Political economy of pension reforms: Selected general issues and the Polish pension reform case

Gó a, Ma ek A icle Poli ical economy o pension e o ms: Selec ed gene al issues and he Polish pension e o m case IZA Jou nal o Labo & De elopmen P o ided in Coope a ion wi h: IZA – Ins i u e o Labo Economics Sugges ed Ci a ion: Gó a, Ma ek (2013) : Poli ical economy o pension e o ms: Selec ed gene al issues and he Polish pension e o m case, IZA Jou nal o Labo & De elopmen , ISSN 2193-9020, Sp inge , Heidelbe g, Vol. 2, pp. 1-31, h ps://doi.o g/10.1186/2193-9020-2-2 This Ve sion is a ailable a : h ps://hdl.handle.ne /10419/92291 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 p://c ea i ecommons.o g/licenses/by/2.0/ ORIGINAL ARTICLE Open Access Poli ical economy o pension e o ms: selec ed gene al issues and he Polish pension e o m case Ma ek Gó a Co espondence: ma ek.go a@sgh. waw.pl Wa saw School o Economics, Wa saw, Poland and IZA Abs ac The ins i u ional s uc u es o he a ious ypes o Eu opean wel a e s a e we e es ablished a ound ex a e enues called he "demog aphic di idend" ha used o be easily a ailable h oughou he decades o he 20s cen u y. They, howe e , ceased o be a ailable a he end o ha cen u y. The challenge socie ies ace oday is how o manage he ins i u ions ha ing ela i ely less esou ces a ailable o hei inancing. In his pape I analyse selec ed poli ical economy issues ela ed o pension e o ms, such as he wo ke - e i ee con lic o in e es , dis ibu ion by age o he cos s o he loss o demog aphic di idend, and key goals behind pension e o ms. The ocus o his pape is on pension e o ms in Poland, which ac ually was an implemen a ion o a ho oughly new pension sys em. I s design is analysed and discussed om a poli ical economy iewpoin . Fu he mo e, analysis o he Polish pension sys em is p esen ed in a b oade Eu opean con ex . Jel codes: H55; I31; J11; J18. 1. In oduc ion Th oughou Eu ope, pension sys ems as well as public inance in gene al a e s ongly exposed o he challenges o he 21 s cen u y. This applies o he mo e a luen EU15 as well as o he less a luen new EU membe s a es (Alesina & Glaese 2004). In bo h cases he eason is he same: popula ion changes lie a he oo o he p oblem, as he ins i u ional s uc u es o he a ious ypes o Eu opean wel a e s a e o Eu opean Social Model we e es ablished a ound ex a e enues called he “demog aphic di idend” ha used o be easily a ailable h oughou he decades o he 20s cen u y. They, howe e , ceased o be a ailable a he end o ha cen u y when mos , i no all, Eu opean socie ies eached s age ou o hei demog aphic ansi ion (A kinson 1995). Pension sys ems a e deeply oo ed in he public pe cep ion o socie ies. This applies no only o de eloped coun ies bu also o hose a ious coun ies ha we e his o ically linked o he Eu opean ins i u ional adi ion. Pension sys ems belong o co e ins i u- ional sys ems ha a e expec ed o be e y s able and eliable. Tha pe cep ion has been de eloped o e many decades when consecu i e gene a ions o ci izens pa icipa ed and bene i ed om ha pa icipa ion. O e ime, he link be ween wel a e and wo k e o has weakened. The public in e nalised ha si ua ion and began o pe cei e wel a e as being g an ed, solely on he p emise hey we e he ci izens o a luen and well o ganised coun ies. © 2013 Gó a; licensee Sp inge . This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License (h p://c ea i ecommons.o g/licenses/by/2.0), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed. Gó a IZA Jou nal o Labo & De elopmen 2013, 2:2 h p://www.izajold.com/con en /2/1/2 Eu opean coun ies s ill y o manage hese pension ins i u ions as i he di idend was s ill a ailable. Go e nmen s ocus mos ly on he ine i able iscal cu s om such a si ua ion, e en hough i is deep social e o ms ha a e needed. Eu ope is less a luen han Eu opean ci izens end o assume. Wi h smalle esou ces a ailable, hey need o e hink he way hese may be dis ibu ed wi hin socie ies. Pension sys ems a e p obably mos di ec ly exposed o he p oblems s emming om he loss o he “demog aphic di idend”. A he same ime, in e ms o GDP, pension sys ems in Eu ope a e la ge o e y la ge. Consequen ly, hey in luence he economic and social si ua ions he mos . Va ious pension e o ms a e on he op o policy agendas h oughou Eu ope. These e o ms a e e y o en con es ed by socie ies. Managing he disillusion is he challenge. I should be s essed he e ha I do no e e o he discussion on whe he he wel a e s a e is in gene al a he a good idea (Ba 2001), o a he an idea leading o p oblems (Bodo & Ru kowski 2012). In his pape I ocus on he a ious consequences o he si ua ion, when esou ces needed o inancing wel a e s a e ins i u ions a e g owing (in e ms o GDP) due o popula ion changes, i espec i e o whe he he ins i u ions a e good o bad. In his pape I analyse selec ed poli ical economy issues ela ed o pension e o ms, such as he wo ke - e i ee con lic o in e es , dis ibu ion by age o he cos s o he loss o demog aphic di idend, and key goals behind pension e o ms. The ocus o his pape is on he pension e o m in Poland, which ac ually was an implemen a ion o a ho oughly new pension sys em. I s design is analysed and discussed om a poli ical economy iewpoin . Fu he mo e, analysis o he Polish pension sys em is p esen ed in a b oade Eu opean con ex . 2. The young, he old and he no -ye -old Economic li e a u e on pensions is as . I do no e e o any pa icula pa o his. Ra he , I e e o hose basic ea u es o pension sys ems ha a e c ucial o he poli ical economy o ien ed analysis p esen ed in his pape . F om he mac o pe spec i e he pension sys em is a way o di ide he cu en GDP be ween he wo king gene a ion ha has p oduced i , and he e i ed gene a ion ha has no con ibu ed o ha bu ecei es a pa o he GDP as a pension ans e (gi en o he ans e s). Tha can be illus a ed as in Figu e 1. Popula ion ageing has led o a subs an ial inc ease in he sha e o GDP spen on pen- sion ans e s T 2 >T 1 which implies R 2 <R 1 . This means he nex gene a ion o wo ke s (2) is, in e ms o GDP, mo e poo ly emune a ed o hei ac i i y and p oduc i i y g ow h han he p e ious gene a ion (1). The igu e abo e ocuses exac ly on he con- lic o in e es be ween he gene a ions. Figu e 1 also illus a es ano he c ucial ea u e o popula ion ageing: P oduc i i y g ow h, being essen ial o social wel a e, does no sol e he c ucial p oblem caused by he ageing, namely changing p opo ions in which GDP is di ided be ween he wo king and he e i ed gene