Monetary Explanations of the Weimar Republic's Hyperinflation: Some Neglected Contributions in Contemporary German Literature Author(s): David E.

Laidler and George W. Stadler Source: Journal of Money, Credit and Banking, Vol. 30, No. 4 (Nov., 1998), pp. 816-831 Published by: Blackwell Publishing Stable URL: http://www.jstor.org/stable/2601130 Accessed: 18/10/2010 18:20
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=black. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Blackwell Publishing and Ohio State University Press are collaborating with JSTOR to digitize, preserve and extend access to Journal of Money, Credit and Banking.

http://www.jstor.org

DAVID E. W. LAIDLER GEORGE W. STADLER

Monetary Explanations theWeimar of Republic's Hyperinflation: SomeNeglectedContributions in Contemporary German Literature

Contemporary analysesof the Weimarhyperinflation, Hahn,Bortkiewicz,andMises, by have been inadequatelyappreciatedby earliercommentators.They used the quantity theory,supplementedby analysis of inflationexpectations,to explain hyperinflation's stylized facts. The lattertwo treatedexpectationsas forwardlooking, andraisedthe fiscal situation,in the spiritof Sargent' lateranalysis.They also arguedthatthe effects of s expectationson price-settingbehaviorcould createa shortfallof money currently cirin culationfrom the demandfor it, thus sketchinga disequilibrium analysisof hyperinflation thathas no exact parallelin moderntreatments the topic. of

ITISGENERALLY BELIEVED the role of expectationsin the that monetarydynamicsof hyperinflation not well understoodbefore the appearance was of Bresciani-Turroni (1931, tr. 1937), with Phillip Cagan (1956), and later Thomas Sargent(1982), makingfurthersubstantialadvanceson the basis of his insights. The WeimarRepublic' hyperinflation 1921- 1923 figuresprominently this work,but s of in it this paperwe arguethatsome contemporary German(or at least German-speaking) commentators this experience,notablyLadislaus(Ludwig)von Bortkiewicz,Ludon wig von Mises, and L. AlbertHahndisplayeda sophisticatedgraspof the role of expectationsas well as fiscal and political factorsin the inflationary process.l We also
Workon this paperbegan when Laidlerwas the holderof a SeniorVisiting Fellowship at the University of Newcastle-upon-Tyne.Laidler also acknowledges the financial supportof the Lynde and Harry BradleyFoundation,andthe SSHRCC.All translations frompreviouslyuntranslated poorlytranslated) (or German-language sources aredue to Stadler.This paperhas been presentedat a workshopat the University of WesternOntario,at the Universityof Toronto-York UniversityWorkshopon the Historyof Economic Thought,and at a conference held in honor of BernardCorry at Queen Mary and Westfield College London. The comments received on those occasions and the advice of Haim Barkai, Peter Bernholtz PhillipCagan,JuneFlanders,Malte Krtiger,Denis O'Brien, ChristofRuhl, and an anonymousreferee are gratefullyacknowledged.LorettaNott providedcapableresearchassistance. 1. Mises, nowadays still well known, was at this time a Privatdozentat the University of Vienna and economist for the Vienna Chamberof Commerce.In 1926 he foundedthe AustrianInstitutefor Business Cycle Research,with which his formerstudent,Friedrichvon Hayek, was also associatedas a junior staff member.Bortkiewicz,of Polish origin,buteducatedat the Universityof St. Petersburg Mathematicsand in Physics, is nowadaysremembered, at all, as a critic of the capitaltheoryof Bohm-BawerkandMarx,as a if contributor the MarxianTheoryof relativevalues andprices andas a statisticianwho appliedthe Poisson to

DAVID W.LAIDLER E. is SpecialAdvisorat the Bankof Canadaandprofessor of economics at the Universityof WesternOntario.GEORGE STADLER W. is lecturerin economicsat the Universityof Newcastle-upon-Tyne.
Journalof Money, Credit,and Banking,Vol.30, No. 4 (November 1998)

Copyright by OhioState 1998 University Press

DAVID E. W. LAIDLERAND GEORGEW. STADLER : 817

show thatthe firsttwo of these raised issues thathave no parallelin modernwork on but the hyperinflation, which may throwextralight on the natureof the "shortageof money"thatit appearedto create.2 Then We begin with a descriptionof the key facts generatedby the hyperinflation. we discuss the state of monetaryeconomics in Germanybefore and shortly afterthe GreatWar,stressingthe antipathyof the majorityof Germaneconomists to the quanexplanationsof the hypertity theoryof money, anddescribeorthodoxcontemporary Cagan's analysis and the contributionsmade by Bortkiewicz, Mises, and inflation. by Hahnin 1923-24 arediscussed in turn,along with the treatment the latterof fiscal factors.We end with a summaryof our findings. andpolitical
1. STYLIZEDFACTSAND GERMANMONETARYECONOMICS

in The Germaninflation,which culminatedin hyperinflation 1921-23, generated four contentiousstylized facts: (1) a phenomenalincreasein prices, well in excess of the increase in the money supply from mid-1921 onward;(2) a more rapidrise (depreciation)in the exchangeratethanin domesticprices;(3) a widely perceived"shortage of money";and (4) an extremelylarge increasein realbalancesin the wake of the stabilizationwhich broughtinflationto an end.3These facts were seen as incompatito distribution the analysis of the frequencyof deathsfrom horse-kicksamong Prussiancavalrymen.Hahn luam was a bankerbased in Frankfurt Main, who became well known laterin the decade for a particularly cid exposition of the process of bankcreditexpansion, and an analysis of the role of the banks in tradecyof prominentlyin debatesaboutthe reconstruction the cle. In the post-WorldWar 2 years, he participated Germanmonetarysystem. reasonsperhapsexplainwhy the workwe referto herehas not previouslyat2. A numberof interrelated amongGermaneconwe tractedthe attention believe it merits.First,the dominantview of the hyperinflation as omists of the period,usuallycharacterized the Balance of PaymentsTheor, was closely associatedwith dealingwith so the RealBills Doctrine.As this view becamediscredited, too did the whole Germanliterature years.HowardEllis of the inflation.Second,this workis not kindlytreatedin the laterliterature the interwar eitherignoresor mis(1934) discusses it only briefly(cf. Ellis 1934,p.217,280,294); andBresciani-Turroni it: interprets Hahnand Mises are referredto only once each in his book, the latterin a footnote. The third, 1931 tr. 1937, pp. 175-82); Bortkiewicz,attractsa full seven pages of commentary(cf. Bresciani-Turroni augmentedvermisleadinglytreatshim as an opponentof an inflation-expectations but Bresciani-Turroni sion of the quantitytheory.Also, as Feldman(1983) has pointedout, most recentresearchon the hyperinflaeconomists,who could not reasonablybe expected to be familiar tion has been carriedout by non-German with whatwas, when all is said anddone, a minorityopinion withinthe body of Germanmonetarythought. thoughnotthose of Hahnor Miss GeraldMerkin(1982) has, however,discussedBortkiewicz' contribution, given is open to question.(see footnote 15.) es; but hereagainthe interpretation 3. Until the end of the war, prices tended to lag significantlybehindthe increasein the money supply. Frombases of unity in 1913, prices increasedto 2.17 by 1918, but the money supplyto 3.75. The accepted duringthe warthatincreasedthe deof explanationof this fact rests on the breakdown creditarrangements mand for money, the extension of territoryover which the Reichsmarkwas legal tender,an increase in hoardingcausedby wartimeanxiety,not to mentionthe influenceof pricecontrols.As soon as the warended, prices andthe money supplybeganto rise significantly,so thatby July 1921 both indexes stood at 14.3, andby January1922 the price index stood at 36.7 andthe money index at 20.5. By the end of 1922 the two indexes stood at 1475 and 213.4 respectively,and the tendencyfor prices to surgeaheadof the money supply persisteduntil stabilizationwas achieved in November 1923. The exchange ratealso rose (that is, the far currencydepreciated) morethanthe money supply,and,until 1923, morethandomestic pricestoo. Furthermore,exchange ratemovementsafterfirstlagging behindthem generallytendedto lead movementsin took hold. As to complaintsof a shortageof money, these occurred domestic price levels as hyperinflation as early as the winterof 1918- 19, when prices still lagged somewhatbehindthe money supply, but at this early stage they probablyreferredto a shortageof small change. After the Treatyof Versailles becameeffective in January1920, such complaints became more frequent,and indeed persisteduntil stabilization a was achieved in 1923. Thereafter, continuinglarge rise in the nominal money supply was absorbedwith

