Studies on the Relationship between Social Security and Personal Saving. A Tabular Survey

Kuné, Jan B. A icle S udies on he Rela ionship be ween Social Secu i y and Pe sonal Sa ing. A Tabula Su ey K edi und Kapi al P o ided in Coope a ion wi h: Duncke & Humblo , Be lin Sugges ed Ci a ion: Kuné, Jan B. (1983) : S udies on he Rela ionship be ween Social Secu i y and Pe sonal Sa ing. A Tabula Su ey, K edi und Kapi al, ISSN 0023-4591, Duncke & Humblo , Be lin, Vol. 16, Iss. 3, pp. 371-380, h ps://doi.o g/10.3790/ccm.16.3.371 This Ve sion is a ailable a : h ps://hdl.handle.ne /10419/292969 S anda d-Nu zungsbedingungen: Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen Zwecken und zum P i a geb auch gespeiche und kopie we den. Sie dü en die Dokumen e nich ü ö en liche ode komme zielle Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich machen, e eiben ode ande wei ig nu zen. So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen (insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en, gel en abweichend on diesen Nu zungsbedingungen die in de do genann en Lizenz gewäh en Nu zungs ech e. Te ms o use: Documen s in EconS o may be sa ed and copied o you pe sonal and schola ly pu poses. You a e no o copy documen s o public o comme cial pu poses, o exhibi he documen s publicly, o make hem publicly a ailable on he in e ne , o o dis ibu e o o he wise use he documen s in public. I he documen s ha e been made a ailable unde an Open Con en Licence (especially C ea i e Commons Licences), you may exe cise u he usage igh s as speci ied in he indica ed licence. h ps://c ea i ecommons.o g/licenses/by/4.0/ S udies on he Rela ionship be ween Social Secu i y and Pe sonal Sa ing A Tabula Su ey By Jan B. Kune, Ams e dam I. In oduc ion The pas decade has wi nessed an inc eased deba e on he ques ion whe he he social secu i y sys em1 has a nega i e impac on p i a e sa ing. Reduced p i a e sa ing induces a capi al sho all and inhibi s g ow h in p oduc i i y. Pa icula ly since he publica ion o Felds ein's 1974 - s udy he e ec s o social secu i y on p i a e sa ing ha e been he subjec o a s ill ongoing esea ch con o e sy. In his s udy Felds ein claims ha his empi - ical analysis o Uni ed S a es ime se ies da a o he pe iod 1929 - 1971 sugges s ha he social secu i y p og am has educed p i a e sa ing by 38 pe cen in 1971. In a clima e o anxie y o e a capi al sho age his es ima e aised se ious ques ions abou he economic consequences o he in oduc- ion and u he g ow h o a social secu i y p og am2. This pape e iews he empi ical con en o he mos impo an s udies ca ied ou o he Uni ed S a es, Canada and six Eu opean coun ies deal- ing wi h he impac o public social secu i y on agg ega e pe sonal sa ing. Theo e ical conside a ions a e le aside. II. Empi ical S udies By including social secu i y Felds ein has ex ended he li e cycle/pe ma- nen income heo y o Ando and Modigliani [1963], Modigliani and B um- be g [1954] and F iedman [1957] in an impo an way. Felds ein concludes 1 Social secu i y gene ally includes old-age, su i o s and disabili y p og ams. 2 A e iew o he capi al sho age li e a u e is ound in Eisne [1977]. See also on Fii s enbe g and Malkiel [1977] and Felds ein [19771]. Pogue and Sgon z [1977] a gue ha he U.S. social secu i y sys em has p o ided an incen i e o one gene a ion o sa e o e i emen by in es ing in he human capi al o he nex gene a ion. OPEN ACCESS | Licensed unde CC BY 4.0 | h ps://c ea i ecommons.o g/abou /cclicenses/ DOI h ps://doi.o g/10.3790/ccm.16.3.371 | Gene a ed on 2023-01-16 