices. The mo e a luen end o use hem mo e o en, especially o ou ine and p e en i e check-ups (33.6 pe cen o he 10 pe cen poo es and 37.1 pe cen o he 10 pe cen weal hies ). A he same ime, lowe income le els a e mo e ep- esen ed among hose using heal h se ices because o illness. Nega i e pe cep ions o he s a e-p o ided wel a e p o isions also a ec he wel a e habi us in he ace o loss o income ela ed wi h e i emen o unemploymen . Many in e iewees in Lisbon and Ba ce- lona illus a e he conce ns o lowe income amilies in B azil o e he low public e i emen pensions. Fo example, Ca la om Ma o G osso do Sul in Lisbon explains ha when he a he e i ed, she ealised ha she could no ely on he B azilian public e i emen sys em. Despi e wo king many yea s and con ibu ing o he public pension scheme, he D. Joli e and S. Pe ei a Geo o um 120 (2021) 58–66 63 a he couldn’ li e only om his e i emen pension. She concluded ha she had o u n o p i a e a angemen s o secu e he sel an income when he ime would come. Ca la summa izes he endency in B azil o cope wi h i s c ises and mal unc ioning h ough in o mal a angemen s: “I emembe ha B azil’s economic si ua ion was eally di icul , and i was ha d o ind a job, so he e was a cons an eeling o insecu i y. The e used o be much in o mal wo k and B azilians we e p oud o being able o make a li ing e en in he ha des condi ions. B azilians we e s imula ed by he go e nmen o manage he si ua ion doing any kind o in o mal wo k because he c isis was widesp ead”. Consequen ly, al hough he exis ing li e a u e has ocused on he ole o s a e-p o ided wel a e p o isions in he coun ies o o igin o mig a ion (de Jong & de Valk, 2019; Ku eko ´ a, 2013; Mahend a, 2014), p i a e and in o mal wel a e p o isions in he o igin coun ies a e also likely o play an impo an ole. In he es o he a icle, we explo e he o igin coun y e ec in wel a e p o isions in con ex s o mig a ion. We pay pa icula a en ion o he in e play o he wel a e habi us wi h people’s e ol ing si ua ion and hei o ien a ions owa ds he u u e. Doing so, we explo e he a ying deg ee o impo ance o pas , p esen and u u e aspec s o wel a e p o ision a di e en s ages o he mig a ion p ocess. The nex sec ion analyses he ole o he wel a e habi us in ini ial mig a ion p ojec s. 6. The o igin coun y e ec o wel a e in mig a ion p ojec s Mig a ion is d i en by a combina ion o d i e s (Van Hea e al., 2012, 2018) and, aken alone, wel a e- ela ed ac o s could no explain he in e iewees’ mig a ion decisions. Howe e , wel a e- ela ed mig a ion aspi a ions a e men ioned by a ew in e iewees, especially hose expe iencing speci ic li e ansi ions and e en s linked o illness, educa ion, o e i emen . The aspec s o wel a e p o ision ha he pa icipan s conside a e o e looked in Bo jas’ wel a e magne hy- po hesis. I is no he di e en ials in o mal wel a e p o ision ha ma e o hem. Ins ead, mig an s end o conside wha is a ailable o hem in hei places o o igin bo h in e ms o o mal and in o mal mechanisms o access wel a e esou ces. Fo ins ance, o Jo˜ ao and Ma ia, men ioned be o e, conce ns o insu icien s a e-p o ided wel a e esou ces a he age o e i emen did play a key ole in hei decision o mig a e. Howe e , when hey we e conside ing he possibili y o mig a ing, hey we e balancing hei si ua ion and p ospec s in hei coun y o o igin in e ms o income, o mal social p o ec ion and in o mal amily ca e esponsibili ies. They we e no assessing he wel- a e bene i s in hei u u e coun y o esidence. They we e d i en by conce ns ega ding hei u u e wel a e esou ces as p o ide s and e- cei e s bu he solu ion hey we e seeking was an inc ease in income h ough (in o mal) labou ma ke pa icipa ion in a coun y ha o e ed, in hei iew, mo e oppo uni ies. The solu ion hey