818

: MONEY, CREDIT,AND BANKING

ble with the quantitytheoryof money, but consistentwith a widely acceptedalternative explanationof inflationcast in termsof balanceof paymentproblems.4 Germanmonetaryeconomics (and indeed Germaneconomics in general) during the firstquarterof the twentiethcenturydisplayedcertainunique characteristics that conditionedthe way in which the hyperinflation was usually analyzed.The German language was then used for scientific communicationwell beyond the borders of Germanyitself, but German economics was far less cosmopolitan in outlook than German-languageeconomics. Withinthe GermanEmpire,the subjectwas dominated by the HistoricalSchool, whose views were systematicallyset out in a widely read and immensely influential treatise by Gustav von Schmoller (1900, 1904). This school taught(exactly contrary,for example, to the doctrinesof CarlMengerand his Austriandisciples) thattherewere no universalscientifictruthsto be discoveredwithin the discipline, and that all economic "laws"were specific to particularhistorical and institutionalsettings. The monetary field was heavily influenced by the Historical School, as Haim Barkai(1989) has shown, and after the first decade of the twentieth century it was dominatedby the Chartalist views of Georg-Friedrich Knapp(1905,3rd ed., 1921, tr. 1924). Knappregardedmoney as the creationof law, and arguedthatthe generalacceptabilityof money derivedfromthe legal authority the state,ratherthanbeing, as of Menger(1892) had it, the unintendedsocial consequence of individualmaximizing behavior.It was a shortstep fromthis view to the position thatthe value of money a question of secondaryimportanceto Knapp was also a legal-institutionalmatter, andhis Chartalism quiteantithetical the quantitytheory.The attitudeof the mawas to jority of Germaneconomists, who usually thoughtof the quantitytheoryas asserting the proportionality the price level to the quantityof money, is typifiedby the folof lowing quotationfrom S. P. Altmann's(1908) survey of Germanmonetaryeconomics in the nineteenthcentury.5
Nobody todaycontendsthatan increasein the money supplyhas a proportional effect on prices.The concept of changes in the value of money as a historicalprocess is more and moregenerallyaccepted,a process thatcannotbe elucidatedby the mechanical juxtaposition of money and goods, but requiresthe total, complicatedorganismof the foundation of society to be properlyunderstood.(Altmann 1908, p. 49)
no increasein prices.Throughout hyperinflation, Reichsbank'sdiscount rateremainedludicrously the the low, being highly negativein realterms.Forexample,by October 1923the discountratestood at 90 percent perannum,butbetweenOctober2 and 30, prices rose by a factorof 221. 4. Indeed,this Balanceof Paymentstheorywas so widely accepted,particularly the Reichsbankitself, at and in politicalcircles, thatit was sometimes characterized "official."Thus, referringto it, Bortkiewicz as says that"onlyrecentlyare theresigns of a clear rejectionof this official and semi-official view of the ruin of our currency" (1924a, p.267). See also footnote 9. 5. Further evidence of the slight regardin which the quantitytheorywas held is providedby HowardEllis and Schumpeter.Ellis (1934) talks of "the traditionalanti-quantity theory alignmentof Germaneconomics" (p. 183), while Schumpeter(1954, p. 1104) also implies that the work of eminent German economists was antitheticalto the quantitytheoryand did not move monetarytheoryin Germanyforward in a way comparable developmentsin, say, Britain,America,or Sweden. Therewas some debatein Gerto many as to how much responsibilityKnappbore for the hyperinflation.The consensus seems to be that Knapphimself did not advocateinflation,but thathis State Theory fosteredan antitheoretical climate (see Bortkiewicz 1924b; Melchior Palyi 1924; Felix Somary 1924). However, there is no doubt that some of Knapp'sfollowers "didindeedadvanceinflationary measures"(Ellis 1934, p. l9).

DAVID E. W. LAIDLERAND GEORGEW. STADLER : 819

Insularthough it was, Germanmonetaryeconomics neverthelessfound room for one importantforeign idea, namely, the "real-bills"doctrine, which argued that a securibankingsystem, which confineditself to discountinggood-qualityshort-term ties, would therebyautomaticallyprovideonly the money neededto facilitatethe current volume of trade at currentprices, and would run no risk of excessive money creationcapableof causingprices to increase.This doctrinehadbeen muchdiscussed in the writingsof the BritishBankingSchool in the 1840s, andwas thencetransmitted to the German literature,initially through the writings of Adolph Wagner and Bendixenduringthe wartimephaseof the GerSchmoller,andlaterthose of Friedrich man Inflation.It was a naturalcomplementto the State Theoryof Money, and tailorexplanation attackson any nonmonetary madeas a defense againstquantity-theoretic of inflation,for one of its centralimplicationsis thatmonetaryexpansionis a passively endogenousconsequenceof rising prices, and in no sense theircause.6 conventionalGermanmonetaryeconomics ofIn short,before the hyperinflation, fered no clear-cuttheoryof the price level to replacethe quantitytheory, but nevertheless argued that any observed relationshipbetween money and prices was the resultof "reversecausation."
2. THE BALANCEOF PAYMENTSEXPLANATIONOF HYPERINFLATION