a ion 5 . The p oblem we y o sol e is ela ed o he p opo ions o he di ision, no o he scale o GDP. The same p oblem o he sh inking emune - a ion o p oduc ion ac o s applies o he mo e a luen coun ies as well as o he less a luen . The pension p oblem is how o keep s able p opo ions o GDP spen on emune a ions as well as on ans e s. I he sha e o GDP spen on pensions is Gó a IZA Jou nal o Labo & De elopmen 2013, 2:2 Page 2 o 31 h p://www.izajold.com/con en /2/1/2 inc easing, hen he emune a ion o labou and capi al is dec easing is-à- is p oduc - i i y g ow h, which seems unsus ainable in he pe spec i e o a pension sys em based on he alloca ion o income o e ime. The sh inking (in e ms o GDP) emune a ion o p oduc i i y g ow h will end o supp ess p oduc i i y g ow h. The demog aphy d i en change in p opo ions o emune a ion o p oduc ion ac o s and social (no only pension) ans e s indica es a e y deep change in he s uc u e o in e es wi hin socie ies. The con lic o in e es be ween labou and capi al has been supplemen ed wi h ano he con lic , namely he con lic o in e es be ween he ac i e ( inancing ans e s) and inac i e ( ecei ing ans e s) ci izens. The scale o his la e con lic is g owing oge he wi h he g owing sha e o GDP spend on inancing ans- e s. Howe e , he con lic o in e es has no been ully ecognised publicly ye 6 . The pension sys em is he mos di ec ly exposed o he consequences o ha si ua ion. The pension sys em is an ins i u ional s uc u e o in e gene a ional exchange be ween he wo king gene a ion (wo ke s) and he e i ed gene a ion ( e i ees) 7 .The amewo k can be s anda dised, based on uni e sal co e age, hence public, o based on indi idual decisions and esponsibili y, hence p i a e. In bo h cases, e i ees exchange hei pension igh s o a pa o he p oduc ion o he wo ke s. These pension igh s can be exp essed in a ious ways om well-de ined owne ship igh s o uzzy poli ical p omises. Popula ion s uc u e by age is he c ucial ac o behind he exchange. The ollowing simple model illus a es he demog aphic backg ound o any pension sys em 8 . The demand side o he ma ke is de e mined by he numbe o wo ke s, hei p oduc i i y and he con ibu ion a e. The numbe o e i ees and hei expec a ions de e mine he supply side. This can be p esen ed as a simple model, shown in Equa ions 1a and 1b. D¼c weLWð1aÞ S¼z wLRð1bÞ whe e Dis he demand o pension igh s (a pa o GDP spen by he wo king gene - a ion on acqui ing pension igh s), Sis he supply o pension igh s (pension expec a- ions o he e i ed gene a ion in e ms o p oduc i i y o he wo king gene a ion), cis he eal con ibu ion a e (including subsidies ou o gene al e enues), zis he Figu e 1 GDP p oduced by consecu i e gene a ions (1 and 2) di ided be ween emune a ion o p oduc ion ac o s (R) and pension ans e (T). No e: O he ans e s in his igu e equal ze o. Gó a IZA Jou nal o Labo & De elopmen 2013, 2:2 Page 3 o 31 h p://www.izajold.com/con en /2/1/2 eplacemen a e 9 ,w(ba ) is he a e age wage, L W is he numbe o wo ke s, eis he employmen a e and L R is he numbe o e i ees. In equilib ium, he eplacemen a e is gi en by Equa ion (2) z¼ce dð2Þ Whe e dis he demog aphic dependency a io (d = L R /L W ). Gi en he demog aphic s uc u e, he employmen a io and he ins i u ional ame- wo k ( e i emen age and so o h), he eplacemen a e depends on he con ibu ion a e alone and does no depend on he ype o pension sys em o i s design. The pension ma ke de e mines he sha es o GDP alloca ed o each gene a ion. A key conclusion s emming om he simple model abo e is ha he way pension igh s a e exp essed does no ma e o he a e age le el o pensions as measu ed by he eplacemen a e 10 . In o he wo ds, nei he poli icians no inancial ma ke s can make a wiza dly ick and pe manen ly gene a e a eplacemen a e on a e age la ge han z, as i is p esen ed in Equa ion 2 11 . This conclusion b ings us back o he c ucial ade-o wi hin he pension sys em: whom shall he policy measu es p o ec –wo ke s o e i ees? Figu e 2 helps o answe ha ques ion. The s o y behind he igu e is as ollows. Ini ially z 1 was es ablished based on he consump ion needs o he e i ees. Then, c 1 equi ed o inance pension ans e s was imposed. I e lec s he en i e cos o he wo king gene a ion. I does no ma e he e whe he he cos is well de ined and explici ly paid as a con ibu ion, o he cos is hidden in gene al e enues spen on subsidisa ion o he pension sys em 12 . The goal o he sys em designe s was o keep he a es s able o o ine une hem i needed. Tha would ha e been he case i he popula ion s uc u e by age had s ayed s able (d 1 ). The ini ial dependency a io de e mining he ela ion o he eplacemen a e and he con ibu ion a e has been inc easing o se e al pas decades and i is expec ed o keep inc easing in he decades o come (d 2 ,d 3 ). In o de o keep he eplacemen a e a he le el p omised o e en legisla ed, he con ibu ion a e has o be subs an ially inc eased, and/o la ge open and/o hidden deb s ha e o be made (c 2 ,c 3 ). Gi en he demog aphic p ojec ions, i is e y likely, ac ually su e, he eplacemen a es will – soone o la e –be educed (z 4 ). Figu e 2 Con ibu ion a e s. eplacemen a e. Gó a IZA Jou nal o Labo & De elopmen 2013, 2:2 Page 4 o 31 h p://www.izajold.com/con en /2/1/2 The si ua ion as p esen ed abo e can be modi ied when, on op o he coun y’s GDP, he pension sys em is inanced ou o e enues om in es men s ab oad. Tha op ion is a ailable when he sys em is based on asse s aded in inancial ma ke s ( he so- called unded sys em). I he a es o e u n ha can be ea ned ab oad a e la ge han domes ic a es, hen eplacemen a es can be la ge han “z”as de ined in equa ion (1). This si ua ion can be called “exploi a ion”o “exchange”o capi al o labou . Ne e - heless, his op ion p obably also canno subs an ially inc ease “z”in he long un. In any case, in e na ional in es men can p o ide pension sys ems wi h mo e isk di e si ica ion. Aiming a an unchanged eplacemen a e means p o ec ing he e i ees a he expense o he wo ke s. Tha has been he dominan policy op ion h oughou Eu ope. Remune a ion o p oduc ion ac o s e lec less and less he supply and p oduc i i y g ow h ac o s because o he g owing cos o pension ans e s. Mo eo e , his si u- a ion leads o labou ma ke p oblems, and in gene al social p oblems aced by he younges gene a ion ying o en e he labou ma ke . They ail o s a hei adul li e enjoying he same le el o wel a e hei pa en s did. So as he “occupy mo emen ”, hey exp ess hei ange in he s ee s o Eu opean ci ies. They