"The Existence of a World Demand for Money Function". A Reply

F owen, S. F.; Kou is, G. A icle "The Exis ence o a Wo ld Demand o Money Func ion". A Reply K edi und Kapi al P o ided in Coope a ion wi h: Duncke & Humblo , Be lin Sugges ed Ci a ion: F owen, S. F.; Kou is, G. (1979) : "The Exis ence o a Wo ld Demand o Money Func ion". A Reply, K edi und Kapi al, ISSN 0023-4591, Duncke & Humblo , Be lin, Vol. 12, Iss. 1, pp. 73-82, h ps://doi.o g/10.3790/ccm.12.1.73 This Ve sion is a ailable a : h ps://hdl.handle.ne /10419/292836 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/ The Exis ence o a Wo ld Demand o Money Func ion A Reply By S. F. F owen, Guild o d and G. Kou is, London In his commen s Zis (1978) main ains, in sho , ha he e is nei he heo e ical no empi ical ounda ion in he claim ha ou indings o a signi ican in e es elas ici y and ins abili y o he wo ld demand o money suppo Keynesian p oposi ions. This would imply ha he Pa kin (1976) conclusions ha wo ld in la ion can be con olled by con- olling he wo ld money supply is co ec . In his commen s Zis e oneously assumes ha in ou iew demand o money unc ions a e iden ical in all coun ies. In ac , we en i ely ag ee wi h him ha he indi idual coun y unc ions migh well di e . As explained in he F owen and Kou is pape (1977, page 3), ou esul s e e o he sample o he en pooled coun ies and ine i ably ou es ima ed elas ici ies would ep esen an a e age beha iou iden ical o he wo ld beha iou e lec ed in he esul s ob ained by G ay, Wa d and Zis (1976) when g ouping he en coun ies oge he and es ima ing a unc ion on he agg ega e a iables. In bo h cases he hypo hesis es ed is he same. Zis is no doub awa e ha adi ionally in mac o economics many issues such as consump ion o p oduc ion beha iou ha e been app oached bo h by using ime se ies and c oss-sec ion analysis. Ou me hodology is simply a combina ion o he wo ap- p oaches which has he dis inc ad an age i s ly o being able o ope a e wi h a sample la ge enough o ake accoun o di e ences be ween each c oss-sec ion uni and secondly o i ually elimina ing he dis ibu ed lag bias. Ou s a ing poin was ha because o di e ences in he demand- o -money beha iou ac oss egions, a simple agg ega ion o he ele an da a o coun ies o be conside ed in a s udy a wo ld le el o ally neglec s hese di e ences. The e is no need o emind Zis ha F iedman (1963), in his s udy on money holdings exp essed in e ms 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.12.1.73 | Gene a ed on 2023-01-16 12:44:02 74 S. F. F owen and G. Kou is o weekly incomes ac oss coun ies, such as he U. S., Is ael, India, Yugosla ia e c., showed ha he a ios a e somewha di e en be ween coun ies. I is p ecisely hese kinds o he e ogenei ies ha we wished o sc een ou in an a emp o es ima e a wo ld demand o money. Al hough unde a ixed exchange a e egime he money supplies o indi idual coun ies could legi ima ely be iewed as one en i y, i is qui e logical o assume ha spending pa e ns (and hus he desi ed demand o hold money balances) di e ges be ween coun ies. Thus hese di e ences when no allowed o in a wo ld agg ega e would impai he alidi y o he exe cise. Fo his eason we allowed o such di e ences and upon s a is ical es ing ound hem signi ican . As o he c i icism ha he in e p e a ion o ou esul s lacks heo- e ical ounda ion in so a as he claim suppo ing he Keynesian case is conce ned, he poin Zis makes does no seem o be a all p o ound. When one se s ou o es ima e a wo ld demand o money unc ion unde he assump ion ha du ing a ixed exchange a e pe iod he en i e wo ld can be iewed as one economy ... e c., i ollows by de ini ion ha bo h he mone a is and he Keynesian heo ies can be applied as i we we e dealing wi h an indi idual coun y. Fo