The most prominentexponentof whathas come to be called "TheBalance of Paywas KarlHelfferich.7In the final (6th) ediments"explanationof the hyperinflation, tion of his book Das Geld (Money) (1923) he argued that an adverse balance of payments(resultingfrom the difficultyof obtainingcreditabroadandthe need to imin to by portrawmaterials,factorsexacerbated the requirement makereparations kind to Belgium and France)had caused the Reichsmarkto depreciateon the foreign exhe changes.The depreciation, then argued,had resultedin upwardpressureon prices andwages, which in turnhadcreateda shortageof money;in orderto averta complete breakdownin economic relations,the CentralBankhad been requiredto increasethe circulatingmediumto facilitatetransactionsat the higher prices. He concluded that
Act, andthe debatethat 6. Wagner(1857) containsan extensive discussionof Peel's 1844 BankCharter This discussion shows him to have been an admiringpartisanof the Banking surrounded introduction. its pp. School, particularly Tooke, and an advocate of the Real Bills Doctrine. See, in particular, 119-29. Schumpeter(1954, p. 707, fn.) says that "theGermanenthusiasmfor Tooke as a theorist was, I think, in greatpartdue to the influenceof AdolphWagner"(emphasisin original).Note that,in additionto being an exponentof BankingSchool ideas, Bendixen was second only to Knappas a creatorand exponent of the statetheoryof money. (See Schumpeter1917/18, tr. l 956, p.149, fn.) On the role of BankingSchool ideas in GermanMonetaryEconomics,see also Ellis (1934) p. 180 et seq. 7. Helfferichhadbeen a studentof Knapp,buthe was not an exponentof the StateTheoryof money, but TheodoreGregory remarkedin his Introductionto the 1927 translationof Money (Helfferich 1923, tr. 1927), that "Helfferich's. . . attitudeis perhapsat times more colouredby Knapp'sterminologyand opinions than he was himself awareof." (Gregory 1927, p. viii) Helfferich was Secretaryto the Treasuryand Secretaryof the Interiorduringthe FirstWorldWar, and a Reichstagdeputyfrom 1920 until his deathin architectof 1924. As Haim Barkaiand an anonymousrefereehave remindedus, he was thus an important the wartimepolicies of deficit financeandpricecontrolswhich ensuredthat,in 1919, the Germanmonetary pressures,pressuresthatcould only be exacerbatsystem would be subjectto enormouslatentinflationary ed by the Reichsbank'sdiscountratepolicy. He was, therefore,by no meansa disinterestedobserverof the hyperinflation. The referee suggests, indeed, that it was the debt overhangcreatedby the war, in partby policies for which Helfferichhimself was responsible,thatlay at the root of postwarinflation.

820

: MONEY,CREDIT,AND BANKING

because the currencyhad depreciatedby a far greaterfactor than the money supply had risen, "thecollapse of the Germanexchanges will be seen to be in no way related to the increaseof the note circulation" (1923, tr. 1927, p. 599). The appealof the Balanceof PaymentsTheorylay in its apparent ability to explain the fourcontentiousstylized facts mentionedearlier.In Helfferich'swords:
Firstcame the depreciationof the Germancurrencyby the overburdening Germany of with international liabilitiesandby the Frenchpolicy of violence. Thencefollowed a rise in the prices of all importedcommodities.This led to a generalrise in prices and wages, which in turnled to a greaterdemandfor currencyby the public andby the financialauthoritiesof the Reich; and finally,the greatercalls upon the Reichsbankfrom the public andthe financialadministration the Reich led to an increasein the note issue. In conof trast,therefore,to the widely held view, it is not "inflation" the currency]but the de[of preciationof the currencywhich is the firstlink in this chainof cause andeffect. Inflation [of the currency]is not the cause of the rise in pricesandof the depreciated currency,but the latteris the cause of the higherprices and of the greatervolume in the issue of paper money. (Helfferich 1923, tr. 1927, p. 601)8

Moreover,it was not just data generatedwhile inflation was rising that seemed to Helfferichto confirmhis view. The simultaneous,but short-lived,appreciation the of markand the large increase in the nominal money supply, that took place between Februaryand April 1923, just as Helfferich was completing the final edition of his book, providedfurthersupportfor his position:
It is scarcelypossible more clearly to prove thatprices are independentof the quantitative factor of increase in circulation,and to demonstratetheir dependence upon the courseof the exchangesthanby notingthe fall in the level of pricesandin the ratesof exchange while the note issue was increasing in excess of anything previously known. (Helfferich 1923, tr. 1927, p. 602)

Now balanceof paymentsconsiderationsmust figureprominentlyin any account of the Germaninflation.Such commentators Graham(1930), Feldman(1993) and as Kindleberger(1994) provide ample evidence that the hyperinflation was at least as much a sociopolitical as an economic phenomenon,and that the dislocation of Germany's exportcapacityassociatedwith the war and its aftermath, to mentionthe not prospectiveburdenof meetingpunitivereparations obligationsimposedat Versailles, put important economic and political limits on the WeimarGovernment'sroom for maneuverin economic policy. But Helfferich's position is just as far removedin one directionfromthe kindof balancedeclecticism thatwe findin this historicalliterature as would be a simple mono-causalapplicationof a naive, exogenous money, version of the quantitytheory.He reallydid arguethat:
In consideringthe monetaryconditions in Germany,the view widely held, especially abroad,is basedon the purequantitytheory,and accordinglyregardsthe increasein the
8. We insertthe phrase[of the currency]in the foregoing quotationbecause the use of the word "inflation"to referto risingprices is fairlyrecent.The Concise OxfordDictionary,5th ed. 1967, defines inflation as "anundueincreasein the quantityof money in relationto goods availablefor purchase." This is the sense in which Helfferichand his contemporariesusually used the term.We add similarclarifying phrasesto a numberof quotationsbelow. As an anonymousrefereehas pointedout to us, the Helfferich's descriptionof the "facts"in this quotation is not accurate.The markbegan to depreciateeven beforethe warended (see Feldman,p. 93, Table4).

DAVID E. W. LAIDLERAND GEORGEW. STADLER : 821

circulationof papercurrencyin Germanyas the cause of the rise in the level of German prices and of the depreciationof the currency.On closer examination,however, we find and thatcause and effect are here interchanged, thatthe increasein the amountof paper money circulatingin Germanyis not in fact the cause butthe resultof the fall of the German exchanges and of the consequentialrise in wages and prices. (Helfferich 1923, tr. 1927, p. 598)

The prime evidence he cited to supportthis position was the disproportionamong changesin the exchangerate,the price level, andthe money supply.9
MODELOF HYPERINFLATION 3. CAGAN'S QUANTITYTHEORETIC

Now to a modernreader,the evidence thatso impressedand indeed, in the opinion of many, misled many contemporarycommentatorson the Germanhyperinflation into rejecting any causative role for the quantityof money is easily dealt with: it is expectationson the demandfor only necessary to refer to the effects of inflationary real balances. Even before the event, Marshall(1899) and Pigou (1917) had noted from time to time thatdesiredcash balanceswould be smallerif the currencywas "liable to discredit,"andclosely relatedcommentsareto be found in Fisher's (for example,1911) writingsas well. Afterthe War,bothCannan(1921) and,of course, Keynes attention,also discussed this phepaid particular (1923) to whom Bresciani-Turroni nomenon. Dynamics of Hyperinflation" Even so it was only with Cagan's (1956) "Monetary thatpreciseanalysis of these phenomenabased on the postulateof a stableandconbetween the demandfor real money balancesand the relationship functional tinuous expectedinflationrateappeared. Cagan'smodel may be written
mt-Pt

= a(Etpt+1-Pt) + Y

(1) (2)