blame u hless poli icians and banke s o hei si ua ion. Those blamed a e in many cases guil y o i esponsibili y, g eed, sho -sigh edness and s upidi y. Howe e , hey a e no guil y o he loss o “demo- g aphic di idend”. Reduced esou ces (no di idend any mo e) a e spen on hose who a e al eady in, while hose who a e s ill ou , namely he young, ha e li le chance o join since he esou ces a e spen . The di icul si ua ion o he young Eu opean gene a ion can be seen ia da a on un- employmen as well as ia da a on he isk o po e y by age. These a e young Eu opeans who ha e been hi by he cu en di icul si ua ion. Table 1 and Table 2 p o ide da a on he isk o po e y by age di e ence be ween 2005, he p e-c isis imes, and 2010 when he e ec s o he c isis we e e ealed. The ables p o ide da a on he la ges six EU coun- ies and addi ionally G eece and Po ugal, belonging o he g oup o coun ies acing he la ges p oblems in he pe iod analysed on one side, and Sweden being in a ela i ely good si ua ion on he o he . Sweden is also impo an in his pape because o he pension issues discussed below. Table 1 A - isk-o -po e y a e by age (pe cen age change be ween 2005 and 2010 a e ans e s (2005 le el = 100%); selec ed coun ies) To al 65+ 50-64 25-49 18-24 -18 EU (27 coun ies) 100% 85% 101% 104% 108% 103% Eu o a ea (17 coun ies) 106% 80% 105% 114% 119% 114% Ge many 128% 105% 135% 131% 127% 143% Spain 105% 74% 110% 118% 140% 108% F ance 102% 65% 83% 110% 128% 124% I aly 96% 73% 90% 109% 104% 105% Poland 86% 195% 101% 76% 84% 77% Uni ed Kingdom 90% 86% 92% 95% 85% 89% G eece 103% 76% 95% 117% 129% 113% Po ugal 92% 76% 94% 97% 117% 95% Sweden 136% 153% 137% 133% 116% 128% Sou ce: Eu os a [ilc_li02], own calcula ions. Gó a IZA Jou nal o Labo & De elopmen 2013, 2:2 Page 5 o 31 h p://www.izajold.com/con en /2/1/2 In he ables an inc ease (Table 1) o a high le el (Table 2) o exposu e o he isk o po e y is ma ked in bold, while he opposi e is ma ked in i alic. The ables p esen a gloomy pe spec i e o young Eu opeans. In he coun ies p esen ed in bo h ables, he e i ees a e p o ec ed, while he si ua ion o he young is bad and also wo sening. The e a e wo exemp ions, namely Poland and Sweden (discussed la e in his pape ). P o ec ing he e i ees a he expense o he young is no only a p oblem o he la e g oup bu also a gene al social ap. Socie ies ace an asymme ic choice. P o ec ing he old con ibu es o he weake g ow h o wel a e p oduced by he young, whe eas p o ec ing (o a leas aking in o accoun ) he in e es s o he young con ibu es o he s onge g ow h o wel a e o bo h he old and he young. T adi ional pension sys ems, as hey s ill domina e in Eu ope, ocus hei a en ion on he old while he in e es s o he wo king gene a ion (which is emune a ion o hei wo king ac i i y) is esidual in hose sys ems. The old a e assumed o be weak while he ac i e gene a ion is assumed o be s ong. So, he old should be suppo ed by he ac i e gene a ion. Tha was co ec in he pas when li e expec ancy was much lowe and people e i ed a ound he age o 70 anyway 13 . The e is much less a ionale when people a e li ing much longe and e i e a a ound he age o 60. Those who e i e a e no -ye -old (NYO). Financing hei pensions educes esou ces a ailable o he emune a ion o p oduc ion ac o s p o ided by he wo king gene a ion, as well as con ibu ing o he g owing expos- u e o he young o he isk o po e y, as p esen ed in Table 1 and Table 2. Redis ibu ion channels should be adjus ed o he change. P obably hey should be pa ially e e sed. I no , hey will be c ea ing social p oblems a he han mi iga ing hem. 3. The old pension sys em in Poland A ypical adi ional con inen al Eu opean pension sys em exis ed in Poland un il he end o 1998. The sys em applied a de ined bene i (DB) egime. Expendi u es on pensions we e high by Eu opean s anda ds (one o he highes ). Acco ding o p ojec ions, pe spec i es o he u u e we e gloomy. The old sys em was non- anspa en . People did no unde s and i s ules and did no ca e abou hei own pa icipa ion in he sys em (see nex sec ions). The sys em did Table 2 A - isk-o -po e y a e by age as compa ed o he isk o he en i e socie y (selec ed coun ies) To al 65+ 50-64 25-49 18-24 -18 EU (27 coun ies) 16.4 98% 82% 90% 129% 125% Eu o a ea (17 coun ies) 16.1 96% 84% 91% 130% 124% Ge many 15.6 90% 109% 90% 121% 112% Spain 20.7 105% 87% 90% 112% 127% F ance 13.3 80% 62% 89% 177% 135% I aly 18.2 91% 72% 97% 126% 136% Poland 17.6 81% 93% 91% 119% 128% Uni ed Kingdom 17.1 125% 84% 80% 122% 119% G eece 20.1 106% 86% 90% 137% 114% Po ugal 17.9 117% 90% 83% 104% 125% Sweden 12.9 120% 46% 82% 229% 102% Sou ce: Eu os a [ilc_li02], own calcula ions. Gó a IZA Jou nal o Labo & De elopmen 2013, 2:2 Page 6 o 31 h p://www.izajold.com/con en /2/1/2 no know people un il he momen hey came in and claimed a bene i . Con ibu ions we e inc easing e y apidly, con ibu ing o g owing ax dis o ions on one hand and g owing labou cos s and/o declining ne emune a ions on he o he . Howe e , he key p oblem o he sys em was he e y low e i emen age. Ea ly e i e- men schemes exis in many coun ies. His o ically hey e lec he special p e e ences gi en o selec ed g oups, such as ailway wo ke s o mine s, which we e pa icula ly impo an du ing he p ocess o indus ialisa ion. Thei p i ileges a ied ac oss coun- ies. In Poland, he ea ly e i emen scheme o ailway wo ke s da ed om he ea ly wen ie h cen u y. The p i ilege o mine s was based on he c ucial ole coal mining used o play in he economy o Poland. The ea ly e i emen schemes became a pa o he wel a e s a es du ing he pos Sec- ond Wo ld Wa p ospe i y pe iod. People we e g an ed e i emen a ages ea lie han becoming eally old. Tha d ama ically changed he ole o pensions. Ini ially hey we e o suppo he e y old. As a pa o he wel a e s a e hey become a “gi ” o he no - ye -old. The demog aphic di idend gene a ed enough esou ces o pay o ha . Pa a- doxically, du ing he same p ospe i y pe iod, li e expec ancy subs an ially inc eased. Consequen ly, he pe iod o ime spen on a pension inc eased om “bo h ends”: wo ke s s a ed e i emen ea lie and enjoyed i ill la e ages. In Poland ha p ocess was weake . Pe manen sho ages in he labou ma ke p e en ed pushing wo ke s ou o jobs oo ea ly. Also li e expec ancy did no inc ease o he le els obse ed in Wes e n Eu opean coun ies. Howe e , jus a e ma ial law (1981), a speci ic ac o s ongly changed he pension landscape. In o de o p e en so- cial un es in ha e y poli ically u bulen pe iod, au ho i ies in oduced a massi e ea ly e i emen scheme o many g oups o wo ke s. The o icial explana ion o ha was o acknowledge hei ha d wo k. In p ac ice, he scheme was a ailable o a la ge numbe o wo ke s i espec i e o he ype o wo k. Al hough he legally obliged public sys em e i emen age s ayed unchanged, he ac ual a e age e i emen age s a ed d opping yea by yea ollowing he ageing o coho s en i led o he ea ly e i emen 14 . The low e i emen age oge he wi h he popula ion ageing led o a d ama ic inc ease in pension sys em expendi u e. Up o 1981 he con ibu ion a e was 15.5 pe cen o he en i e social secu i y. In he mid 1990s he a e was 45 pe cen and was insu icien . 