example, in he same way as we ega d England, Wales and Sco land as in eg al pa s o one closed economy, he Uni ed Kingdom, we can also ega d he indi idual coun ies included in ou s udy as in eg al pa s o a ixed exchange a e wo ld, a wo ld which o all p ac ical pu poses is jus ano he closed economy. We a e g a e ul o Zis o aising he ques ion o he con e sion o dolla s o he na ional money supply and na ional income se ies. This p o ides us wi h he oppo uni y o commen ing on wha appea s o us as an inconsis ency in he G ay, Wa d and Zis pape , an incon- sis ency we neglec ed o poin ou ea lie . When agg ega ion o e a numbe o coun ies wi h dissimila cu encies akes place, bo h he money s ock and income mus be con e ed o a common uni o measu emen , say he U. S. dolla . The e a e basically h ee al e na i es o his app oach. Fi s ly, o con e bo h magni udes by using cu en exchange a es, i. e. he exchange a e p e ailing a each qua e (op ion chosen by G. W. Z.). Secondly, o use cons an exchange a es o a gi en yea h oughou . Thi dly, o con e he money supply by ussing cu en exchange a es and income by using cons an exchange a es (al e na i e chosen by us). A ou h op ion, o con e money by using cons an exchange a es and income by cu en exchange a es, is 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.12.1.73 | Gene a ed on 2023-01-16 12:44:02 The Exis ence o a Wo ld Demand o Money Func ion 75 logically p ecluded as will be made e iden in he a gumen s ha ollow. The e a e s ong a gumen s o using cons an exchange a es in con e ing eal na ional GNPs o a common denomina o . This is because we wan he a ia ion o GNP o e ime o e lec changes in li ing s anda ds and hus explain he ansac ions demand o money. When GNP is con e ed a cu en exchange a es we a i icially impa an ex a a ia ion on he se ies which has no hing o do wi h ac ual changes in income. Fo example, in he 1967/68 pe iod eal GNP in he U. K. ose by abou 2.3 °/o while a he same ime he ex e nal alue o he pound agains he dolla declined by 14 <Vo. This means ha i he U. K. GNP is con e ed in o U. S. dolla s i would depic a sha p all o 11.7 Vo om 1967 o 1968! This would clea ly be absu d and such an anomaly can only be a oided i GNP is con e ed a cons an ex- change a es. In eali y wha we ac ually need is pu chasing powe pa i ies so ha a p ope weigh ing could be gi en o na ional GNPs be o e agg ega ion akes place. Pu chasing powe pa i ies o he yea s and he numbe o coun ies co e ed in ou s udy un o una ely do no exis . Thus he second bes me hod is o use as weigh s he exchange a es p e ailing a a gi en poin in ime which o con e sion pu poses would be kep cons an du ing he whole pe iod. A leas in his way he ue a ia ion o GNPs is no dis o ed. Likewise one may a gue ha na ional money s ocks should also be con e ed a cons an exchange a es so ha changes in he money supply would e lec only he ise in money ha ac ually ci cula es wi hin coun ies. I ins ead we exp ess na ional money supplies in dolla e ms by using cu en exchange a es, he money supplies ac oss coun ies would appea a imes o ise o slow down in esponse o exchange a e changes a he han o changes in he eal g ow h a e o he money supply ha akes place wi hin he domes ic economy. Thus whene e he e has been an app ecia ion o he DM, he money supply in Ge many, iewed in e ms o dolla s, would seem o ha e inc eased e en i a he gi en yea o app ecia ion he ac ual s ock o money had no isen by one ex a DM. Hence ollowing he G. W. Z. heo y ha unde ixed exchange a es he en i e wo ld can be iewed as one economy whe e e e y coun y cons i u es a me e egion in i , he e a e powe ul a gumen s o wishing o explain in his wo ld he ue mo emen s o he money supply, incomes and in la ion a es and no he dis o ed