12:49:16 372 Jan B. Kune ha he in oduc ion and g ow h o a pay-as-you-go social secu i y sys em would dep ess pe sonal sa ing and hence na ional capi al o ma ion3. Indi- iduals iew an icipa ed social secu i y bene i s du ing e i emen as a sub- s i u e o hei own p e e i emen sa ing and hus a e less mo i a ed o sa e du ing hei wo king yea s o p o ide o e i emen consump ion. This e ec is called he sa ing eplacemen e ec . On he o he hand, howe e , sa ing is inc eased because o he "induced e i emen e ec " o he sys em: social secu i y bene i s encou age ea lie e i emen , he eby inc easing sa ing. The e ec o he social secu i y p og am on pe sonal sa ing is he e- o e heo e ically ambiguous. Ba o added o his a u he sou ce o ambigui y. He poin ed ou ha he ins i u ion o a social secu i y p og am migh eplace an exis ing sys em o in a amily (in e gene a ional) ans- e s. To he ex en ha social secu i y bene i s esul in o se ing adjus - men s in p i a e ans e s - educed ans e s om child en o hei aged pa en s o inc eased beques s om pa en s o hei child en - he e ec on p i a e sa ing would be diminished. The au ho s included in his su ey conside al e na i e de ini ions o sa ing (consump ion) and use a ious unc ional o ms o he sa ing (con- sump ion) unc ion o di e en ime pe iods encompassing a a he wide ange o explana o y a iables. Among hese a iables a e cu en and lag- ged disposable income, pe manen income, ansi o y income, e ained ea nings, household weal h, he s ock o household du ables, human capi al, o he pension und weal h, he unemploymen a e, a ious in e es a es, he labou o ce pa icipa ion a e o he aged, go e nmen su plus, eal money balances and o he s. Va iables assumed o ep esen he in luence o he social secu i y sys em a e, (1) g oss social secu i y weal h (SSWG), i.e. he p esen alue o expec ed u u e social secu i y bene i s ha can be claimed by all hose who a e ei he in he labou o ce o al eady e i ed, (2) ne social secu i y weal h (SSWN), i.e. g oss social secu i y weal h minus he p esen alue o social secu i y con ibu ions o be paid by hose who a e cu en ly in he labou 3 In his su ey only s udies dealing wi h he impac o un unded public social sec- u i y p og ams on sa ing a e conside ed. The in luence o p i a e pensions on sa ing is conside ed by Cagan [1965], Ka ona [1965], Felds ein [1978] and Munnell [1976, 1980]. Cagan and Ka ona ound ha people co e ed by p i a e pensions did no educe hei sa ing bu a he sa ed mo e han people no co e ed. Cagan explained his esul s in e ms o a ecogni ion e ec : pension plans di ec a en ion o e i emen needs. Ka ona added o his a second explana ion, he goal g adien hypo hesis: people inc ease hei sa ing e o s he close he da a o e i emen . The s udies o Felds ein and Munnell e eal ha educed p i a e sa ing is o se by inc eased pen- sion sa ing by a small amoun . OPEN ACCESS | Licensed unde CC BY 4.0 | h ps://c ea i ecommons.o g/abou /cclicenses/ DOI h ps://doi.o g/10.3790/ccm.16.3.371 | Gene a ed on 2023-01-16 12:49:16 Rela ionship be ween Social Secu i y and Pe sonal Sa ing 373 o ce, (3) social secu i y con ibu ions (SSC) and (4) social secu i y bene i s (SSB). I will be clea ha he a iables SSWG and SSWN a e jus ic ional quan i ies. I should be kep in mind ha he o m and speci ica ion o he sa ing o consump ion unc ion used by an au ho is only one o many al e na i es ha could ha e been chosen. Ca e ully sc u inizing he wo k o di e en in es iga o s e eals ha es ima ion esul s a e o en e y sensi i e o changes in speci ica ion. In o de o educe mul icollinea i y p