4. A wo- old challenge o Poland: ansi ion owa ds a ma ke -based econ- omy, and wel a e ca ch-up du ing he imes since he demog aphic di idend was los Poland s a ed he ansi ion om a cen ally planned economy owa ds a ma ke based economy a ound 1989 15 . The ansi ion was success ully comple ed by he la e 1990s 16 . The ansi ion p ocess u ned smoo hly in o he Eu opean Union p e-accession p ocess which was also success ully comple ed in 2004. Since ha momen (being s ill less a luen han mos o he Eu opean coun ies) Poland has aced uni e sal challenges simila o hose aced by he en i e EU. In oducing a ma ke -based economy o ally changed economic and social li e in Poland. A sha p social and economic upg ade o he coun y was achie ed. Howe e , ha success was di icul o people. They had o adjus o a new en i onmen . In oducing Gó a IZA Jou nal o Labo & De elopmen 2013, 2:2 Page 7 o 31 h p://www.izajold.com/con en /2/1/2 ma ke -based policies was he ask, while he in oduc ion o Wes e n ype wel a e ins i u- ions was an expec ed ewa d. I was an a ac o , a d i ing ac o behind he ansi ion e o s. Ac ually, he p e- ansi ion wel a e ins i u ions in Poland, as well as in o he coun- ies o he o me So ie bloc, we e less de eloped ye simila o hose in he Wes . Only he o e all low p oduc i i y did no allow o any gene ous ou comes om hei exis ence. Ne e heless, people expec ed no only an inc ease in wel a e bu also an easy way o achie e ha a e joining he Wes , whose wel a e egime was ypically pe cei ed as he kind o policy ha p o ided wel a e wi hou wo k - as i wel a e was g an ed o Wes e n Eu opean ci izens jus because hey we e EU ci izens. Ac ually, ha pe cep ion is, o an ex en , also common in he Wes e n pa o con inen al Eu ope 17 . A he end o he 20s cen u y he demog aphic/ inancial py amid wen bank up h oughou Eu ope. The di idend uelling he Eu opean Social Model was los . E en e y a luen coun ies became less a luen han hei ci izens ended o assume. On op o ha , o me ansi ion coun ies los hei p e iously clea and commonly accep ed di ec ion o de elopmen . A e comple ing one ansi ion (in a na ow sense) hey had o pass a second. The ansi ion om a demog aphic di idend based wel a e s a e o a wo k e o based egime will be coped wi h alongside he es o he EU. Howe e , in his la e case, he e is no clea p esc ip ion o a success ul economic and social policy. Pension e o ms a e on he op o policy agendas h oughou Eu ope. Coun ies ha e ei he implemen ed e o ms, a e in he cou se o e o ms, o a e a leas planning e o ms o hei public pension sys ems. Poland is in he g oup ha ha e al eady implemen ed a adical pension e o m. This e o m was based on he documen “Secu i y h ough Di e si y”and was implemen ed in 1999. Figu e 3 indica es selec ed c ucial momen s o he e o m. The wo challenges, namely he ansi ion and he pension e o m ( oge he wi h o he s uc u al e o ms no discussed in his pape ), we e no di ec ly connec ed. Ra he , hey we e wo di e en pa hs. Howe e , he ime coincidence c ea ed a unique oppo uni y o Figu e 3 Time pe spec i e o he yea s since 1989 a . a No e: Figu e 1 p esen s selec ed momen s in he ecen his o y o Poland. The uppe po ion o he igu e e e s o he pension e o ms. Blue a ows indica e success ul ou comes o he e o m p ocess, while he ed a ows deno e ailu es (see la e in he ex ). S D – “Secu i y h ough Di e si y”–a bluep in o he pension e o m design (see Gó a and Ru kowski 1997). Gó a IZA Jou nal o Labo & De elopmen 2013, 2:2 Page 8 o 31 h p://www.izajold.com/con en /2/1/2 The simila i y o he new sys ems in oduced in Poland and Sweden is no ob ious as in public deba e he pension sys em issues we e ypically mixed up. The e a e h ee le els a which pension sys ems can be discussed (see Figu e 4). The i s e e s o he mos gene al issues such as social goals and undamen al economic p inciples. Exam- ples: alloca ion o income should co e he en i e li e o each pa icipan ; expendi u e on pensions canno g ow as e han GDP; a pension sys em is public i i co e s he en i e popula ion and applies he same s anda d ules ac oss all pa icipan s, while i is p i a e i i co e s indi iduals o g oups whose pa icipa ion is based on indi idual decisions and esponsibili y. The second le el e e s o he me hods applied wi hin he pension sys em. Me hods should e lec he si ua ion and jus be e icien . A he hi d le el, egula ions should se e he me hods and should be logical, clea and anspa en . A good design o pension sys em should be p epa ed on all h ee le els. They a e all impo an . Howe e , hey should be subo dina ed acco ding o he pa e n: goals ➔ me hods ➔ egula ions. Discussions can lead o a a ional conclusion i hey ollow ha Table 4 Manda o y con ibu ions in Poland be o e and a e implemen a ion o he new pension sys em To al NDC indi idual accoun FDC indi idual accoun O he elemen s o he sys em be o e 1-Jan-1999 Manda o y con ibu ion 36.59 a –– 36.59 since 1-Jan-1999 Manda o y old-age (OA) con ibu ion [un il Ap il 2011] 19.52 12.22 7.3 – Manda o y old-age (OA) con ibu ion [since May 2011] 19.52 17.22 b 2.3 – O he manda o y social wel a e (NOA) con ibu ions 17.07 –– 17.07 a Equi alen o 45 pe cen (a e g ossing-up o labou income). b In ac since 2011 pa icipan s ha e had wo NDC accoun s. Con ibu ion (17.22) is spli be ween hem: 12.22 and 5.00 pe cen age poin s. La e in he ex an addi ional explana ion is p o ided. Figu e 4 Th ee le els o discussion on pensions. Gó a IZA Jou nal o Labo & De elopmen 2013, 2:2 Page 15 o 31 h p://www.izajold.com/con en /2/1/2 pa e n. In public deba e he opposi e pa e n is commonly a gued. Ra ional a gumen s a e discussed wi h espec o egula ions wi h li le e e ence o he goals behind he public pension sys em. Fo ins ance, is he e any sense behind manda ing people o pa icipa e (pay con ibu ions) i hei pensions may no