a icula ed was deeply embedded in hei own iews and unde s andings o wel a e egimes om wha hey had expe ienced in B azil – hei wel a e habi us he e. In hei case, hey we e no used o elying on he s a e o p o ide income when i was no a ailable h ough o mal employmen . Hence, we could ask o wha ex en s a e-p o ided wel a e p o ision, e en in mo e de eloped wel a e s a es is uni e sally unde s ood as su icien o mee social p o ec ion needs – an unde lying assump ion o he wel a e magne hypo hesis. Fu he mo e, we obse e ha he e ec in mig a ion p ojec s o ames o e e ence based on p e ious expe iences o wel a e egimes a e likely o be in luenced by people’s socio-economic backg ound. Fo example, Bea iz, a pos g adua e om Juiz de Fo a li ing in Ba celona, who desc ibes he sel as pa o he com o able B azilian middle class, mig a ed o he i s ime o ca y ou he educa ion p ojec : “I s udied English philology and hen I le . I had a godmo he ha , when I was bo n le me some money, she opened a sa ings accoun , which I could only use when I was 18 yea s old o wha e e I wan ed and when I u ned 18 yea s old i was decided ha I would use i o go o London and I spend 6 mon hs he e, s udying English”. I was he a luen backg ound ha enabled he educa ion-mo i a ed mig a ion. Bea iz’s example sugges s ha i is no only he lack o wel a e p o ision ha d i es mig a ion. Access o (p i a e) wel a e p o isions in he coun y o o igin can also lead o mig a ion aspi a ions. This ep esen s a complemen a y pe spec i e o he s udies ocusing on he e aining e ec s o wel a e p o isions in he o igin coun y (Ku eko ´ a, 2013; Mahend a, 2014). These o e look he acili a ing mig a ion e ec s o he b oade o mal and in o mal esou ce en i on- men in he o igin coun y. Things wo ked ou di e en ly o Amanda. She was bo n in a u al a ea in Minas Ge ais in an economically com o able en i onmen . She ma ied and had h ee child en, bu she hen di o ced and became he main b eadwinne o he household. Amanda mig a ed a he age o 45 due o sho alls in he B azilian wel a e egime ha she could no o e come wi h he inancial esou ces. Be o e he in oduc ion o he quo a sys em in public uni e si ies, when access o public uni e si ies in B azil was limi ed o he bes s uden s o en om he mo e a luen classes, he only possibili y o Amanda’s child en o s udy was o access a p i a e uni e si y: “As my child en s udied in public schools, hey we e no [educa ionally] p epa ed o be accep ed in a public uni e si y”. The e o e, when he oldes daugh e was in he second yea o uni e si y and he son was p epa ing o access he uni e si y, she decided o mo e o Spain o inc ease he income o a o d he cos s o p i a e e ia y educa ion o he h ee child en. Addi ionally, she wan ed o accumu- la e economic esou ces o cope wi h he loss o income a e he e i emen . All hese expe iences illus a e possible in e plays be ween he o igin coun y’s wel a e esou ce en i onmen , and indi idual socio-economic and li e cou se ac o s in shaping he ole o wel a e p o ision in mig a ion p ojec s. The examples show mul iple social dimensions in which wel a e ope a es as a d i e o mig a ion. Fa beyond he po en- ially mo e gene ous wel a e bene i s in he coun y o des ina ion, we ha e iden i ied wel a e- ela ed ac o s ha can in luence mig a ion om he indi idual, household, amily and wide socie y pe spec i e in he coun y o o igin. Fu he mo e, in he cases o Jo˜ ao and Amanda, we obse ed ha he wel a e habi us and he p esen ci cums ances a pe sonal and s uc u al le el shape in a simila deg ee ideas on wha would ha e been hei wel a e esou ce en i onmen in he u u e i hey had s ayed in B azil. Taking in o conside a ion hei wel a e habi us acqui ed h ough pas expe iences in he B azilian wel a e egime and he wel a e esou ces accessible o hem, Jo˜ ao and Amanda conside ed mig a ion as an al e na i e o econ igu e hei wel a e habi us ha hey conside ed unsus ainable in he u u e. The nex sec ion explo es how mig an s’ pe cep ions and decisions o wel a e upon a i al a e also shaped by ames o e e ence o med by he p e-exis en wel a e habi us. I also sheds ligh on he ans- o ma ions in he unde s andings o wel a e egimes ha occu a e mig a ion and how his ela es o conside a ions linked o he h ee empo al dimensions o agency. 