oblems mos au ho s use se ies in cons an dolla s and on a pe capi a basis o use he a io's o ele an quan- i ies o na ional income. All au ho s, excep Gul enkin and Logue use yea ly da a, which makes i less meaning ul o ake in o accoun he possi- bili y o sho - e m disequilib ia o , in o he wo ds, he possibili y ha ac ual sa ing di e s om desi ed sa ing. All au ho s excep he ones o he On a io-s udy use a single-equa ion app oach and o dina y leas squa es. The es ima ion esul s a e he e o e subjec o possible simul aneous equa- ion bias. The On a io-s udy uses a simul aneous equa ions model ha is comp ised o wo s uc u al equa ions, he dependen a iables being he sa ing o he households and he labou - o ce pa icipa ion o he aged. The able below summa izes he mos impo an ime-se ies s udies ca - ied ou o he Uni ed S a es, Canada and six Eu opean coun ies4. A en- ion is only gi en o he sign and signi icance o he coe icien o he social secu i y a iable. Un o una ely i was no possible o calcula e elas ici ies. F om each au ho we choose hose equa ions which a e in ou iew mos cha ac e is ic o on which he au ho himsel places mos in e es . 4 S udies on he ela ionship be ween social secu i y and p i a e sa ing ac oss coun ies ha e been ca ied ou e.g. by Felds ein [1977n, 1980], Ba o and McDonald [1979] and Kopi s and Go u [1980]. Felds ein concludes ha his indings con i m he hypo hesis o a nega i e in luence o public social secu i y on pe sonal sa ing, whe eas he esul s o Ba o and McDonald and Kopi s and Go u do no suppo ha hypo hesis. The di e gence o pa ame e es ima es o Felds ein and Ba o and McDonald is discussed by Ho ioka [1980]. C oss-sec ion analyses o indi idual house- holds a a poin in ime o he U.S. ha e been ca ied ou e.g. by Munnell [1976], Felds ein [1976], Felds ein and Pellechio [1979] and Ko liko [19791]. The indings o Munnell, Felds ein and Felds ein and Pellechio sugges ha he e is a nega i e impac o public social secu i y on sa ing, whe eas he e idence o Ko liko indica es incon- clusi eness. A use ul e iew o c oss-sec ion analyses o he Uni ed S a es is gi en by Danzinge e al. [1981]. The main conclusion o G een [1981] o he Uni ed Kingdom is ha pension sa ing does no subs i u e wi h o he ypes o sa ing. Agains his Schoeplein [1970] ound in Canada some e idence o subs i u abili y be ween pen- sions and o he e i emen sa ing. OPEN ACCESS | Licensed unde CC BY 4.0 | h ps://c ea i ecommons.o g/abou /cclicenses/ DOI h ps://doi.o g/10.3790/ccm.16.3.371 | Gene a ed on 2023-01-16 12:49:16 374 Jan B. Kune 1. Uni ed S a es Felds ein [1974] uses agg ega e ime-se ies da a o he pe iod 1929 - 1971. His es ima es a e based on a li e-cycle model including g oss and ne social secu i y weal h in u n5. The eg ession coe icien ound on SSWG o he en i e pe iod has he igh sign and is highly signi ican . Howe e , his coe icien is no signi ican when he equa ion is es ima ed o he pos wa pe iod 1947 - 1971. Mo e impo an , he esul ound o he en i e pe iod depends c i ically upon dele ing he unemploymen e m om he equa ion. This howe e does no appea jus i ied. Ko liko [197911] no es ha he e ec o he implied dec ease in sa ing on he a e o e u n on cap- i al should be aken in o accoun . Munnell [1974] uses agg ega e ime-se ies da a o he pe iod 1900 - 1971. He es ima es a e also based on he li e-cycle model. Va iables measu ing he e ec o social secu i y a e g oss and ne social secu i y weal h and social secu i y con ibu ions. Fu he mo e a a iable ep esen ing he induced e i emen e ec is included, which howe e appea s