co e hei en i e li e (in he case o a lump-sum paymen ). Regula ions a e easie o unde s and. They a e also con enien o incon enien o he a ious ac o s ac i e in he a ea o pensions, so ha egula ions a e in he cen e o public deba e as well as expe deba e, i espec i e o hei ela ion o he me i s o he pension sys em i sel . The Swedish and Polish sys ems a e almos iden ical a he i s le el. They a e simi- la a he second. Di e ences a e mos ly loca ed a he hi d le el. This is opposi e o he simila i ies be ween he Polish and he Chilean sys ems. Simila i ies can be ob- se ed a he hi d le el while he sys ems di e s ongly a he second and especially so a he i s le el. In pension li e a u e wo app oaches a e dominan , namely he social policy based and he inancial ma ke app oaches. Discussion be ween p oponen s o he wo app oaches is p obably endless. Nei he o he dominan app oaches ocuses on eco- nomics in a way ha would pe mi analysing he long e m ea u es and uni e sal co e age o he pension sys ems oge he . The ocus on he hi d le el o discussion, which domina es public deba e, causes s ong misunde s anding o he essence and oo s o he Polish pension sys em. Table 5 highligh s key di e ences be ween he Secu i y h ough Di e si y and he h ee pilla app oach o pension e o m (see also Gó a (2003)). The idea behind he h ee pilla app oach is he pa ial p i a isa ion o pensions: he second pilla is he same as he hi d one, i is jus manda ed. A pa o he public sys em is pushed ou o he p i a e sec o . The idea behind Secu i y h ough Di e si y is he opposi e. I is he di e si ica ion o isks wi hin he public sys em: p i a e asse manage s a e hi ed o un a pa o he public sys em using inancial ma ke s. The sys- em s ays public and is s ill ocused on unchanged social goals. Running a pa o he pension sys em has jus been con ac ed ou o he p i a e sec o . This a angemen can be called a kind o public-p i a e pa ne ship wi hin he pension sys em. Table 5 Al e na i e app oaches o pension e o m Typical “ h ee pilla ” e o m Secu i y h ough Di e si y ▸Downsized and possibly a ionalised old sys em, named he “ i s pilla ” ▸Spli ing old sys em in o OA and NOA ▸Te mina ion o he OA pa o he old sys em ▸New “ unded”sys em named “second pilla ”c ea ed in he oom le a e he downsizing o he old sys em ▸C ea ion o en i ely new OA sys em ▸Ra ionalisa ion o NOA ▸Con ibu ion spli be ween wo accoun s wi hin he new OA ▸Con ibu ion spli be ween he old and he new sys em ▸Fi s indi idual accoun –non- inancial; a e o e u n de e mined by GDP g ow h ▸Swi ching p ocess based on indi idual p e e ences ▸Second indi idual accoun – inancial; a e o e u n de e mined in inancial ma ke s ▸Va ie y o pay-ou me hods in he “second pilla ”▸Annui isa ion o bo h accoun alues ▸P omo ion o a ious o ms o addi ional sa ings ▸P omo ion o a ious o ms o addi ional sa ings, named he “ hi d pilla ” Gó a IZA Jou nal o Labo & De elopmen 2013, 2:2 Page 16 o 31 h p://www.izajold.com/con en /2/1/2 The ype o app oach applied in he Polish pension sys em is commonly mispe cei ed. In he pension deba e bo h expe s and he gene al public a e hos ages o he e minology used and assump ions in ui i ely made. The mispe cep ion leads o public communica ion p oblems as well as economic p oblems s emming om he s anda d accoun ing me hods applied. 6. Minimum gua an ee and he cap on con ibu ions The en i e new old-age pension sys em is based on indi idual accoun s. This applies o bo h he pa un by ZUS as well as he one un by PTEs. The way bo h ypes o ac- coun a e managed is di e en . Howe e , in Poland he ole hey play in he public sys- em is exac ly he same, namely hey p o ide people wi h a me hod o income alloca ion o e hei li e cycle. The e is no o he ole he accoun s play; he e is no o he elemen o he sys em o he han he accoun s. This means ha he income e- dis ibu ion – ypically p esen in public pension sys ems –has been mo ed om he pension sys em o he budge . The sole ole o he sys em is jus income alloca ion. Shi ing edis ibu ion ou o he OA pension sys em o he budge ollows he demo- g aphic changes. On op o ha , p o iding edis ibu ion ia he budge a y channel has a numbe o ad an ages o e doing i ia he pension sys em. See Table 6. In he OA sys em i sel a minimum bene i gua an ee does no exis . On he day o e i emen accoun alues a e u ned in o annui ies i espec i e o he amoun accumu- la ed and consequen ly i espec i e o he le el o co esponding bene i s. Howe e , i a bene i is below a le el ha is cu en ly legally assumed oo low, hen a budge a y sub- sidy comes in o inance a op-up paymen inc easing he bene i le el o he minimum le el. This le el is le o poli ical disc e ion. Figu e 5 illus a es he idea o he op-up pension bene i . This app oach has been implemen ed in a simila way in bo h Poland and in Sweden. The design o he Polish sys em in ol es a cap on con ibu ions. Sys em pa icipan s pay con ibu ions based on s anda d ules bu only o a le el co esponding o 30 imes he a e age wage in he economy. Income abo e ha le el is no subjec o con ibu- ion. This leads o an imposed cap on u u e bene i s. Tha a angemen looks like a p i ilege o he ich. This is no he case since he sys em is no ax-based any mo e. Con ibu ions o bo h NDC and FDC accoun s a e indi idual sa ings. The ich can ake ca e o hemsel es, addi ionally sa ing ou side he public sys em. The cap helps o educe ax dis o ions e en u he han hey s em om using indi- idual accoun s wi hin he sys em. The con ibu ions a e paid obliga o ily. I hey a e oo high hen ax pe cep ion and consequen ly ax dis o ions will always appea . A lack o edis ibu ion causes he addi ional con ibu ions o no help he less a luen Table 6 Channelling edis ibu ion ia he budge e sus ia he pension sys em Via he budge Via he pension sys em Redis ibu ion base La ge Smalle Who pays (1) Gene ally axes a e usually p og essi e –so he ich pay mo e Pension con ibu ions a e linea –no p og ession Who pays (2) Labou and capi al Labou Adjus abili y o changing needs Rela i ely easy –each yea Rigid –long e m liabili ies Gó a IZA Jou nal o Labo & De elopmen 2013, 2:2 Page 17 o 31 h p://www.izajold.com/con en /2/1/2 pa icipan s. While no paying addi ional con ibu ions, he ich ha e o pay axes on he addi ional income hey ea n. 7. Implemen a ion o he new OA sys em The new sys em eplaced he old one o people bo n a e 31 Decembe 1948. Tha means ha many people we e s ongly ad anced in hei wo king ca ee s. D opping he old sys em did no mean a loss o p e