7. Expe iences o wel a e: be ween old concep ions and new eali ies We ha e a gued ha people a e used o speci ic ways o hinking and doing ha a e in luenced by he esou ce en i onmen accessible o hem wi hin he exis ing wel a e egimes in he coun y o o igin. I is likely ha hese ames o e e ence shape ini ial ac ions a ound wel a e p o ision in he des ina ion coun y, a p ocess ha we called he o igin coun y e ec . This migh explain why mig an s a e gene ally unawa e o he implica ions o mig a ion on how hey will o ganise hei social p o ec ion a e mo ing (de Jong & de Valk, 2019; Scheibelho e & Holzinge , 2018 among o he s). Ou esul s also e lec ha D. Joli e and S. Pe ei a Geo o um 120 (2021) 58–66 64 ( ansna ional) wel a e- ela ed ac ions a e econside ed en ou e, on a need o need basis. We can di e en ia e wo s ages in he e olu ion o he wel a e habi us a e mig a ion: Upon a i al he pas dimension o agency domina es, and old con- cep ions om he o igin coun y s ongly shape ini ial wel a e- ela ed ac ions in he coun y o des ina ion. The na a i e o Ma iana om Bahía and li ing in Lisbon is illus a i e o how he wel a e habi us was in luen ial in he case o sea ching o a school o he daugh e when she a i ed: “I s a ed sea ching o mo e in o ma ion a e I go he e, specially, abou he schools, he eali y he e is o ally di e en i we compa e o schools in B azil. Public schools he e end o be e y bad, ha ’s why he e a e so many p i a e ins i u ions. So, when I go he e, I s a ed looking o a p i a e school o my daugh e , because ha was he e e ence I had abou schooling”. Ma iana’s immedia e needs upon a i al had mo e weigh han any u u e wel a e p o ision needs o aspi a ions. She d ew on he wel a e habi us in an unknown new wel a e egime. I Ma iana had elied on social ne wo ks, he pas wel a e habi us migh ha e had less impo - ance in he judgemen s o choose a school. The wel a e habi us keeps playing a ole o e ime in he des ina ion coun y. Júlia’s expe ience is one example. In si ua ions o unemploy- men , his a is in Ba celona who has been jumping om one in o mal job o ano he lea ned om he wel a e egime in B azil o cope wi h he lack o income in Spain. He na a i e e lec s ha i is no only he o mal wel a e esou ces in B azil ha shaped he wel a e habi us ha she ep oduces in Ba celona. Wha she has lea ned om he pa en s’ agency and eac ions o he s uc u al cons ain s in B azil also played a ole: “In B azil, people imp o ise a lo . (…) B azil is always in c isis so we a e ma ina ed in he necessi y o su i e. (…) E e y hing I do he e I lea ned om seeing how i wo ked he e and how I ein en ed mysel was om my a he and mo he as my eache s. We always had o in en ou sel es o pay he bills wi h wha we had”. So, a e some ime in he place o des ina ion, mig an s end o keep e e ing o old ames o e e ence o assess wel a e esou ces in he place o esidence. This is he case o Ra aela, om S˜ ao Paulo and li ing in Ba celona, when she e lec s upon he heal hca e p o isions in he coun y o des ina ion: “(…) o me, in his coun y [Spain], he heal h sys em is g ea , because hey ha e ne e le me down a any momen . On he con a y—I hink he public heal h ca e sys em is almos p i a e o me (…)”. Howe e , he expe ience o Ra