ne e o be signi ican . The social secu i y a iables a e no signi ican in explaining pe sonal sa ing du ing he whole pe iod; he a iable SSWN is signi ican in one case o he pos wa pe iod 1949 - 1969. Al hough agg ega e pe sonal sa ing is no a ec ed, social secu i y seems o ha e an impac on he ype o sa ing. Ba o [1978] uses agg ega e ime-se ies da a o he pe iod 1929 - 1974. His unc ion is also inspi ed by he li e-cycle model. A new a iable used o measu e he e ec o social secu i y is he bene i -co e age, which appea s ne e o be signi ican . The unemploymen a iable is always signi ican , whe eas he SSWG a iable is no signi ican in all pe iods analyzed excep when he unemploymen e m is dele ed. Da by [1979] uses ime-se ies da a o he pe iod 1929 - 1974 and he unc ions he es ima es a e based on he pe manen income hypo hesis. The SSWG and SSWN a iables seem o be signi ican in explaining pe sonal sa ing du ing he whole pe iod, bu Da by himsel is doub ul on his esul because o speci ica ion e o s. Reg essions un o he pos wa pe iod 1947 - 1974 ne e show a posi i e o nega i e e ec o social secu i y on sa ing. Gul enkin and Logue [1979] i s use yea ly ime-se ies da a o he pos - wa pe iod. In one o wo cases he SSC a iable appea s o be signi ican . 5 Though Felds ein's heo e ical o mula ion clea ly ecognizes he induced e i e- men e ec o social secu i y, he e is no a iable explici ly designed o accoun o i in his speci ica ion. OPEN ACCESS | Licensed unde CC BY 4.0 | h ps://c ea i ecommons.o g/abou /cclicenses/ DOI h ps://doi.o g/10.3790/ccm.16.3.371 | Gene a ed on 2023-01-16 12:49:16 Rela ionship be ween Social Secu i y and Pe sonal Sa ing 375 The coe icien o SSWN is ne e signi ican . Nex he au ho s es ima e some sa ing equa ions based on a s ock adjus men model using qua e ly da a o he pe iod 1952: III-1974: II. The long- un e ec s o social secu i y appea o be nega i e o he en i e pe iod and wo chosen subpe iods. Un o una ely s anda d e o s o he long- e m coe icien s a e no gi en. Leime and Lesnoy [1982] disco e ed an e o in Felds ein's cons uc ion o he SSWG a iable. The esul s o he au ho s based on he Felds ein eplica a iables do no p o ide s a is ically signi ican suppo o he hypo hesis ha social secu i y dec eased pe sonal sa ing. Leime and Les- noy also de eloped a ious al e na i e SSWG a iables. The coe icien s a e gene ally insigni ican and (in he au ho s' iew) o he w ong (posi i e) sign. 2. Canada Boyle and Mu ay [1974] use annual ime-se ies da a o he pe iod 1954 - 1975. Also hei sa ing unc ion is inspi ed by he li e-cycle model. The SSWG a iable appea s no o be signi ican in any o he equa ions es i- ma ed. In a s udy o he Go e nmen o On a io [1977] a sa ing unc ion and a e i emen unc ion a e es ima ed simul aneously o he pe iod 1953 - 1975. The social secu i y a iable used he e is eal go e nmen public pen- sion bene i s pe aged eligible indi idual. The indings a e ha he e is no signi ican in luence o his a iable on pe sonal sa ing. 3. England Hemming [1978] ex ends Felds ein's s udy o England o he pe iod 1949 - 1973. One o he explana o y a iables is expec ed income, which is app oxima ed by an in ini e geome ically declining lag on pas incomes. The coe icien o SSWG is ound o be insigni ican , whe eas ha o he SSWN a iable is signi ican and has a posi i e sign. 4. Sweden The model o Ma kowski and Palme [1979] employs o he pe iod 1952 - 1975 he amewo k o he pe manen income hypo hesis o iginally o mu- la ed by F iedman. Expec ed pension bene i s a e used as he social secu i y a iable. This a iable becomes signi ican in he equa ion on which he au ho s hemsel es place mos in e es . I appea s howe e ha educed OPEN ACCESS | Licensed unde CC BY 4.0 | h ps://c ea i ecommons.o g/abou /cclicenses/ DOI h ps://doi.o g/10.3790/ccm.16.3.371 | Gene a ed on 2023-01-16 12:49:16 376 Jan B. Kune pe sonal sa ing is mo e han o se by inc eased pension sa ing, since he Swedish pension scheme is pa ially unded. 