Etpt+-Et lPt= (Et lPt Pt)

wherem is the logarithmof the money supply,p is the logarithmof the pricelevel, and ' E the expectationsoperator. In this system the higheris actual,andeventuallythere9. Helfferichwas by no means alone in drawingattentionto this evidence. For example, the American economist Allyn Young, having served as the principaleconomic advisor to the Americandelegation at Versailles, was familiarenough with the Germansituationto see that governmentdebt problems,rather thanbalanceof paymentsissues per se lay at the heartof her economic problems,and he was no exponent of the real bills doctrine.Nevertheless, in 1923, with referenceto Europein general he wrote that the sein quence of cause and effect, particularly the periodfollowing the war, has not been inflation[of the note exchanges,inflation budgets,disordered exchanges,butunbalanced budgets,disordered issue], unbalanced (Young 1923, p. 402). In the specific case of Germanyhe cited the same phenomenaas did Helfferich,conthanthe cause of the decludingthat,"Inlargemeasureinflation[of the note issue] has been the resultrather preciationof the value of the currency"(Young 1923, p. 403). See Mehrling( 1996) for a comprehensive accountof Young's monetaryeconomics. 10. Cagan's system is dynamicallystableif a,B < I, a conditionwhich datageneratedby the Germanhy(amongothers)seem to satisfy. ThoughCagan's ( 1956) point estimatesfor a,B exceeded unity perinflation Khan( 1975) found thatcorrectingfor residualserial for Germanyand the Soviet Union's hyperinflations,

822

: MONEY, CREDIT,AND BANKING

fore expected, inflation,the lower is the quantityof realbalancesdemanded.Hence, it easily accounts for one of the two stylized facts to which, in the 1920s, Germanadherents of the real bills doctrine pointed as being allegedly inconsistent with the QuantityTheory,namely, thatthe Germanprice level (and implicitly also the foreign exchangevalue of the Reichsmark) moved aheadof the quantityof money as inflation gatheredspeed. The Cagan model leaves somethingto be desired, however, when it comes to explainingthe "shortageof money" phenomenon.There are pitfalls inherentin asking people what they think ratherthan observing what they do; but, although remarks aboutthe extra costs associated with managinga smaller, but still equilibrium,cash balance might be expected in Cagan's world, the frequencyof referencesto "shortages" of money duringthe Weimar inflation suggest that agents often found themselves in possession of less cash than they wanted. Their complaints are easier to squarewith the existence of disequilibrium the monetarysystem, with the supplyof in money (nominalandreal)falling shortof the demandfor it, thanwith a simple voluntaryandsuccessfullyexecuteddecision to makedo with a smallerinventoryof cash.l l As we shall see, some contemporary Germancommentaryaddressedthis issue.
4. CONTEMPORARY EXPONENTSOF THE QUANTITYTHEORY

Therewas some discussionof the quantitytheoryin the Germanlanguageliterature before the onset of hyperinflation. had, for example, a prominentplace in Wicksell It ( 1898), Ludwigvon Mises ( 1912) and in JosephSchumpeter( 1917/18), a paperwritten "fromthe conviction . . . thatcurrencypolicy is a field in which thereis no sense

correlationresults in point estimatesof a of less thanone for all the seven hyperinflations studiedby Cagan. Cagan( 1991 ) has revisitedhis studyrecently,and makesthe following remark: While the use of adaptive measuresof expected price changes in estimating the money-demand function in hyperinflation does not appearfar from reality, a more serious problem . . . is that the money stock cannotbe treatedas exogenous. A plausibleway to endogenize the money stock is to model the revenueneeds of the governmentfor an inflationtax. Unfortunately, one has successno fully shown how to expressthose needs andtheirrealizationas a functionof independent variables." (Cagan 1991, p. 559, fn.) The main thrustof this paperis to show that adaptiveexpectationsperformwell in a demand for money function,relativeto rationalexpectations,as exemplifiedby Frenkel' ( 1977) use of the ratioof the forward s exchange rateof the Reichsmark its concurrentspot rate.Caganinterprets resultas the consequence to this of "rational learning"in the face of a major,initially unperceived,long-runshock to the time pathof inflation. 11. We areawareof the logical possibility that,in an economy in which the aggregatedemandand supply for money are in equilibrium,some individualagents may find themselves with excessive cash balances, and some may experiencea shortage.However, in a monetarysystem in which idiosyncraticshocks can have such effects, it is also possible thatan economy-wide disturbance push the majorityof agents can "off' theirfunctionsin the same direction.Since the monetarydisturbancesassociatedwith hyperinflation are predominantly economy wide, and since we are unawareof any evidence of complaintsof cash surpluses thatmightcounterbalance complaintsof shortagesto which we herecall attention,we do not find the the above-mentioned logical possibilityof zero aggregateexcess demandfor money a compellingexplanation of these complaints.Hence, we suggest thatthe "shortages" questionwere indeedan aggregatephein nomenonwhose explanationalong the lines we discuss here is worthtakingseriously.

DAVID E. W. LAIDLERAND GEORGEW. STADLER : 823

whateverin approaching practicalproblemsbefore we have become clear on the fundamentalquestionsof money"(Schumpeter1917/ 18, tr. l 956, p. l 49). l2 Even so, while the hyperinflationwas in progress, only a few German-speaking, and even fewer German,economists argued that its root cause was excessive creation of money. Perhaps the best known of these nowadays is Walter Eucken (1923), who emphasized the monetizationof budget deficits as the driving force in the inflation. Eucken, however, did not discuss a potential role for expectations in the inflationaryprocess, and hence had difficulty in coping with some of the hyperinflation's salient empirical characteristics, as Ellis (1934, pp. 224-30) noted. Some exponents of a monetary explanation of the hyperinflation, notably Hahn, Bortkiewicz, and Mises did, however, rely on an analysis of inflation expectations to reconcile that explanationwith the facts, including the "shortage"of money. All three attendedthe 1924 meeting, dedicated to a discussion of the hyperinflation,of the Verein Sozialpolitik, organizationof German-speaking fur an economists similar to the American Economic Association, but founded earlier, in 1872. Bortkiewicz presenteda paper at this meeting, while Hahn and Mises made important contributionsto the discussions which took place there, based on their recent writings (Mises 1923, Hahn 1924a).13 Confidencein the currencyor, moreprecisely,expectationsof changes in its value, were repeatedlystressedby these economistsas centralto an analysisof the inflation. Hahn(1924a, 1924b)in particular emphasizedthe effect on the demandfor money of expected inflation.For him, "confidence," what he sometimes called "thequalitative factor"affectingthe value of a currency,dependedon the value of goods thatcurrency could buy in the future,andthus on expected price inflation.Furthermore, expected future price movements, though they had effects on the desirability of money holding similarto interestratechanges, were likely to be more important.
Hopes of a rise or fear of a fall in the value of money means that a positive or negative premiumis addedto the real value of money. The effect is thereforeprincipallythe same as thatof a rise or fall in the interestrate.However, thereare two differences.Firstly,the effect of variationsin confidencein the currencyareusuallymuchstronger,becausethey concernfar largermargins,thanfor regularinterestratechanges. Forexample, a currency depreciationof some magnitudeis usually not only equivalentto a fall in the interest rateto zero, but is equivalentto a negativeinterestrate.(Hahn 1924a, p. 305)
12. In the light of the dominanceof the HistoricalSchool withinGermany,it is not accidentalthatnone of these contributions came from Germaneconomists. Note thatthe 1912 edition of Mises' work remains untranslated. second edition(Mises 1924, tr. 1934), a considerablyextendedandrevisedwork,is availThe able in English. It is worthnoting, as Denis O'Brien has pointedout to us, thatMises became increasingly hostile to the quantitytheory,andparticularly emphasison the conceptof an aggregateprice level, as his its careerprogressed.For Mises, even in 1924, the criticalconsequenceof monetaryexpansionwas its effect on the relativepriceof capitalandconsumptiongoods. However,in 1923-24, his antipathyto the quantity theorywas far less intensethanit would laterbecome. 13. Unlike the othercontributions discussedhere,Mises' 1923 paperhas been previouslytranslated into English underthe title "Onthe Manipulation Money and Credit."See Mises ( 1923, tr. 1977). We prefer of to use our own translation, however,because this translation been re-edited,with, for example, section has headingsinsertedthatare not in the originalpaper.It also misses the subtletyof Mises' meaningin several places as readerswill see if they comparesome of our translated passages to those of this version. In fairness to its translator, 1977 version was clearly aimedat a generalcontemporary this audiencerather thanat specialisthistoriansof thought.