aela also illus a es how, o e ime, mig a ion e eals i sel as a lea ning p ocess ha shapes ames o e e ence, gene al pe cep ions and unde s andings o wel a e p o ision as mig an s begin o expe ience and lea n abou new eali ies: “I didn’ ha e a compa ison. I hadn’ li ed in a wo ld o highe ed- uca ion and quali y o li e. (…). And now I can say ha B azil is ho ible, in e ms o quali y o li e, educa ion, heal h, in as uc u e, e e y hing. (…) I knew B azil was ho ible, bu I didn’ know i was ha ho ible. When I came he e, I saw ha ”. Fo Ra aela, he new expe ience in he des ina ion coun y opened up new op ions o wel a e p o ision and in oked he cons uc ion o new pe spec i es on he wel a e egime in he o igin coun y. P esen expe- iences in new wel a e egimes a e hus likely o shape wha mig an s ecall om pas expe iences and hei wel a e habi us. This is a p ocess ha akes place o e he mig a ion p ocess and in unan icipa ed ways. Recen s udies ha e unde lined he ole o necessi y in adop ing new wel a e a angemen s in he des ina ion coun y (de Jong & de Valk, 2019; Scheibelho e & Holzinge , 2018). Wha we aim o highligh wi h his s udy is he ole o he changing meaning o wel a e esou ces caused by a socio-spa ial eloca ion (Hal ac ee & Ri e a, 2012) ha emains o e looked in he mig a ion li e a u e, especially in Bo jas’ wel a e magne hypo hesis. We ha e iden i ied ou in e ela ed ac o s causing ans o ma ions in he wel a e habi us ha a e likely o eme ge wi h mig a ion. We illus a e hem h ough he expe iences o some in e iewees. Fi s ly, he lea ning- h ough-expe ience ha akes place in he new con ex . This is illus a ed in he case o Ma iana. Fo p ac ical easons, she had o egis e he daugh e o a public school. Tha was agains he ini ial p e e ences, bu he scep icism and doub s owa ds he public educa ion sys em based in he expe iences in B azil aded away when she ealized ha he public school ha he daugh e a ended “was a good school wi h good eache s” and his al e ed he unde s anding o public schools in he new con ex . Zuleida’s expe ience is also an example o hese empo al shi s in concep ions o wel a e p o ision. In o mal wel a e esou ces we e he main d i e o wha was ini ially a sho - e m mig a o y p ojec . Once he p essing needs o he son upon he a i al we e me , wi h mig a ion and new expe iences in Lisbon, Zuleida was able o iden i y di e ences in he wel a e egimes o B azil and Po ugal and, o ien ed owa ds he own u u e wel a e p o ision needs, she inco po a ed aspec s o he local wel a e egime o he wel a e habi us. S a e-p o ided wel a e esou ces played a mo e impo an ole han in o mal esou ces when she econ- igu ed he p ojec o s ay longe in Lisbon combining o mal wel a e esou ces bo h in B azil and Po ugal. Secondly, he new se o oppo uni ies and cons ain s in he immi- g a ion en i onmen . In he domain o heal hca e, he expe ience o Vinicius ( om he s a e o S˜ ao Paulo and li ing in Lisbon) e lec s how p e ious ames o e e ence combined wi h he cons ain s ha he aced in Po ugal as an i egula mig an shaped his ac ions in e ms o wel a e p o ision. His choice o a p i a e heal h insu ance was based on wo assessmen s. On he one hand, his lack o us on he B azilian public heal hca e sys em. On he o he hand, his limi ed access o s a e- p o ided wel a e esou ces in Po ugal as an i egula mig an : “As ou si ua ion is no ully legalized he e ye , I decided o ge he easie way, which would be ha ing he p i a e heal h ca e plan o bo h o us. I didn’ e en ge in o ma ion abou [ he public heal hca e]”. In sum, a combina ion o wel a e habi us and s uc u al cons ain s econ igu ed his ac ions once in Lisbon. Thi dly, li e e en s ha can shape, o example, ames o belonging. This is wha happened o Bea iz who also had a