ones ha a ise 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.12.1.73 | Gene a ed on 2023-01-16 12:44:02 76 S. F. F owen and G. Kou is when we ake in o accoun luc ua ions in he ex e nal alues o domes ic cu encies. In hese ci cums ances i is p ope o use cons an exchange a es o con e ing bo h incomes and money s ocks. Ne e heless, G. W. Z. chose o con e GNPs and money s ocks using cu en exchange a es. Thus he a iables used o hei empi ical wo k a e subjec o a dis o ion. Fu he mo e, he e is an e en mo e se ious a gumen agains he use o cu en exchange a es. Take, o example, a hypo he ical case whe e du ing a pe iod o h ee yea s bo h eal income and he money supply wi hin a coun y ha e emained cons an while ex e nally he na ional cu ency has dep ecia ed agains he dolla . The me e con e sion o income and money in o dolla s a cu en exchange a es will c ea e an a i icial posi i e co ela ion be ween he wo a iables while in eali y he e was none. This p ocedu e would lead o a wo ld demand o money which is mo e s able han would o he wise be he case. This is due, o cou se, o he con e - sion in o dolla s which qui e alsely made he co ela ion be ween a ia- ions in he money supply and a ia ions in income s onge han i eally is. We would no be su p ised i a imes he use o cu en exchange a es p oduced a s able wo ld demand o money e en unde loa ing exchange a es. In ac , by ollowing his me hod he mo e equen he luc ua ions in exchange a es he s onge he associa ion be ween money and income and hus he be e de ined he demand o money would appea on es ima ion. The easoning ou lined so a poin s o he ac ha o wo ld ag- g ega es he bes p ocedu e is o use cons an exchange a es. We ha e ollowed a middle cou se, howe e , in ha we used cons an exchange a es o con e ing GNPs bu cu en exchange a es o con e ing money s ocks. The jus i ica ion o doing so is as ollows: he e a e h ee possible ypes o exchange a e egimes, namely igidly ixed exchange a es, eely loa ing exchange a es and an adjus able peg sys em wi h pe iodic changes in pa i ies. In he i s case, es ima ing a wo ld demand o money is a legi ima e exe cise. In he second case he e is no ixed link be ween na ional cu encies and hus he exe cise is a oid one. In he hi d case, howe e , which is he one ha cha ac e ised he pe iod unde in es iga ion, a wo ld demand o money may be iden i ied bu i mus be ecognised ha e e y ime an exchange a e change has aken place, he alidi y o he whole exe cise is impai ed. I one is o be hones in empi ical es ima ion, some allowance o he "shocks" ha he wo ld demand o money has 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.12.1.73 | Gene a ed on 2023-01-16 12:44:02 The Exis ence o a Wo ld Demand o Money Func ion 77 sus ained, mus be made. To ou mind, one way o allowing o hese shocks is o con e he money supply a he cu en exchange a e so ha whene e he e has been a ealignmen o exchange a es, he dis o ion is aken ca e o in he wo ld demand o money unc ion. This me hod appea s o be p e e able o dis ega ding he ac ha some luc ua ion in exchange a es has occu ed du ing he pe iod unde e iew and o using cons an exchange a es as i his pe iod had been one o igid exchange a es. Ins ead, we main ain ha he use o cu en exchange a es o con e ing he money supply and cons an exchange a es o con- e ing income is p obably he bes p ocedu e o ollow. In his way e en when he e a e luc ua ions in he money s ock caused by ex- change a e changes, which o cou se a e no accoun ed o in he ex- plana o y side o he equa ion, his will be p ima ily abso bed by he e o e m. Acco ding o he G. W. Z. app oach, howe e , his ex a sou ce o a iabili y will be abso bed di ec ly by he income a iable hus p oducing an income elas ici y which would be biased