824

: MONEY.CREDIT,AND BANKING

Hahn arguedrepeatedlythatconfidence in the futurevalue of the currencydetermined the demandfor it, and thus the velocity of money. As demandfor money fell, and people began to economize on money balances, so prices, exchange rates,not to mentionthe bankmoney multiplier,would begin to rise, along with velocity:
The individualtransactiondetainedcash balances for a shortertime period than previously. The phenomenonof a progressivedepreciationin the value of money, causingreduction of cash balances known from previous inflationaryperiods manifested itself. This has the consequencethatthe quantityandvelocity of bankmoney circulating in the economy can rise andtherewiththe price level andforeignexchangerate even if the quantityof banknotesin circulationdoes not rise accordingly,in otherwords, that prices andexchange ratesrise fasterthanthe quantityof notes in circulation.... The reasonfor the manifestationof a rise in velocity of money is well known. On the expectationof continuedlosses associatedwith every kind of money holding, everyone attemptedto exchange money, that until now he had been accustomedto retainingfor longeror shorterperiods,as quickly as possible be it againstcommoditiesor foreign exchange in orderto transfer loss involved onto the next recipient.A mentalitydethe veloped towardthe Germanmarkthat one can most succinctly describe as "playinga game of SchwarzerPeter." (Hahn 1924a, p. 295) l 4

ThusHahn,whose analysisis in this respectsimilarto thatof Cagan,andof BrescianiTurroni,too, saw the rise in velocity as the consequenceof a dramaticdecline in demandfor money as the inflationprogressed. The positiontakenby Bortkiewicz( 1924a) was rather different.He arguedthatexpectationsof futurenote issues hada directimpacton current prices, which in turnrequiredmoney to turnover morerapidlyto financecurrenttransactions: 5
I readilyadmitthatsuch a rise in the velocity of money occursin reality.In additionI believe that mistrustin the currencyplays a role in this; but I envisage the chain of causation in questiondifferentlyto most authorswho have recentlyapproached question. this For I am of the opinionthatmistrustin the currency,the expectationof furtherissues [of money] and the consequentfall in the value of money has an immediateimpacton the price level, so directlystrengthening impactof an increasedmoney supply.Thatthe the price level rises proportionally morethanthe money supplyin this mannernow causes a shortageof money, which one attemptsto overcomeby changingpaymentspracticesaccordingly, and thereby, in one way or another, raising the velocity of money. This process of adjustment eventually lead to a very considerablerise in the velocity of can money. Accordingto this, the rise in the velocity of circulationof money is not the cause of the sharper depreciation money,buta consequenceof the latter,a kindof adaptation of to changedcircumstances.(Bortkiewicz 1924a,p. 266)

14. In the Germancardgame SchwarzerPeter the objective is to rid oneself of a particular cardby passing it on to otherplayers.The playerholding it when the game ends is the loser. Merkin(1982, pp. 28-30) refersto the SchwarzerPeter Theorieof hyperinflation, without,however,citing Hahn's work. 15, Merkin( 1982, p. 30) also quotes the bulk of the following threepassages, albeit with minordifferences in translation, his discussion stresses Bortkiewicz's comments about "mistrust" but ("lackof confidence" in his translation)of the currency,and not his view of the source of this mistrustas being "the expectation of furtherissues of money" which strengthensthe effects of "an increased money supply" ("quantity money"in Merkin's translation).Hence MerkintreatsBortkiewicz's views as being further of removedfromthe quantitytheorythanwe believe can be justified.He arguesthatBortkiewicz'sfinal insistence on the prime importanceof an increasingmoney supply as the factorundermining confidence,containedin the fourthpassagequotedbelow, is inconsistentwith whatprecededit. We disagree.(See Merkin 1982, pp. 30-32, p. 45.)

DAVID E. W. LAIDLERAND GEORGEW. STADLER : 825

Furthermore, Bortkiewiczsaw the exchangerateadjustingfasterthandomesticprices andpulling domestic prices up behindit:

The foreignvaluation thecurrency a countxy of of pursuing policy of inflation[of the note a issue] can manifestitself as a factorthatdrivesup the domesticprice level, especiallyif it is less favorablethanis justifiedby theactualvolumeortempoof issues of money.At a certainstage of the inflation[of the note issue] it is the externalvalueof the currencythatdeterminesthe internal value, andnot the otherway round.(Bortkiewicz1924a,p. 266).

And, he went on,

the true chain of causality is thus in my opinion: mistrust,resultingin a disproportionately large rise in exchange rates,consequentlydomestic prices and wages thathave no relationshipto the actual rise in the money supply, wherefrom - a furtherlink in the causalchain a shortageof money arisesthatthen resultsin a rise in the velocity of circulationin one way or another.(Bortkiewicz 1924a, p. 267)

This sounds suspiciously like a concession to the Balance of Paymentstheorists, and indeed this is precisely the misleading interpretationthat Bresciani-Turroni (1931, tr. 1937, p. 176) was laterto give to Bortkiewicz's theory.However, and crucially, Bortkiewiczexplicitly criticizedHelfferichfor
overlookingthe fact thatit is in the finaljudgmentinflation[of the note issue] thatstands at the beginningof the chainof causation.For how does lack of confidence,the unfavorable foreign valuationof the money of a paper-currency countryarise, if not throughan excessively largeincreasein its money supply?(Bortkiewicz 1924a, p. 267)

By giving a centralrole to the impactof expectationsof the futurelevel of the money supply on exchange rates and prices, Bortkiewicz could thus explain the disproportionaterise in exchangeratesandprices, as well as the shortageof money; and the lag in prices behindexchange ratesalso followed naturallyin this chain of causation. Bortkiewiczthereforeexplicitly concludedthatone did not have to rejectthe quantity theoryof money in orderto explain the hyperinflation:
One does not have to standthe QuantityTheoryon its headin orderto understand disthe proportionality between the rise in the money supply and the rise in the price level that occurs at a particular phase of the inflation.It is far more valid to shape it into a version thattakes accountof the fact thatthe level of the value of money does not exclusively depend on the quantityof the actualnote issue, but simultaneouslyon the expectedvolume offuture issues. The expectationsof the public with respectto the shapingof monetary circumstancesin the futureplay a role here. Confidenceor lack of confidencebecomes relevant as an independentfactor besides the actual quantityof money (Bortkiewicz 1924a,pp. 167-68, ouremphasis)

Although his argumentwas not couched in the mathematicalrigor of the present day, the use of forward-looking expectationsaboutmoney creationsurelyrepresented a considerableinsight on Bortkiewicz's part.If equation(1) above is solved forwardin the spiritof rationalexpectationstheorizing,one obtains 1- a (a _ 1) (Ermr+e y), (3

andprices today are seen to dependon the entirefuturetime pathof the money stock. To this extentBortkiewicz's analysiswas closer in spiritto thatof Sargent( 1982) than