5. Ge many (FR) P a , Hu le and Denne lein [1979] conside he pos wa pe iods 1958 (1964) - 1975. Social secu i y bene i s and con ibu ions a e used as a i- ables e lec ing he in luence o he social secu i y sys em. Thei main ind- ing is ha he empi ical e idence indica es ha ei he no e ec exis s a all o else ha i would end o be negligible. 6. F ance Kessle , Masson and S auss-Kahn [1980] ex end he li e-cycle model s udies o Felds ein and Munnell o F ance o he pe iod 1959 - 1977. Wi h espec o he SSWG a iable he indings a e inconclusi e. The a iable SSC also does no seem o exe a signi ican in luence. 7. Belgium Pes ieau and Pe elman [1980] pe o m he same calcula ions as he abo e men ioned au ho s o Belgium o he pe iod 1962- 1977. Thei indings a e ha he e is nei he a nega i e no a posi i e in luence o social secu i y on na ional sa ing. 8. Ne he lands Kune [1981] in es iga es he impac o social secu i y on pe sonal sa ing o he pe iod 1952 - 1978. I appea s ha no unambiguous empi ical sup- po can be ound o he hypo hesis o a nega i e e ec o social secu i y on pe sonal sa ing. m. Conclusions In his pape he empi ical esul s o he majo s udies ha in es iga e he e ec o social secu i y on agg ega e pe sonal sa ing in he Uni ed S a es, Canada and six Eu opean coun ies a e e iewed. The su ey is summa ized in abula o m. OPEN ACCESS | Licensed unde CC BY 4.0 | h ps://c ea i ecommons.o g/abou /cclicenses/ DOI h ps://doi.o g/10.3790/ccm.16.3.371 | Gene a ed on 2023-01-16 12:49:16 Rela ionship be ween Social Secu i y and Pe sonal Sa ing 377 The E ec o Social Secu i y on Pe sonal Sa ing coun y au ho (s) pe ioda) social secu i y a iable e idence b) Uni ed Felds ein 1929- 1971 SSWG signi ican S a es 1947 - 1971 SSWG no signi ican Munnell 1900- 1971 SSC no signi ican 1929 - 1969 SSWG no signi ican 1929 - 1969 SSWN no signi ican 1946 - 1971 SSC no signi ican 1946- 1969 SSWG no signi ican 1946- 1969 SSWN signi ican Ba o 1929 - 1974 SSWG no signi ican 1947 - 1974 SSWG no signi ican Da by 1929 - 1974 SSC no signi ican 1929 - 1974 SSWG signi ican 1929 - 1974 SSWN signi ican 1947 - 1974 SSC no signi ican 1947 - 1974 SSWG no signi ican 1947 - 1974 SSWN no signi ican Gul enkin 1947 - 1974 SSWN no signi ican and Logue 1947 - 1974 SSC signi ican 1952:111- 1974:11 SSWN unknown and wo SSC unknown subpe iods Leime and Lesnoy 1930 - 1974 SSWG no signi ican 1930 - 1974 SSWN no signi ican 1947 - 1974 SSWG signi ican , posi i e sign, implausible size 1947 - 1974 SSWN implausible size 1930 - 1974 SSWG no signi ican , a ian s posi i e sign 1947 - 1974 SSWG a ian s posi i e sign Canada Boyle and Mu ay 1954- 1975 SSWG no signi ican Go . o On a io 1953 - 1975 SSB no signi ican England Hemming 1949 - 1973 SSWG no signi ican 1949 - 1973 SSWN signi ican , posi i e sign Sweden Ma kowski 1952 - 1975 SSB signi ican and Palme signi ican Ge many P a e al. 1958(1964)-1975 SSC signi ican , (FR) posi i e sign 1958(1964)-1975 SSB no signi ican F ance Kessle e al. 1959 - 1977 SSWG no signi ican 1959 - 1977 SSWG.! signi ican 1959- 1977 SSC no signi ican Belgium Pes ieau 1962 - 1977 SSWG no signi ican and Pe elman Ne he lands Kune 1952 - 1978 SSC