upwa d. We now come o wha appea s o be he c ux o he ma e , namely he s a is ical and economic di e ences be ween he G. W. Z. s udy and ou s. To s a wi h we es a e wha G. W. Z. ega d as hei bes model, which in ac is he speci ica ion we ha e chosen ou sel es. Using he agg ega e da a o he en coun ies o he pe iod 1957 (2) - 1971 (2) we es ima e he unc ion: M = a + bY + c + dM _ whe e all a iables a e in na u al loga i hms and M -i is he esul o a pa ial adjus men hypo hesis on he dependen a iable. The in e es a e used is he Eu odolla a e. The esul s ob ained a e as ollows: M = - 1.1887 + 0.1794 Y - 0.02269 + 0.7911 M _1 (4.57) (7.49) (6.68) (17.24) R2 = 0.997 DW = 1.30 These esul s came as a su p ise o us since hey a e no iden ical o hose quo ed by G. W. Z. o he same ime pe iod and iden ical speci ica ion*. In ac , he esul s quo ed by G. W. Z. we e as ollows: * The basic di e ence is ha he GNPs se ies ha we used we e ac ual qua e ly igu es while he se ies used by G. W. Z. we e ob ained by in e - pola ion om annual igu es. 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.12.1.73 | Gene a ed on 2023-01-16 12:44:02 78 S. F. F owen and G. Kou is M = 0.663 + 0.234 Y - 0.0125 R + 0.610 M _ R2 = 0.99 Ou use o he G. W. Z. da a e eal a much highe lagged money s ock coe izien which implies a slowe speed o adjus men . Con- cen a ing on he empi ical esul s, as one would expec , he lagged money ss ock a iable is explaining oo much, hus educing he income e ec o explaining e y li le, he classical case o he dis ibu ed lag bias. In mos cases whe e lagged endogenous a iables a e used as eg esso s hey end o domina e he explana o y side o he equa ion, so ha he es o he coe icien s a e biased downwa ds. In his sense he dis ibu ed lag bias appea s o be simila o he ype o bias encoun- e ed in he p esence o mul i-collinea i y which is de imen al o he co ec es ima ion o any o he coe icien s. As is well known his bias anishes only as he sample inc eases (i. e. asymp o ically) some hing we claim o ha e accomplished in ou s udy. A measu e o he bias could be ob ained om compa ing he simple co ela ion coe icien be ween Y and M (which is 0.985) and he o e all mul iple co ela ion coe icien (which is 0.998). Appa en ly hese a e qui e close. One can p oceed by using i s di e ences o logs ins ead o le els so ha he associa ion be ween Y and M (and hence he bias) becomes weake . The esul s a e as ollows: span 1957 (3) - 1971 (2) DM = 0.3083DY - 0.0225D + 0.5805DM _ (4.69) (3.69) (5.81) = 0.554 DW = 1.94 whe e he ope a o D signi ies i s di e ences. The esul s look plau- sible wi h he lagged money s ock becoming less signi ican and he pa - ial adjus men coe icien aking a asonable alue o 0.42. The simple co ela ion coe icien be ween DY and DM -1 is 0.35 while he mul- iple co ela ion coe icien is 0.75. The size o he o me is less han hal he size o he la e , e ealing ha he dis ibu ed lag bias has been conside ably educed. We can now es , by spli ing he pe iod, whe he he wo ld demand o money is s able o no . We will ollow he G. W. Z. p ocedu e o es ima ing he same unc ion o he i s o y and he las o y obse a ions. We a e awa e, o cou se, ha by so doing he o e lap be ween he wo sub-samples is qui e con- side able ( wen y- h ee obse a ions), and hence he e is an in-buil 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.12.1.73 | Gene a ed on 2023-01-16 12:44:02 The Exis ence o a Wo ld Demand o Money Func ion 79 s abili y al eady in he unc ion. By again using i s di e ences we ob ained he ollowing esul s: span = 1957 (3) - 1967 (2) DM = 0.3352 DY - 0.0275 D + 0.5608 DM _ (4.81) (3.56) (4.95) R2 = 0.588 DW = 1.70 simple co ela ion be ween DY and DM . i = 0.396, R = 0.767. span = 1961 (3) - 1971 (2) DM = 0.2950 