826

: MONEY,CREDIT,AND BANKING

of Cagan, though unlike either of them, and unlike Hahn, he saw inflation as a disequilibriumphenomenonwhere "theexpectationof furtherissues . . . has an immediate impact on the price level," so that prevailingprices do not necessarily bear any relationshipto the current quantityof money.16 Mises (1923) took a similarposition. He too treatedthe expected time path of the money supply as exertingan influenceon the currentbehaviorof the price level, over and above thatof the quantityof money currentlyin circulation.
If the futurepathof the value of money is judged to be unfavorable,then, in anticipation of its expecteddecline in value, it will be valuedat less todaythanwould accordwith the actualcircumstancesof money demandand supply. Prices will be set and paid that do to not correspond the presentquantityof money in circulationandnot to the presentstate of demandfor money, but to futureconditions.(Mises 1923, p. 6, our emphasis)

Hence, Mises saw both buyersand sellers as willing to transactat prices thatdepended on expected, not actual, monetaryconditions, and was thus able to explain the shortageof money:
Thereis not enough money availableat the momentto pay for prices thatcorrespondto the expected futuresupply and demandfor money. Hence tradebegins to suffer from a shortageof notes: there is insufficientcurrencyon hand for the completion of agreed The marketmechanism,thatbrings aboutequality between aggregatedetransactions. mandandaggregatesupply,no longeroperatesto createthe exchange relationshipsthat exist between money and other economic goods. One could see this clearly in the late Fall of 1921 in Austria.The completionof business transactionssufferedseverely from a shortageof money. (Mises 1923, p. 7)

In such circumstancesincreasingthe money supplyfurtherwould only make matters 17 worse and furtherfuel the inflation:
Were one, as some people suggest, to increasethe note issue even further,the situation would only deterioratefurther.Because, in this case, the panic would increase further, between the depreciationof money andthe quantityin circulation the disproportionality would become intensified.(Mises 1923, p. 7)
AND THE VELOCITYOF MONEY 5. CURRENCYSTABILIZATION

Bortkiewicz,Hahn,and Mises all realizedthat stabilizationof the currencywould reducethe velocity of money significantly.During a deflation,said Bortkiewicz,the
16. The Vereinproceedingsof 1924 recorda ratherfutile debatebetween Bortkiewiczand Hahn,which this seems to have arisenbecause neitherauthorappreciated differencebetween theirpointsof view. In the paperhe presentedto the Verein,Bortkiewiczconsidereda simple numericalexample which laterattracted (1931, tr. 1937, pp. 175 et seq.). It was aimed at Hahn,and unfavorableattentionfrom Bresciani-Turroni thatthe mere fact of consumersspendingmoney more quickly would not necespurported demonstrate to sarily alter velocity; the money might insteadsimply stay longer in the tills of producers,and it would requirea reductionin the periodbetween income paymentsbefore velocity rose (Bortkiewicz 1924a,p.263). Hahnrepliedthatthis example was unrealistic:duringa time when money w,aslosing its value, producers would be no more willing to hold extra balancesthanconsumers,and velocity must rise (Hahn, 1924b, p. Hahn(1924b, pp. 303-304) arguedthat if lack of confidence in the currencywas seen 303). Furthermore, to bid up prices independentlyof changes in the money supply or its velocity, then this would violate the quantitytheory.Hahndid not admitthe possibility thatP could rise unless MVrose simultaneouslyto perwas mit higherprices.He missed the pointthat,for Bortkiewicz,hyperinflation not an equilibriumprocess, thatpricescould be set in anticipationof a continuingrise in the money supply,and,crucially,thatthe volmightbe reducedas a consequenceof a resultingshortageof money. ume of transactions similarvein in the second edition of The Theor}of Mo)1ew 17. Mises expresses himself in a remarkably bInd Credit,which appeareda year afterhis Vereinpaper.Indeed,he seems to have lifted passages from the pp. Vereinpaperalmost verbatim.See particularly, 228-29 of Mises (1924, tr. 1934).

DAVID E. W. LAIDLERAND GEORGEW. STADLER : 827

velocity of money would tend to fall as people increasedtheircash balances.This was not only a possibility in the case of an actualdeflation,"butalso applicablein the case of stabilizationof the currency without shutting down the note-printingpress, for which Austriaprovidesan example. In this case a rise in the price level is avoidedby the startof an intensive savings process"(Bortkiewicz 1924a, p. 273). Mises (1924) also discussed recentAustrianexperience.
The fact thatthe circulationof notes has grownconsiderablyin the pastyearwithoutrenderingthe maintenanceof the actualconvertibilityof notes [into dollars]impossiblefor the Bank shows that today the economy needs more Austriancurrencyagain; foreign money, which in the summerof 1922 had alreadysatisfieda significant,possibly even the greaterpartof the Austriandemandfor money, has, now thatconfidencein Austrian monetarypolicy has returned,been replacedby Austrianmoney. (Mises 1924, p. 279)

Referringto events in Germanyduring the period Februaryto April 1923, when prices remained roughly constant, the currency appreciated,but the money stock doubled, events which Helfferich had treated as incompatible with the quantity theory, Hahn commented that "this requiresno furtherexplanation.The phenomenon rested on the fact that as a result of the returnof a certain confidence in the mark, the velocity of circulation of bank deposits as well as of currency was reduced" (Hahn 1924a, p. 295). Hahn also discussed the stabilization programme proper,which began in November 1923. He describedits effect on money balances as follows:
If the velocity of bankdepositsdeclines in consequenceof a rise in confidencein the currency, so no alleviation [of the shortageof money] will initially occur insofar as the reduction in velocity follows a maximum level because the money base [Reichsbanksgeld;] also change its velocity . . . will A strikingillustration the case describedhereis providedby the unusualsituationin of the money marketafterthe introduction the Rentenmark. of Because of the spontaneous reductionin the velocity of money froma maximum,no relief of the strainon the money marketoccurredfor the time being, and money remained incredibly short [unerhort knapp].First of all the cash boxes and cash registersfilled themselves. Only when the cash registersand pocket book balances had acquireda normallevel did money again take on the form of depositswith the banks,whereupona quiteextraordinarily strongrelief of the situationin the money marketcertainlyoccurred,comparedto the time before the introduction the Rentenmark. of (Hahn 1924a, pp. 310-1 1)

Thus,like Sargent(1982), the Germanquantitytheoristssaw a rise in confidencein the currencyas increasingthe demandfor money, so reversingthe processthathadled to such a dramatic decline in realcash balancesduringthe hyperinflation. Unlike Sargent, however, they did not view the disproportionate in the nominalmoney suprise ply relative to prices after stabilizationas a fact thatmight seem to be "violatingthe quantitytheoryof money"(Sargent1982, p. S4).