p i a e insu ance in Ba celona. Howe e , when he son was bo n, she s a ed using he public heal hca e sys em. This happened almos by p ocedu al ine ia. She isi ed a paedia ician in he neighbou hood public su ge y as she was old o do o he pe iodical check-ups on he baby’s weigh and de elopmen and o his accina ions. As a mig an , she also el ha i was a way o con i ming he baby’s belonging o his coun y o bi h: “I hink I wan ed o ha e him egis e ed he e, like he exis s, bo n he e and so on, wi h his accina ions”. Heal h- ela ed li e e en s can also lead o econ igu a ions o he wel a e habi us. Fo ins ance, p io i ies in wel- a e p o ision changed o Amanda when she su e ed a domes ic acci- den a e some ime in Ba celona. Due o he ac u es caused by he acciden she was no able o wo k o se e al mon hs. She e u ned empo a ily o B azil un il she eco e ed. The e she elied on he in o mal ca e o he child en and he o mal wel a e bene i s om Spain. Un il she eco e ed, he p io i ies changed om he ini ial o ien a ions owa ds he u u e educa ion o he child en and he pension o he mo e p essing needs o heal h ca e and loss o income o he p esen . Finally, mig a ion ne wo ks and mig a ion his o y also play a ole in he change o wel a e habi us. This is appa en in he case o Lua om San a Ca a ina in Lisbon. Be o e mig a ing, she disco e ed ha she and he son had a degene a i e disease. A doc o ecommended ha hey wen o Po ugal since hei disease was mo e common he e and ea men mo e ad anced. She was di ec ly e e ed o doc o s in Po ugal. The ans e was possible h ough he PB4 bila e al ag eemen men ioned ea lie . The wel a e ag eemen s esul ing om he his o ical links be ween B azil and Po ugal eased he o mali ies o mo e o D. Joli e and S. Pe ei a Geo o um 120 (2021) 58–66 65 Po ugal and opened new pe spec i es o Lua o sea ch a mo e sui able heal h ca e elsewhe e. The con ex o he in ense and long mig a ion his o y be ween Po ugal and B azil also played a ole as mo he and son we e ecei ed in Po ugal by he amily o a Po uguese iend in B azil. These ou in e ela ed elemen s show he s ongly in e ela ed ole o his o ical and geog aphical ac o s (including he immig a ion en i- onmen and mig an s’ es ic ed access o wel a e) in shaping meanings o wel a e p o ision and p ac ices a ound i (Raghu am, 2016). 8. Conclusion In his a icle, we ha e sough o highligh mul iple ways h ough which mig a ion and wel a e cons i u e a dynamic ela ionship. This is an aspec ha has been igno ed in Bo jas’ wel a e magne hypo hesis and, al hough acknowledged in he ecen li e a u e on ansna ional social p o ec ion, i has also emained unde s udied in esea ch applying his amewo k. We a gue ha , beyond he e ec s on mig a ion o mo e gene ous wel a e egimes in he coun ies o des ina ion and he ad- ances made by he ansna ional social p o ec ion amewo k s essing he a ie y o wel a e p o ide s beyond he s a e in con ex s o mig a ion (Fais e al., 2015; Le i e al., 2016), i is s ill possible o deepen and e ine ou unde s anding o his mu ual and changing ela ionship. The a icle con ibu es o he exis ing li e a u e ocusing on he wel a e habi us and how i changes o e he mig a ion p ocess. I explo es h ee aspec s so a o e looked. Fi s , i looks a he ole o wel a e egimes in he o ma ion and ede ini ion o mig a ion p ojec s. Second, i analyses how he wel a e habi us eme ged om expe iences in he coun y o o igin deeply shape concep ions o wel a e egimes and a ec expec a- ions on a i al and ini ial ac ions a ound wel a e p o ision in he des ina ion coun y. Finally, i sheds ligh on he changing meaning o wel a e p o ision o e ime, h ough he li e cycle and ac oss geog a- phies. The a icle d aws on Emi baye and Mische (1998) o p opose an analysis o he