iously acqui ed pension igh s. They we e ecalcula ed in o accoun alues co esponding o he discoun ed alue o hei u u e pensions o be paid beginning om he e i emen age o 60 and 65 o women and men espec i ely. Thanks o ha , he closu e o he old sys em was de ini i e. Calcula ion o he ini ial capi al was ela i ely easy. This was done by ZUS. Howe e , he ewe ealo o p oblemss emming om heine iciencyo heoldsys em.In ad- i ional sys ems people pa icipa ed anonymously. The sys em s a ed ecognizing hem om he momen hey applied o a pension. Each coho o wo ke s p og essed h ough ha p ocedu e be o e 1999. Howe e , hesewe epeoplewhowe eap- p oaching e i emen age so hey we e p epa ed. Knowing he ules o he old sys em, hey had all documen s eady. The pension e o m had suddenly exposed millions o young people o a si ua ion hey we e no p epa ed o . They had o collec a numbe o documen s heyhadno expec ed obenecessa y 35 . The new sys em will ne e need any addi ional documen s in he u u e. So collec ing hem was no an addi ional adminis a i e bu den, bu jus shi ing o he bu den in a li e ime. Re o m o he pension sys em in Poland was in oduced on 1s Jan. 1999. The old sys em ceased o exis o people bo n a e 31 s Dec. 1948. They we e co e ed by he new sys em au oma ically. Olde people did no ha e any op ion o swi ch o he new sys em. Ac ually, he e was no swi ching a all, nei he o he young no o he old. See Table 7. The employed we e only o make a decision whe he o spli hei OA con ibu ion be o e he end o Sep embe 1999. No decision was legally in e p e ed as he decision o lea e he con ibu ion non-spli and consequen ly o ha e he en i e con ibu ion managed by ZUS. Figu e 5 Minimum gua an ee and he cap on con ibu ions. Gó a IZA Jou nal o Labo & De elopmen 2013, 2:2 Page 18 o 31 h p://www.izajold.com/con en /2/1/2 The choice o ha pa icula implemen a ion scheme by age was c ucial o he ol- lowing easons: ▸A oiding gene al p o es s ypically aised by e i ees and people close o e i emen ; ▸A oiding speci ic p o es s agains inc easing he e i emen age; ▸A oiding a emp s o make a bi age by hose who could make a choice o swi ch o no ; ▸A kind o choice was in oduced wi hin he new sys em o gi e people mo e com o . I should be no ed ha he decision o spli o no spli he OA con ibu ion did no ha e a s ong impac on u u e pensions. As discussed ea lie in his pape , in he public pension sys em he eplacemen a e weakly –i a all –depended on he way ( ype o pension sys em) in which con ibu ions we e managed. The gain ollowing he spli o con ibu ion was in isk di e si ica ion. The dis ibu ion o he decisions sugges s he choices o ages as p esen ed in Table 7 co esponded well o indi idual p e e ences. The as majo i y o hose who in 1999 we e only sligh ly o e 30 had decided o spli hei con ibu ions while he as majo i y o hose who we e abou o each 50 had decided o ha e hei con ibu ion unchanged. Tha dis ibu ion also sugges s ha he people had made qui e a ional decisions. Figu e 6 illus a es he dis ibu ion by age o hose decisions. Decisions on he spli o con ibu ions wi hin he new OA sys em we e in e p e ed by people, as well as he media and poli icians, as decisions “ o join he new sys em”. Tha was a comple e misunde s anding as he decision e e ed o wo a ian s o he new sys em. Lea ing he con ibu ion non-spli also mean co e age by he new sys em. The e we e no pa icula consequences om his Table 7 In oduc ion o he new sys em (age g oups) New sys em (people bo n a e 31 Dec. 1948) Old sys em (people bo n be o e 1 Jan. 1949) People bo n a e 31 Dec. 1968 People bo n be o e 1 Jan. 1969 Au oma ically co e ed by he new sys em; OA con ibu ion au oma ically spli be ween wo accoun s [NDC +FDC] Au oma ically co e ed by he new sys em; OA con ibu ion ei he spli be ween wo accoun s o paid in o one accoun [(NDC+FDC) o NDC] S ay in he old sys em (no possibili y o pa icipa e in he new one); no accoun s Figu e 6 Age dis ibu ion o OA con ibu ion spli in 1999 o hose who had he op ion o choose. Sou ce: Chlon-Dominczak (2002). Gó a IZA Jou nal o Labo & De elopmen 2013, 2:2 Page 19 o 31 h p://www.izajold.com/con en /2/1/2 misunde s anding, i jus indica ed a s ong need o public educa ion o be o- cused on he a ea o pensions. Implemen a ion o he new sys em mean an inc ease in he e ec i e e i emen age by a ound 5 yea s, as he esul o changing he o me s a u o y e i emen age o be- come a minimum age. Pension p i ileges we e le ou side he new sys em. Acco ding o new legisla ion he e was no way o e i e be o e eaching he new minimum age. Mo eo e , e i ing ea lie would mean ex emely low pensions, which is he e ec o he ac ua ially neu al way o hei calcula ion. People in gene al we e in a ou o he new sys em, so hey implici ly accep ed he e i emen age. These we e a he young people so hey had no s a ed hinking abou hei own e i emen ye . People o e he age o 50 we e le ou o ha discussion. The cos o such an app oach was in no iscal sa ing jus a e he new sys em was in o- duced. Today he sa ing is la ge and is inc easing yea by yea , once new egula ions on he e i emen age ha e become e ec i e. 8. B idging pensions The inc ease in he e i emen age –e en i no con es ed in 1999 –was e y ab up . In o de o a oid ex-pos p o es s, he concep o b idging pensions was de eloped. I was pa o he en i e p ojec o e o m implemen ed in he la e 1990s. The e mina ion o ea ly e i emen schemes was legisla ed oge he wi h he en i e e o m bill package. Howe e , he legisla ion did no include egula ions on how he b idging pensions would be o ganised and inanced. Tha legisla ion gap was no illed du ing he 10 yea s a e adop ion o he basic law bills. Poli icians we e a aid o ouching ha issue again, and his led o p oblems because he inal da e by which all ea ly e i emen schemes we e o be e mina ed was 2006. Poli icians wice pos poned ha da e by one yea . The p oblem was e en ually sol ed a he end o 2008 when he law on b idging pensions was inally passed h ough Pa liamen . Passing he law h ough Pa liamen was no easy bu inally success ul. The law had been e oed by he P esiden and sen back o Pa liamen whe e i was o ed on again. Fo una ely one opposi ion pa y, namely SLD, joined he uling coali ion causing he P esiden ’s e o o be o e uled. The law became e ec i e om he beginning o 2009 and o ally changed he pension landscape in Poland. Key ea u es o he b idging pensions: ▸Based on labou medical g ounds. The e a e no o he c i e ia o selec ing he en i led wo k pos s. Only