6. FISCALPOLICYAND SOCIOPOLITICAL FACTORS

Sargent(1982) arguesthata necessaryconditionfor ending hyperinflation a fisis cal policy reformthatobviates the need for inflationary finance.Examiningthe great inflations that occurred in Austria, Hungary,Germany,and Poland after the First

828:

MONEY CREDIT.ANDBANKING

WorldWar, he concludes thatit was not merely a change in monetarypolicy per se, but the coordination fiscal andmonetaryreformthatendedthese inflations,andthat of a merecommitmentto price stabilityby the CentralBank would not have been credible in the absence of a fiscal reform.The economists whose work we have discussed in this paperwere also well awareof the role of governmentfinancesin the inflation though they did not integrateit into their analysis of inflationexpectations with the same systematicthoroughness. 18 Hahnarguedthatprice stabilitycan be maintainedin the face of changes in the velocity of money providedthe money supply can be altered to offset such changes. Problemsonly beganto arise when
it is not possible to stop the increasein notes from enteringcirculationand to withdraw notes fromcirculation.This is particularly case when the statebegins to cover its exthe penditureswith the help of the CentralBank ratherthan throughtaxation or issuing bonds.Forin this case, the CentralBankdoes not have the opportunity withdrawbankto notes fromcirculations becausethe debtor,the State,is normallynot in a positionto meet its obligations.As a generalrule, the financialplight of the State is the cause why it, instead of financingits expendituresthroughtaxes or borrowing,looks for access to the CentralBankthatcan cover its financialneeds throughad hoc newly createdpurchasing powersusuallyprovidedin exchangefor TreasuryBills. (Hahn 1924a, pp. 3 13-14)

Furthermore, best way of preventinginflationand currencydepreciationwas "inthe creasingthe economic strengthof the statethroughfiscal policies so thatit can repay its floating debt." For, "if there are no more governmentbonds in circulation,they cannotbe encashedat the CentralBank"(Hahn 1924a,p. 314). Even so, Hahndid not view a fiscal reformas the sine qua non of ending inflation.He went on to say thatif fiscal reformis impossible,an alternativesolutionmightbe to returnto the gold standard,which would restoreconfidencein the currency. Bortkiewicz also argued (1924c) that it was fiscal mismanagementthat initially gives rise to inflation,and suggestedthatthe circumstanceunderlyingsuch mismanagement is invariablya war, regardlessof its outcome. Indeed, in his view, the extreme fragility of Germany'ssociopolitical fabric in the wake of World War I may well have madeinflationthe most appropriate policy for the authoritiesto follow.
In monetaryand in CentralBank policy one cannot separatethe political from the economic factorsandcannotoverlooksocial factorseither.It is no doubtproperto mention a remark PrivyCounsellorHarmshere,who has generouslyabstainedfrom speaking: by the inflationsaved us froma social revolution.I subscribeto this opinion, naturally without identifyingmyself with all the measurestakenin the areaof money andcreditduring the warandpostwarperiod.(Bortkiewicz 1924c, p. 321)

Mises too conceded that,given the sociopoliticalconditionsprevalentin Germany in the early 1920s, inflatingthe currencyto defrayreparations paymentshadbeen unavoidable:
18. Eucken( 1923) was also awareof these issues. See above. Note that,in general,the Germanquantity theoristssaw the root of the inflationas Iying in the financingof the GreatWar,because the government had to an extent resortedto the printingpress for this purpose.However, they differedas to whetherinflation of the money supplyhadbeen essentialto financethe war.See, for example,GeorgBernhard 1924, pp. ( 28 1-82).

DAVID E. W. LAIDLERAND GEORGEW. STADLER : 829

The Germangovernmenthas no alternativeway of covering its reparations obligations. It would have no success if it triedto raise the sums demandedby issuing bonds or raising taxes. Given the way matterscurrentlystand with the Germanpeople, a policy of compliancecould not count on the consent of the majorityif its economic consequences were clearly understoodandthey were not deceived as to its costs. Publicopinion would turnwith elemental force against any governmentthat were to try to fulfill the obligations unde}taken towardthe Allied Powers completely. (Mises 1923, p. 31)

He thus suggested thatthe hyperinflation was the best responseavailableto the German govetument in the face of politically unsupportable demandsfor reparations.l9
7. CONCLUSIONS

In this paperwe have shown thattherewere economists in Germany,albeit a small minority,who, at the time of the Weimarhyperinflation, favoreda monetaryexplanation of the phenomenon,understoodthatexpectationsof inflationwould cause veloc19. Mises was, of course, always a strong advocate of sound money, and the view quoted in the text stemmedfroma politicaljudgment,which receives some supportfromothersources. In his classic studyof the Germaninflation,FrankGrahamsays that the will to check depreciationin Germanywas muchweakerthanin othercountriessince the Germans were convinced, by no means withoutjustification,that improvementin the public finances would lead to still more severe exactions on the partof the victors in the war. (Graham1930, pp. It would be wrong, however,to assume the hyperinflation the resultof a cynical calculationby the was Germangovernment.It was a time of "frequent changes of government,but also of repeatedputschistefforts, political violence andassassinations" (Feldman1993, p. 310). Feldmanconcludes that Germany'sleaderscertainlydid not planeitherthe inflationor the hyperinflation. the face of exIn tremedomestic and international constraint,however,they eitherfound it necessaryor chose to exploit inflationaryopportunitiesat crucial points, especially in the fall of 1922, and failed to take appropriate measuresto controlthe avalanche." (Feldman 1993, p. 838) The view, that the fiscal and monetaryauthoritiesof the countriesvanquishedin the FirstWorld War were simplyoverwhelmed the sociopoliticalconsequencesof theirdefeat,was echoedby BaronAlexander by von Spitzmullerat the 1924 Verei)lmeeting. He had been governorof the AustrianCentralBank throughout the Austrianhyperinflation, which had ended in September1922, morethana year before thatof Germany,and he confessed that,in theoryat least, he had hadthe power to stop the note-printing presses. During the war the Austro-Hungarian bank was led by two people of impeccable gold standard credentials . . . They, and also especially the HungarianFinance Minister at the time, Teleszky, madethe greatesteffort to containthe inflationduringthe war,but withoutsuccess. I thinkthatthis failurecan be tracedbackto the process thatHerrGoldscheidtodaycalled "beingforced to accommodateto existing political imperatives."[ZvtaelgscExpcessuelg cl bestehendepolitische Machtserhaltnisse.] To a certainextent this was also true in the postwarperiod, when I functionedas Bank governor. I believe that the social conditions in Austria after the collapse [of the Empire in late 1918] were such that one could not have managedwithout inflation[of the money supplyl. This shredof landthatremainedafterthe appallingwreckage,experiencedsocial circumstanceswhich, if one had also added restrictivemonetaryconditions, would undoubtedlyhave led to social collapse. I was the firstto try to workagainstthe inflationary tendencyby stoppingnote issues; but that was out of the question.Then I pointedto my legal rights thatwere, mindyou, empty of any content, althoughI did in fact assertthem anddeclaredthatI could shutdown the note-printing press any day. ThereuponI was told: Quite right, you will then create complete economic chaosileiz7 u irtschaftlichesTohuviabohu] for a week and then we will establish a state bankingdepartment, like they have in Czechoslovakiaand Hungary.In otherwords, by shuttingdown the printingpress I would have accomplishedan act of futile heroics [herostrcitisch], withoutalteringthe long-term course of events in the slightest. (Spitzmuller1924, p. 312)