dynamic ela ionship be ween mig a ion and wel a e conside ing ha ac ions a ound wel a e p o ision a e in o med by pas expe iences (wel a e habi us), in luenced by (pe cep ions o ) he p esen si ua ion in cons an change, and o ien ed owa ds he u u e. This amewo k allows o cap u e aspec s o he wel a e habi us ha emain unchanged and o he s ha change o e he mig a ion p ocess and he li e cou se. The e a e wo main poin s o depa u e o ou a gumen . On he one hand, people’s expec a ions, aspi a ions and decisions a ound wel a e p o ision, including in i s in e ac ion wi h mig a ion, a e based on speci ic ames o e e ence shaped by people’s habi us, hei indi- idual ci cums ances and pas expe iences o he ex e nal s uc u es in he coun y o o igin. On he o he hand, he e is a ans o ming ac o in mig a ion concep ualised as a li e e en ha changes wha wel a e p o isions mean o e ime as he mig an spends ime in a coun y whe e he wel a e egime is o ganized in a di e en way om wha had been expe ienced be o e. The ans o ma i e p ocess is shaped by he con ac wi h a new socio-spa ial con ex , he changing s uc u al op- po uni ies and cons ain s o e he mig a ion p ocess and changes o e he li e cou se. Mig a ion ne wo ks and mig a ion his o y also play a ole in he change o e e ences. Aligned wi h p e ious s udies, ou da a analysis showed ha he ini ial mig a ion p ojec ei he includes li le o no e e ence o wel a e p o ision as a sole d i e o mig a ion. Howe e , when wel a e p o ision is pa o he mig a ion p ojec , i is la gely shaped by he concep ion o wel a e egimes ha is oo ed in he social s uc u es and habi us in he coun y o o igin, hence he o igin coun y e ec . As such, be o e mig a ion, pas expe iences and p esen ci cum- s ances a pe sonal and s uc u al le el shape imagina ions on u u e p ac ices in he coun y o o igin and he coun y o des ina ion. The wel a e habi us based on expe iences in he coun y o o igin is ep o- duced and emains cen al upon a i al. This is especially he case i mig an s do no ely on social ne wo ks o make decisions o co e ini ial p essing needs in wel a e p o ision. O e ime, he wel a e habi us is econ igu ed en ou e. Mig a ion i sel becomes a lea ning p ocess ha shapes ames o e e ence, gene al pe cep ions and unde s andings o wel a e p o ision as mig an s begin o expe ience and lea n abou new eali ies. Despi e such ans o ma ions, mig an s o en keep e e ing o hei p e ious ames o e e ence. This shows ha wi h mig a ion old ames o e e ence a e no immedia ely subs i u ed by new ones. Hence, i is inco ec o conside ha when wel a e p o ision plays a ole in he mig a ion p ocess mig an s look only o wha is (po en ially) a ailable in he coun y o des ina ion. On he con a y, o e ime ames o e e ence o e lap, expand and emain in e connec ed wi h each o he , as he wel a e habi us de elops and mig an s begin o cons i u e ans- na ional wel a e assemblages as hose desc ibed by Fais e al. (2015) and Le i e al. (2016). Finally, ou indings show ha , like ca e p ac- ices (Raghu am, 2016), p ac ices a ound wel a e p o ision a e si e- speci ic, complex and dynamic. The e o e, esea ch should mo e beyond he assump ions ha , be o e mig a ion, s a e-p o ided wel a e esou ces in he des ina ion coun y is pe cei ed posi i ely a ound he globe and p e e ed o o he wel a e esou ces and coping s a egies conside ed in mig a ion p ojec s. Ou indings based on a quali a i e analysis p o ide new pe spec- i es o he complex ela ionship be ween wel a e and mig a ion o be con i med wi h la ge quan i a i e s udies. Fu u e esea ch could explo e mo e in-dep h he e ec o li e-cou se e en s in he econ igu- a ion o he wel a e habi us and hei impac on mig a ion p ojec s. Pa icula li e e en s such as he bi h o a