DY - 0.0177 D + 0.5283 DM _ (3.36) (2.61) (4.24) R2 = 0.411 DW = 1.83 simple co ela ion be ween DY and DM -1 = 0.151, R = 0.641. Long Run Elas ici ies SPAN Y R Adjus men pa ame e 1957 (3) - 1971 (2) 0.735 - 0.054 0.420 1957 (3) - 1967 (2) 0.763 - 0.063 0.439 1961 (3) - 1971 (2) 0.625 - 0.038 0.472 The in e es elas ici y is e y signi ican in bo h pe iods in con as o he G. W. Z. esul s who ound i insigni ican . In ba h pe iods he unc ion is well de ined (especially o an equa ion exp essed in di - e ences) al hough in he second pe iod he explana o y powe de e- io a es. Ou simple indica o o he dis ibu ed lag bias shows ha i has been o a conside able ex en weakened by he use o i s di - e ences. In conclusion, he agg ega e da a e eal ha an in e es a e e ec is e y signi ican in he wo ld demand o money. The unc ion seems o be qui e uns able i i ed in di e en ime pe iods as he in e es elas ici y is almos hal in alue, while he income elas ici y is educed by abou 20%. Adjus men s owa ds he desi ed money holding, howe e , end o be mo e s able o e ime. In gene al, hese esul s ag ee qui e closely wi h ou own indings, whe e he pooling me hod was used, and we see no g ounds o any c i icism o ou s udy by Zis. 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.12.1.73 | Gene a ed on 2023-01-16 12:44:02 80 S. F. F owen and G. Kou is In ac , qui e he opposi e: ou empi ical in es iga ion demons a es ha i some allowance o he dis ibu ed lag bias is made and be e quali y da a a e used, he G. W. Z. esul s end o be e e sed com- ple ely. Re e ences F iedman, M.: In la ion: Causes and Consequences (New Yo k: Asia Pub- lishing House, 1963). — F owen, S.F. and Kou is, G.: "The Exis ence o a Wo ld Demand o Money Func ion: P elimina y Resul s". K edi und Kapi al, ol. 10, No. 1, 1977, pp. 1 -17. — G ay, M.R., Wa d, R. and Zis, G.: "The Wo ld Demand o Money Func ion: Some P elimina y Resul s", in Pa kin, M. and Zis, G. (eds.) In la ion in he Wo ld Economy (Manches e Uni e si y P ess, 1976) — Pa kin, M.: "In la ion in he Wo ld Economy: 1958- 1975, Su ey Pape s in Economics, No. 11, No embe 1976. — Zis, G.: "A No e on 'The Exis ence o a Wo ld Demand o Money Func ion: P eli- mina y Resul s'", K edi und Kapi al, ol.11, No. 3, 1978, pp. 340 - 347. Zusammen assung Die Exis enz eine in e na ionalen Geldnach age unk ion Eine Replik Mi diesem Bei ag wollen wi de Behaup ung on Zis en gegen e en, daß unse e A bei übe die Exis enz eine in e na ionalen Geldnach age- unk ion sowohl de heo e ischen als auch de empi ischen Gül igkei en - beh . Die Behaup ungen on Zis können sowohl du ch die en wickel en heo e ischen Behaup ungen als auch du ch die du chge üh en ökonome- ischen Tes s öllig wide leg we den. Wi hal en die Behaup ung au - ech , daß die zu es ende Hypo hese gül ig is , ob man nun das Ve hal en de in e na ionalen Geldnach age unk ion es e , indem man Lände g up- pen zusammen aß , ode diese Lände in eine zusammenge aß en Zei ei- hen-K euzs ichp obe behandel . Da übe hinaus is unse e Technik de Agg ega ion ökonome isch höhe we ig, weil sie das P oblem de Mul i- kollinea i ä und die Ve ze ungen de Lag-Ve eilung e minde . Wenn man a sächlich den Ve such un e nimm , einen Teil de Ve ze ungen de Lag-Ve eilung aus den Da en on G.W. Zis zu eliminie en, dann ände n sich ih e Resul a e olls ändig und s immen in de Tendenz mi unse en übe eiin. Was die F age anbelang , welches de geeigne s e Weg sei, na ionale Geld- a iablen au einen gemeinsamen Nenne zu b ingen, so haben wi da übe hin eichend disku ie und kamen zu dem Schluß, daß de Ansa z on G. W. Zis die lau enden Wechselku se zu benu zen, eine posi i e Ve ze ung au 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.12.1.73 | Gene a ed on 2023-01-16 12:44:02