830

: MONEY,CREDIT,AND BANKING

ity to rise as inflationrose, and to fall again as stabilizationtook hold, had no difficulty in reconcilingthe behaviorof the exchangeratewith a monetaryexplanationof inflation,anddid not findthe "shortage money"phenomenonpuzzling. We have also of shown thatthey paidattentionto the fiscal situationof the Weimargovernment,andto its political vulnerability,as factors affecting confidence in the currency. Furthermore, two of them, BortkiewiczandMises, advanceda disequilibrium analysis of the effects of expectationsabout the futurecourse of monetarypolicy on currentpricesettingbehaviorand on velocity thatseems to have no exact parallelin modem work. It may be that, upon furtherinvestigation,the line of enquirywhich they opened up would prove to be flawed. We intendto take no position on this matterwhen we suggest that such furtherinvestigation would nevertheless be worthwhile, and that the very fact thatthe literature have here discussedcontainsthese novel ideas makesit we more thana historicalcuriosity.
LITERATURE CITED

Altmann,S. P. "ZurdeutschenGeldlehredes 19.Jahrhunderts." Festgabefur Schmoller:Die In Entwicklung deutschenVolkswirtschaftslehre 19. Jahrhundert. der im Leipzig, vol. 1, sec. 6 (1908),pp.1-67. Barkai,Haim. "TheOld HistoricalSchool: Roscheron Money and MonetaryIssues."Historv of Political Economy21 (Summer1989), 179-200. Bernhard, Georg. "Discussion."VereinfurSozialpolitik: Schriften170 (1924), 280-87. Bortkiewicz,Ladislausvon. "Die Ursacheneiner potenziertenWirkungdes vermehrtenGeldumlaufsauf das Preisniveau." Verein Sozialpolitik:Schriften 170 (1924a), 256-74. fur . "Discussion."VereinfurSozicllp(71itik: Schriften170 (1924b), 290-91. . "Schlusswort." Verein Sozialpolitik:Schriften170 (1924c), 317-21. fur Bresciani-Turroni, The Economicsof Infation, 1931. As translatedby M. E. Sayers, 1937, C. reprinted, New York:AugustusKelley, 1968. Cagan, P. "The MonetaryDynamics of Hyperinflation." Studies in the QuantityTheorwZ In of Money,edited by M. Friedman. Chicago:Universityof Chicago Press, 1956. . "Expectations the GermanHyperinflation in Reconsidered." Journal of International Moneyand Finance 10 (December 1991), 552-60. Cannan,E. "TheApplicationof the TheoreticalApparatus Supply and Demandto Units of of Currency." EconomicJournal 31 (1921), 453-61. Ellis, H. GermanMonetaryTheory,1905-1933. Cambridge,Mass.:Harvard UniversityPress, 1934. Eucken,Walter.KritischeBetrachtungen zum Deutschen Geltproblem.Jena:Gustav Fischer, 1923. Feldman,GeraldD. "TheHistorianand the GermanInflation.9' InXqati(7n In throughthe Ages: Economic,Social, Psychological, and HistoricalAspects, edited by NathanSchmuklerand EdwardMarkus.New York:BrooklynCollege Press, 1983. . The Great Disorder: Politics, Economics, and Sc)cietv in the German Inflation, 1914-1924. Oxford:OxfordUniversityPress, 1993. Fisher,Irving.ThePurchasingPower of Monev.New York:Macmillan, 1911. Frenkel,J. "TheForwardExchangeRate, Expectations,and the Demandfor Money:The German Hyperinflation." AmericanEconomicReview67 (September1977), 653-70.

DAVID E. W. LAIDLERAND GEORGEW. STADLER : 831

Graham, Frank D. Exchange, Prices, and Production in Hyper-Infation: Germany, 1920- 1923. New York:Russell and Russell,1930. Gregory,T.E."Introduction"toHelfferich,1923,translated1927. Archivfur SozialHahn,L. Albert."ZurFragedes sogennanten'Vertrauensin die Wahrung."' wissenschaftundSozialpolitik52:2 (1924a), 289-316. fur . "Discussion."Verein Sozialpolitik:Schriften170 (1924b),301 - 306. by Helfferich,Karl.Money, 6th ed. (1923). Translated L. Linfield,Adelphi. New York, 1927. Keynes,J. M. A Tracton MonetaryReform.London:Macmillan,1923. A Khan,Moshin S. "TheMonetaryDynamics of Hyperinflation: Note." Journal of Monetary Economics(1975),337-60. CharlesP. "The Great Disorder:A Review." Journal of Economic Literature Kindleberger, 32:3 (September1994), 1216-25. as Knapp,G. F. Die StaatlicheTheoriedes Geldes, 1905. 3rd 1921 ed. abridgedand translated TheState Theoryof Moneyby H. M. LucasandJ. Bonar,London, 1924. ficial PaCommission"(1899). Reprintedin Of Marshall,A. "Evidenceto the IndianCurrency pers of AlfredMarshall,editedby J. M. Keynes. London:Macmillan,1926. Mehrling,P. "TheMonetaryThoughtof Allyn Abbott Young."Historyof Political Economy 28 (Winter1996), 607-32. Menger,Carl."Onthe Originof Money."EconomicJournal2 (June 1892), 239-55. Observations." Some Preliminary a Merkin,Gerald."Towards Theoryof the GermanInflation: In The GermanInfation Reconsidered:A PreliminaryBalance, edited by GeraldD. FeldWitte.BerlinandNew Gerhard Ritter,andPeter-Christian A. Holtfrerich, man,Carl-Ludwig York:Walterde Gruyter,1982. 1 Mises, Ludwig von. Theoriedes Geldes undder Umlaufsmittel, st ed. Munich, 1912. Vereinfur Sozialpolitik: . "Die GeldtheoretischeSeite des Stabilizierungsproblems." as Schriften164 (1923), 1-37, translated On the Manipulationof Moneyand Credit.Dodds Ferry,N. Y.: Free Marketbooks (tr.anded. Percy L. Greaves,Jr.) fur . "Discussion."Verein Sozialpolitik:Schriften170 (1924), 275 - 80. . The Theoryof Moneyand Credit, 1924. 2d ed., English tr. by H. E. Batson. London: JonathonCape,1934. Verein Sozialpolitik:Schriften170 (1924),320-23. fur Palyi, Melchior."Schlusswort." Journalof Economics32 (1917), 38-65. Pigou, A. C. "TheValue of Money."Quarterly Sargent,T. J. "The Ends of Four Big Inflations."In Infation: Causes and Effects, edited by R. E. Hall. Chicago:Universityof Chicago Press,1982. (2 derAllgemeinenVolkswirtshaftslehre vols.). Leipzig und Schmoller,Gustavvon. Grundriss Munchen:DunkerundHumblot,1900/04. Schumpeter,Joseph. "Das Sozialproduktund die Rechenpfennige."Archiv fixr SozialwisMargetas "Moneyand 44 senschaftundSozialpolitik (1917/18),627-715. Tr.1956 by Arthur EconomicPapers,no.6, London:Macmillan,pp. l48-211. International the Social Product," . Historyof EconomicAnalysis. New York:OxfordUniversityPress, 1954. fur Verein Sozialpolitik:Schriften170 (1924), 323-25. Somary,Felix. "Schlusswort." Spitzmuller, Alexander Baron von. "Discussion." Vereinfur Sozialpolitik: Schriften 170 (1924), 310-13. Wagner,Adolph.Beitrage zur Lehrevon den Banken.Leipzig:Leopold Voss,1857. Wicksell, K. Geldzins und Guterpreise, 1898. (Interestand Prices, tr. R. F. Kahn, London: Macmillan,for the Royal Economic Society, 1936.) Foreign Affairs2 (PartB) (1923),397-409. Young, A. A. "WarDebts, Externaland Internal."