child, illness, o unemploy- men a e likely o econ igu e he p edominance o pas expe iences, he p esen si ua ion, o u u e o ien a ions in mig an s’ p io i ies and de- cisions a ound wel a e in con ex s o mig a ion. These econ igu a ions emain unde s udied. CRediT au ho ship con ibu ion s a emen Dominique Joli e : Concep ualiza ion, Me hodology, Fo mal analysis, In es iga ion, W i ing - o iginal d a , W i ing - e iew & edi ing, Visualiza ion. Sonia Pe ei a: Concep ualiza ion, Me hodology, Fo mal analysis, In es iga ion, W i ing - o iginal d a . Acknowledgemen s This a icle has been published, hanks o he inancial suppo o he NORFACE T ansna ional Resea ch P og amme on Wel a e S a e Fu u es o he MobileWel a e p ojec . We hank Casa do B asil em Ba celona, Casa do B asil em Lisboa (especially G upo Acolhida) and he Consula e o B azil in Ba celona o hei collabo a ion in he ec ui men o esea ch pa icipan s and o p o iding in o ma ion on he local en i- onmen in which B azilian mig an s li e in bo h ci ies. We a e also g a e ul o he willingness and openness o he in e iewees o ake pa o his esea ch. Finally, we would like o hank Thomas Lac oix o his gene ous commen s o an ea lie e sion o his a icle. Re e ences Be ge , P.L., Luckmann, T., 1991. The social cons uc ion o eali y. A ea ise in he sociology o knowledge, Rep . ed. Penguin Books, London. Bilecen, B., Ba glowski, K., 2015. On he assemblages o in o mal and o mal ansna ional social p o ec ion. Popul., Space Place 21 (3), 203–214. h ps://doi. o g/10.1002/psp.1897. BOE, 1996. Con enio de Segu idad Social en e el Reino de Espa˜ na y la República Fede a i a de B asil de 16 de mayo de 1991. Bole in o icial del Es ado, Mad id. BOE, 2012. Con enio complemen a io de e isi´ on del Con enio de Segu idad Social i mado el 16 de mayo de 1991 en e el Reino de Espa˜ na y la República Fede a i a de B asil, hecho en Mad id el 24 de julio de 2012. Bole in O icial de las Co es Gene ales, Mad id. Bo jas, G.J., 1999. Immig a ion and wel a e magne s. J. Labo Econ. 17 (4), 607–637. Boye , F., 2005. Le p oje mig a oi e des mig an s oua egs de la zone de Bankila ´ e: la pau e ´ e d´ esa ou´ ee. S ichp oben. Wiene Zei sch i ü k i ische A ikas udien 8 (5), 47–67. Bu sz yn, L., 2016. Po e y and he poli ical economy o public educa ion spending: e idence om B azil. J. Eu . Econ. Assoc. 14 (5), 1101–1128. de Jong, P.W., de Valk, H.A., 2019. In a-Eu opean mig a ion decisions and wel a e sys ems: he missing li e cou se link. J. E hnic Mig a ion S ud. 1–19. Emi baye , M., Mische, A., 1998. Wha is agency? Am. J. Sociol. 103 (4), 962–1023. D. Joli e and S. Pe ei a Geo o um 120 (2021) 58–66 66 Fais , T., 2000. The c ucial meso link: social capi al in social and symbolic ies. In: Fais , T. (Ed.), The Volume and Dynamics o In e na ional Mig a ion and T ansna ional Social Space. Ox o d Schola ship Online, Ox o d, pp. 96–122. Fais , T., Bilecen, B., Ba glowski, K., Sienkiewicz, J.J., 2015. T ansna ional social p o ec ion: mig an s’ s a egies and pa e ns o inequali ies. Popul., Space Place 21 (3), 193–202. h ps://doi.o g/10.1002/psp.1903. F iedmann, J. (2005). Placemaking as p ojec ? Habi us and mig a ion in ansna ional ci ies. In: Hillie , J., Rooksby, E. (Eds.), Habi us: a sense o place. Ashga e, Alde sho , Han s, England; Bu ling on, VT, pp. 299–316. Gough, I., 2004. Wel a e egimes in de elopmen con ex s: a global and egional analysis. In: Gough, I., Wood, G., Ba ien os, A., Be an, P., Da is, P., Room, G. (Eds.), Insecu i y and wel a e egimes in Asia, A ica and La in Ame ica: Social policy in de elopmen con ex s. Camb idge Uni e si y P ess, Camb idge, pp. 15–48. Hal ac ee, K.H., Ri e a, M.J., 2012. Mo ing o he coun yside... and s aying: li es beyond ep esen