occupying such a pos o a de ined pe iod en i les wo ke s o b idging. No special a angemen s exis o any pa icula g oup o p o ession. ▸Bene i s a e p o ided only o he sho pe iod o ime be ween exi ing he en i led wo k pos and eaching he minimum e i emen age. The pe iod is de ined as no mo e han ( ypically) 5 yea s o less. ▸Financed (addi ional con ibu ion) by employe s who employ wo ke s in pos s ha expose hei heal h o isks such as wo king unde g ound, unde wa e o hose equi ing a special heal h s a us, such as an ai c a pilo . ▸Planned o a ini e pe iod. E en ually b idging pensions need o be exchanged o equali ica ion p og ammes allowing people o wo k longe in o he pos s. Gó a IZA Jou nal o Labo & De elopmen 2013, 2:2 Page 20 o 31 h p://www.izajold.com/con en /2/1/2 The goals behind he concep : ▸P o ec ing wo ke s heal h –selling heal h o he p i ilege o e i ing ea ly should no be possible ▸Elimina ion o hidden subsidisa ion o hose employe s who bene i om employing wo ke s in pos s exposed o heal h isks – hose who bene i om his employmen should co e he cos o he subsequen heal h isks ▸Inc easing employmen The in oduc ion o he b idging pensions did no ex end he scale o ea ly e i emen schemes. Ra he he opposi e. This was a way o p o ec he wi hd awal o he schemes. The way b idging pensions we e de ined (labou medical) and hei clea cos mean ha he numbe o po en ial cases was oughly 1/10 h o he p e iously possible ea ly e i emen schemes 36 . Mo eo e , ea ly e i emen was condi ional on addi ional con i- bu ions paid ea lie by employe s. The en i e scheme was empo al co e ing only hose wo ke s who had s a ed wo king in special condi ions o in a special capaci y be o e 1999. The law on b idging pensions is commonly e e ed o as he mos impo an achie emen o he go e nmen du ing he e m 2007-2011. 9. Replacemen a es and he dis ibu ion o income by age As al eady p esen ed, he le el o OA pension bene i s (as exp essed by he eplacemen a e) did no depend on he pa icula ype o pension sys em. The e was no public pen- sion e o m ha could inc ease ha le el wi hou educing emune a ion o p oduc ion ac o s and ans e s o he han OA bene i s. Ne e heless, people s ill pe cei ed pension sys ems ia he bene i s hey paid –o ac ually p omised. Consequen ly, changing a sys- em ha gene a ed in la ed pension expec a ions o one ha p o ided ue in o ma ion on u u e pensions in ad ance caused social eelings ha he change was bad. E en young people om whom he e o m eleased he pe spec i e o he highe con ibu ions and/o axes el unhappy. In many cases hey e en p o es ed. Pension expec a ions do ma e o social and economic de elopmen . I hey a e in la ed, p oblems will a ise soone o la e . F om he indi idual’s iewpoin , adi ional DB pension sys ems wo k ex pos , which means pension sys em liabili ies a e exp essed in mone a y e ms, a e a pe son e i es. In his way, people may and do expec mo e han a pension sys em –ac ually he wo king people –can and wish o pay. The indi idual’s pe cep ion does no ake in o accoun equa ion 2 (ea lie in he ex ). The GDP cons ain on expend- i u e is a mac o concep , while a he indi idual le el people s ill expec wha was p omised. The new Polish sys em wo ks along an ex an e app oach: no p omises, jus well-de ined sys em liabili ies as hey a e now. This con ibu es o he p e en ion o such in la ed expec a ions. I he e en ual le el o pensions is gi en by he popula ion s uc u e and he con ibu ion a e, hen i seems be e ha people expec wha hey will e en ually ecei e a he han expec ing mo e. In o he wo ds, a educ ion o expec a ions seems be e han cu ing ac ual bene i s down. In a DC sys em, eplacemen a es addi ionally depend on he e i emen age. The la e a pe son e i es, he highe he bene i . Mo eo e , in ac ua ially neu al sys ems, each addi ional pe iod wo ked be o e e i emen adds mo e o he le el o he bene i Gó a IZA Jou nal o Labo & De elopmen 2013, 2:2 Page 21 o 31 h p://www.izajold.com/con en /2/1/2 han he p e ious pe iod. Consequen ly, he issue o eplacemen a es becomes mo e complex. The le el o he pension is a unc ion o he e i emen age. So asking he ques ion ‘wha will he le el o u u e bene i s be’is no p ecise. One should addi ion- ally speci y he e i emen age co esponding o he ques ion. I espec i e o he abo e, u u e pensions will be smalle han hose he p e ious gen- e a ions enjoyed. This will jus be he e ec o he popula ion change. Pensions sys ems can simply adjus o ha change o no . In he la e case demog aphy will be s onge anyway and he adjus men will come anyway –jus mo e d ama ically and socially pain ully. The ex an e a angemen s a e be e om ha iewpoin . They also p o ec he young much mo e han he ex pos a angemen s. Pos poning he ine i able adjus men leads o pain ul e en ual disillusionmen . The ex an e app oach was implemen ed well in ad ance, allowing o a g adual educ ion o pension expec a ions. People co e ed by he new sys em (bo n a e 1948) do no ha e hei pension igh s simply educed. Thanks o he ini ial capi al ha ca ies o e he igh s al eady acqui ed, he eplacemen a es will go down subs an ially bu slowly, which means he ac ual le el o pensions will no subs an ially change yea o yea 37 . I is oo ea ly –and in gene al e y di icul – o measu e he e ec o he im- plemen a ion o he new pension sys em on he change o dis ibu ion o he isk o po e y by age (as p esen ed in Table 1 and Table 2) and o analy ically dis in- guish his e ec om he e ec s s emming om o he possible easons. Howe e , a clea coincidence can be obse ed in he ables. The shi o ha isk om younge o olde age g oups has been obse ed o Poland and Sweden –coun- ies ha in oduced he same ype o pension sys em ha in ol es he in e es s o he young. The o he coun ies in imes o c isis p o ec ed he olde gene - a ion much mo e s ongly han he younge gene a ion. I should be s essed ha he e i ees’exposu e o he isk o po e y is much smalle , e en a e he sha p change we obse ed ecen ly. Figu e 7 Pension expendi u e un il 2060. Sou ce: Eu opean Commission (2009). Gó a IZA Jou nal o Labo & De elopmen 2013, 2:2 Page 22 o 31 h p://www.izajold.com/con en /2/1/2 P o ec ing he in e es s o he young can also be de i ed om p ojec ions o u u e pension expendi u e 38 . Figu e 7 illus a es he p ojec ions. No going in o de ail, one can see he p ojec ed e ec o Poland is di e en om all o he cases 39 . Poland will enjoy a dec ease ins ead o smalle o la ge inc eases in pension expendi u e in he de- cades o come. This do no