no signi ican a) Excluding he wa yea s 1941 - 1946. b) Unless o he wise no ed he e is (no ) a signi ican e idence o a nega i e impac o social secu i y on pe sonal sa ing. 25 K edi und Kapi al 3/1983 OPEN ACCESS | Licensed unde CC BY 4.0 | h ps://c ea i ecommons.o g/abou /cclicenses/ DOI h ps://doi.o g/10.3790/ccm.16.3.371 | Gene a ed on 2023-01-16 12:49:16 378 Jan B. Kune The o e all conclusion om he s udies su eyed mus be a nega i e one: he e idence ound does no p o ide empi ical suppo o he hypo hesis ha social secu i y dep esses p i a e sa ing. Mo e impo an , he imp es- sion comes h ough ha ei he a posi i e o a nega i e social secu i y e ec on sa ing can be p oduced depending on he de ails o speci ica ion and he leng h o he obse a ional pe iod. The s udies which gi e suppo o he hypo hesis o a nega i e impac o social secu i y lose much o hei alidi y a e ca e ul in e p e a ion6. The e a e many p oblems o economic concep ion, s a is ical heo y and da a cha ac e is ics ha se e ely limi he eliabili y o he es ima es o he e ec o social secu i y on sa ing. Fu he mo e, we a e inclined o conclude ha he mo i es o sa ing a e much mo e complex han hose pos ula ed wi hin he con ex o he li e cycle model. The inconclusi e esul s ound by many au ho s may hus be consis en wi h he iew ha he li e cycle model does no p o ide an adequa e explana ion o he indi idual's sa ing beha io . Re e ences Ando, A. and F. Modigliani: The 'Li e Cycle' Hypo hesis o Sa ing: Agg ega e Implica ions and Tes s, Ame ican Economic Re iew, 53, 1963, pp. 53 - 84. - Ba o, R. J.: The Impac o Social Secu i y on P i a e Sa ing, Ame ican En e p ise Ins i u e o Public Policy Resea ch, Washing on, D.C., 1978. - Ba o, R. J. and G. M. McDonald: Social Secu i y and Consume Spending in an In e na ional C oss Sec ion, Jou nal o Public Economics, 11, 1979, pp. 275 - 89. - Boyle, P. and J. Mu ay: Social Secu i y Weal h and P i a e Sa ing in Canada, The Canadian Jou nal o Economics, 12, 1979, pp. 456 - 68. - Cagan, Ph.: The E ec o Pension Plans on Agg ega e Sa ings, Na ional Bu eau o Economic Resea ch, New Yo k, 1965. - Danzige , S., Ha eman, R. and R. Plo nick: How Income T ans e P og ams A ec Wo k, Sa ings and he Income Dis ibu ion: A C i ical Re iew, Jou nal o Economic Li e a u e, 19, 1981, pp. 975 - 1028. - Da by, M. R.: The E ec s o Social Secu i y on Income and he Capi al S ock, Ame ican En e p ise Ins i u e o Public Policy Resea ch, Washing on, D.C., 1979. - Eisne , R.: Capi al Sho age: My h and Reali y, Ame ican Economic Re iew, 67, Pape s and P oceedings, 1977, pp. 110 - 15. - Esposi o, L.: E ec o Social Sec- u i y on Sa ing: Re iew o S udies Using U.S. Time-Se ies Da a, Social Secu i y Bul- le in, May 1978, pp. 9 - 17. - Felds ein, M. S.: Social Secu i y, Induced Re i emen and Agg ega e Capi al Accumula ion, Jou nal o Poli ical Economy, 82,1974, pp. 905 - 26. - Felds ein, M. S.: Social Secu i y and he Dis ibu ion o Weal h, Jou nal o he Ame ican S a is ical Associa ion, 71, dec. 1976, pp. 800 - 07. - Felds ein, M. S.: Na ional Sa ing in he Uni ed S a es, In: Capi al o P oduc i i y and Jobs, eds. E. Shapi o and W. L. Whi e, P en ice Hall, Englewood Cli s, N.J., 19771. - Felds ein, M. 6 This conclusion eme ges also om heo e ical conside a ions and om wo o he ypes o empi ical e idence ha a e p esen ly a ailable: c oss coun y s udies and analyses o indi idual c oss-sec ions. OPEN ACCESS | Licensed unde CC BY 4.0 | h ps://c ea i ecommons.o g/abou /cclicenses/ DOI h ps://doi.o g/10.3790/ccm.16.3.371 | Gene a ed on 2023-01-16 12:49:16