a ions. Sociologia Ru alis 52 (1), 92–114. IBGE, 1998. Pesquisa Nacional po Amos a de Domicílios (PNAD). INE, 2020. Es adís ica del Pad ´ on Con inuo. Da os po municipios - Ba celona. Re ie ed 02/03/2020 om h ps://www.ine.es/dyngs/INEbase/es/ope acion.h m? c=Es adis ica_C&cid=1254736177012&menu= esul ados&secc=12547361954 61&idp=1254734710990#! abs-1254736195557. Joli e , D., 2019. Pos -2008 Mul i-Si ed Household P ac ices: Be ween Mo occo, Spain and No way. In e na ional Mig a ion. Ku eko ´ a, L., 2013. Wel a e sys ems as emig a ion ac o : e idence om he new accession s a es. JCMS J. Common Ma ke S ud. 51 (4), 721–739. h ps://doi.o g/ 10.1111/jcms.12020. La leu , J.-M., Lizin, O., 2016. T ansna ional heal h insu ances as social emi ances: he case o congolese immig an s in Belgium. In: Nowicka, M., ˇ Se bedˇ zija, V. (Eds.), Mig a ion and Social Remi ances in a Global Eu ope. Palg a e Macmillan UK, London, pp. 211–233. Le i , P., Vi e na, J., Muelle , A., Lloyd, C., 2016. T ansna ional social p o ec ion: se ing he agenda. Ox o d De . S ud. 1–18. Mahend a, E., 2014. Financial cons ain s, social policy and mig a ion. IMI Wo king Pape , 101. Mazzuca o, V., 2009. In o mal insu ance a angemen s in Ghanaian mig an s’ ansna ional ne wo ks: he ole o e e se emi ances and geog aphic p oximi y. Wo ld De . 37 (6), 1105–1115. Mo awska, E.T., 2009. The sociology o immig a ion: ( e)making mul i ace ed Ame ica. Palg a e Macmillan, Basings oke. Ne i, M., Soa es, W., 2002. Desigualdade social e saúde no B asil. Social Inequali y and Heal h in B azil. Cade nos de Saúde Pública (18 suplemen o), 77–87. Nolasco, L., 2012. E oluç˜ ao his ´ o ica da p e idˆ encia social no B asil e no mundo. ˆ Ambi o Ju ídico, Rio G ande XV(98). Nowicka, M., 2018. “I don’ mean o sound acis bu …” T ans o ming acism in ansna ional Eu ope. E hnic Racial S ud. 41 (5), 824–841. OECD, 2015. B azil policy b ie . OECD Be e policies se ies. Re ie ed 02/03/2020 om h p://www.oecd.o g/social/b azil-imp o ing-policies- o- educe-inequali y-a nd-po e y.pd . OECD, 2017. In e na ional Mig a ion Da abase. Re ie ed 04/04/2017, om OECD.S a . h p://s a s.oecd.o g/Index.aspx?Da aSe Code=MIG. O’Reilly, K., 2012. In e na ional mig a ion and social heo y. Basings oke Palg a e Macmillan. Raghu am, P., 2016. Loca ing Ca e E hics Beyond he Global No h. 2016, 15(3), 23. Re ie ed 04/04/2017 om h p://ojs.unbc.ca/index.php/acme/a icle/ iew/1353. Scheibelho e , E., Holzinge , C., 2018. ’Damn I , I Am a Mise able Eas e n Eu opean in he Eyes o he Adminis a o ’: EU Mig an s’ Expe iences wi h (T ansna ional) Social Secu i y. Soc. Inclus. 6 (3), 201–209. Se a Mingo , E., Mazzuca o, V., 2019. Mo ing o a ‘be e wel a e’? The case o ansna ional Sudanese amilies. Global Ne w. 19 (2), 139–157. Sha kh, M.A., Gough, I., 2010. Global wel a e egimes a clus e analysis. Global Soc. Policy 10 (1), 27–58. Sienkiewicz, J.J., Sado skaya, Y., Amelina, A., 2015. The Kazakh-Ge man social space: dec easing ansna ional ies and symbolic social p o ec ion. Popul., Space Place 21 (3), 270–281. Soa es San os, I., Dominguez Ug´ a, M.A., Po o, S.M., 2008. O mix público-p i ado no Sis ema de Saúde B asilei o: inanciamen o, o e a e u ilizaç˜ ao de se iços de saúde. Ciˆ encia & Saúde Cole i a 13 (5), 1431–1440. S ones, R., 2012. Causali y, con ex ual ames and in e na ional mig a ion: combining s ong s uc u a ion heo y, c i ical ealism and ex ual analysis. In . Mig a ion Ins . Wo k. Pap. Se . 62. Van Hea , N., Bakewell, O., Long, K., 2012. D i e s o mig a ion. Mig a ing ou o Po e y RPC Wo king Pape , 1. Van Hea , N., Bakewell, O., Long, K., 2018. Push-pull plus: econside ing he d i e s o mig a ion. J. E hnic Mig a . S ud. 44 (6), 927–944. D. Joli e and S. Pe ei a