mean a educ ion in pensions – his means emune a ion o p oduc ion ac o s will be able o g ow as e han p oduc i i y g ow h wi hou mon- e a y in la ion, which in u n will con ibu e o s onge long e m economic g ow h. 10. Missing elemen s o he new pension sys em, ailu es in he e o m, and a piece o unexpec ed p og ess E en hough he ini ial design o he new pension sys em has almos en i ely been in oduced, a numbe o ailu es can be poin ed ou . Mos o hem a e he ic ims o poli ical p ocess; some mino ones a e ela ed o he design i sel as i was d awn. Co e age ailu es - g oups s ill ou side he public sys em The public pension sys em should, by de ini ion, co e he en i e popula ion. The sys em so named ‘public’in Poland (as well as in many o he coun ies) has le se e al special g oups ou side. So he new sys em does no co e he en i e popula ion. The excep ions a e: ▸Ag icul u al wo ke s (ac ually peasan s) ▸Judges and p osecu o s (exemp ed om 1999; judges because o he cons i u ion; p osecu o s by poli ical ac ion) ▸Uni o med se ices (a my, police, i emen and simila ; exemp ed om 2002; by poli ical ac ion) ▸Mine s (exemp ed om 2005; s ee io s jus be o e elec ions) Each o he excep ions has a di e en his o y and backg ound. Ag icul u e –lagging s uc u al adjus men The la ge sha e o ag icul u e in o al employmen is a sign o s uc u al unde de elop- men 40 , and has pension sys em consequences. Be o e 1972 ag icul u al wo ke s we e no co e ed by any pension sys em. Ins ead o co e ing hem in he public sys em, a special sys em was es ablished o his g oup (KRUS). Tha sys em is in ac a mix u e o a ype o social assis ance o he poo li ing in he coun yside, and a me hod o no paying con i- bu ions. Mo e han 90 pe cen o KRUS expendi u e is co e ed by a budge a y subsidy, so i is a me hod o edis ibu ion om non-ag icul u al wo ke s o ag icul u al wo ke s. The mos impo an p oblem s emming om ha is he mo i a ion o ag icul u al wo ke s o s ay in ag icul u e ins ead o lowing o non-ag icul u al sec o s o he economy. A e age p oduc i i y in ag icul u e is e y low –many imes lowe han in non-ag icul u al sec o s. S aying in ag icul u e is a loss o he ag icul u al wo ke since e en a low income ou side ag icul u e would be highe han wi hin i . Mo eo e , he en i e economy would bene i om he shi o a la ge olume o ag icul u al wo ke s o non- ag icul u al sec o s. Such a shi would no be easy and would depend on hei Gó a IZA Jou nal o Labo & De elopmen 2013, 2:2 Page 23 o 31 h p://www.izajold.com/con en /2/1/2 adap abili y. Howe e , poli ical obs acles p e en he p ocess om s a ing on a sub- s an ial scale 41 . S ong g oup p essu e Judges and p osecu o s we e s ong enough o s ay ou side he sys em om he e y beginning (1999) and ecei e gene ous paymen s o he en i e ime a e s opping hei ac i i y, which hey may do ela i ely ea ly. Misuse o he pension sys em Uni o med se ices we e ini ially co e ed by he new sys em unde s anda d ules. The only di e ence was ha only he newcome s since 1999 we e co e ed. The es s ayed in hei own p e ious sys em. Tha was changed in 2002 o easons s emming om hei commande s’s ong willingness o ha e he “pension”sys em as a de ice o implemen s a policy. Mine s and s ee io s Mine s we e also co e ed by he new sys em in 1999. The mine s wen on s ike o p e en being co e ed, bu we e no success ul. Tha was a g ea achie emen . In 2005 he mine s ac ed again and chose a e y delica e momen . Two elec ions we e sched- uled o he same yea (e en he same pa o ha yea ), namely he p esiden ial elec- ions and pa liamen a y elec ions. Mine s’ io s in Wa saw igh ened poli icians enough o e en ually accep all hei demands. The mos impo an was o wi hd aw hem om he public sys em. The key a gumen he mine s s essed was he absence o he p omised law on b idging pensions. Poli icians will ha e o cope wi h hese ou di e en challenges. Each will ha e o be deal wi h in a speci ic way. One good addi ional ou come o he anspa en ules wi hin he public sys em is ha he es o he popula ion knows and unde s ands mo e on pensions, so is less keen o accep special a angemen s o g oups emaining ou side o he sys em. Design ailu es The NDC pa o he sys em plays a c ucial ole. Implemen a ion o FDC was mo e isible and unde s andable o he people so i played an addi ional ole as an a ac o . I was ac ually he implemen a ion o NDC ha was he essence o he e o m. The ins i u ional s uc u e behind he NDC pa o he sys em has no ye been de eloped well enough. The e should be a specialised ins i u ion unning his pa o he sys em. I would also be good i p ope y igh s o accoun alues we e de- ined and gi en o pa icipan s. By no ha ing in place hese elemen s o he ins i u- ional in as uc u e, he NDC pa is s ill unde de eloped and no ully unde s ood by i s pa icipan s. An illogical a angemen has been adop ed o beques s. They may be le wi h e- spec o he FDC pa , while hey may no wi h espec o he NDC pa o he sys em. The same a angemen should apply o bo h o hem. Too lean incen i es and egula o y mechanisms o educe he cos s o unning he pension sys em. The newly in ol ed PTEs ha un he open pension unds began wi h Gó a IZA Jou nal o Labo & De elopmen 2013, 2:2 Page 24 o 31 h p://www.izajold.com/con en /2/1/2 O ens ein MA (2008) P i a izing Pensions: The T ansna ional Campaign o Social Secu i y Re o m. P ince on Uni e si y P ess, P ince on and Ox o d Palme E (2002) “Swedish Pension Re o m: How Did I E ol e, and Wha Does I Mean o he Fu u e?”. In: Felds ein M, Siebe H (eds) Social Secu i y Pension Re o m in Eu ope. Uni e si y o Chicago P ess, Chicago Van de Kaa DJ (1999) Eu ope and i s popula ion: he long iew. In: an de Kaa DJ, Le idon H, Gesano G, Okólski M (eds) Eu opean Popula ions: Uni y in Di e si y. Kluwe Academic Publishe s, Do d ech Wo ld Bank (1994) A e ing he Old Age C isis: Policies o P o ec he Old and P omo e G ow h. Wo ld Bank, Washing on D.C. doi:10.1186/2193-9020-2-2 Ci e his a icle as: Gó a: Poli ical economy o pension e o ms: selec ed gene al issues and he Polish pension e o m case. IZA Jou nal o Labo & De elopmen 2013 2:2. Submi you manusc ip o a jou nal and benefi om: 7 Con enien online submission 7 Rigo ous pee e iew 7 Immedia e publica ion on accep ance 7 Open access: a icles eely a ailable online 7 High isibili y wi hin he fi eld 7 Re aining he copy igh o you a icle Submi you nex manusc ip a 7 sp inge open.com Gó a IZA Jou nal o Labo & De elopmen 2013, 2:2 Page 31 o 31 h p://www.izajold.com/con en /2/1/2