The Ends of Four Big InflationsFig. 2.1 Wholesale prices in Austria. unbacked money to finance government deficits. 5 This was done on such a scale that it led to a depreciation of

This PDF is a selection from an out-of-print volume from the NationalBureau of Economic Research

Volume Title: Inflation: Causes and Effects

Volume Author/Editor: Robert E. Hall

Volume Publisher: University of Chicago Press

Volume ISBN: 0-226-31323-9

Volume URL: http://www.nber.org/books/hall82-1

Publication Date: 1982

Chapter Title: The Ends of Four Big Inflations

Chapter Author: Thomas J. Sargent

Chapter URL: http://www.nber.org/chapters/c11452

Chapter pages in book: (p. 41 - 98)

The Ends ofFour Big InflationsThomas J. Sargent

2.1 Introduction

Since the middle 1960s, many Western economies have experiencedpersistent and growing rates of inflation. Some prominent economistsand statesmen have become convinced that this inflation has a stubborn,self-sustaining momentum and that either it simply is not susceptible tocure by conventional measures of monetary and fiscal restraint or, interms of the consequent widespread and sustained unemployment, thecost of eradicating inflation by monetary and fiscal measures would beprohibitively high. It is often claimed that there is an underlying rate ofinflation which responds slowly, if at all, to restrictive monetary and fiscalmeasures.1 Evidently, this underlying rate of inflation is the rate ofinflation that firms and workers have come to expect will prevail in thefuture. There is momentum in this process because firms and workerssupposedly form their expectations by extrapolating past rates of inflationinto the future. If this is true, the years from the middle 1960s to the early1980s have left firms and workers with a legacy of high expected rates ofinflation which promise to respond only slowly, if at all, to restrictivemonetary and fiscal policy actions. According to this view, restrictivemonetary and fiscal actions in the first instance cause substantial reduc-

Thomas J. Sargent is with the Federal Reserve Bank of Minneapolis and the Departmentof Economics, University of Minnesota.

The views expressed herein are solely those of the author and do not necessarily representthe views of the Federal Reserve Bank of Minneapolis or the Federal Reserve System.Helpful comments on earlier drafts of this paper were made by Preston Miller, JohnKennan, Peter Garber, and Gail Makinen. General conversations on the subject withMichael K. Salemi and Neil Wallace were most helpful. Gail Makinen directed the author'sattention to the unemployment figures for Poland. Carl Christ of the NBER readingcommittee made several comments that improved the manuscript.

41

42 Thomas J. Sargent

tions in output and employment but have little, if any, effects in reducingthe rate of inflation. For the economy of the United States, a widely citedestimate is that for every one percentage point reduction in the annualinflation rate accomplished by restrictive monetary and fiscal measures,$220 billion of annual GNP would be lost. For the $2,500 billion UnitedStates economy, the cost of achieving zero percent inflation would begreat, indeed, according to this estimate.

An alternative "rational expectations" view denies that there is anyinherent momentum in the present process of inflation.2 This view main-tains that firms and workers have now come to expect high rates ofinflation in the future and that they strike inflationary bargains in light ofthese expectations.1 However, it is held that people expect high rates ofinflation in the future precisely because the government's current andprospective monetary and fiscal policies warrant those expectations.Further, the current rate of inflation and people's expectations aboutfuture rates of inflation may well seem to respond slowly to isolatedactions of restrictive monetary and fiscal policy that are viewed as tem-porary departures from what is perceived as a long-term governmentpolicy involving high average rates of government deficits and monetaryexpansion in the future. Thus inflation only seems to have a momentum ofits own; it is actually the long-term government policy of persistentlyrunning large deficits and creating money at high rates which imparts themomentum to the inflation rate. An implication of this view is thatinflation can be stopped much more quickly than advocates of the"momentum" view have indicated and that their estimates of the lengthof time and the costs of stopping inflation in terms of foregone output($220 billion of GNP for one percentage point in the inflation rate) areerroneous. This is not to say that it would be easy to eradicate inflation.On the contrary, it would require far more than a few temporary restric-tive fiscal and monetary actions. It would require a change in the policyregime: there must be an abrupt change in the continuing governmentpolicy, or strategy, for setting deficits now and in the future that issufficiently binding as to be widely believed. Economists do not nowpossess reliable, empirically tried and true models that can enable themto predict precisely how rapidly and with what disruption in terms of lostoutput and employment such a regime change will work its effects. Howcostly such a move would be in terms of foregone output and how long itwould be in taking effect would depend partly on how resolute andevident the government's commitment was.

This paper describes several dramatic historical experiences which Ibelieve to be consistent with the "rational expectations" view but whichseem difficult to reconcile with the "momentum" model of inflation. Theidea is to stand back from our current predicament and to examine themeasures that successfully brought drastic inflations under control in

43 The Ends of Four Big Inflations

several European countries in the 1920s. I shall describe and interpretevents in Austria, Hungary, Germany, and Poland, countries whichexperienced a dramatic "hyperinflation" in which, after the passage ofseveral months, price indexes assumed astronomical proportions. Thebasic data to be studied are the price indexes in figures 2.1-2.4. Thesedata are recorded in a logarithmic scale, so that they will fit on a page. Forall four countries, and especially Germany, the rise in the price level wasspectacular. The graphs also reveal that in each case inflation stoppedabruptly rather than gradually. I shall also briefly describe events inCzechoslovakia, a country surrounded by neighbors experiencing hyper-inflations, but which successfully achieved a stable currency itself. Myreason for studying these episodes is that they are laboratories for thestudy of regime changes. Within each of Austria, Hungary, Poland, andGermany, there occurred a dramatic change in the fiscal policy regimewhich in each instance was associated with the end of a hyperinflation.Further, though it shared some problems with its four neighbors,Czechoslovakia deliberately adopted a relatively restrictive fiscal policyregime, with the avowed aim of maintaining the value of its currency.

While there are many differences in details among the Austrian,Hungarian, Polish, and German hyperinflations, there are some veryimportant common features. These include the following:

i) The nature of the fiscal policy regime in effect during each of thehyperinflations. Each of the four countries persistently ran enormousbudget deficits on current account.

ii) The nature of the deliberate and drastic fiscal and monetary mea-sures taken to end the hyperinflations.

iii) The immediacy with which the price level and foreign exchangessuddenly stabilized.4

iv) The rapid rise in the "high-powered" money supply in the monthsand years after the rapid inflation had ended.

I shall assemble and interpret the facts in the light of a view about theforces which give money value and about the way the internationalmonetary system worked in the 1920s. Before interpreting the historicalfacts, I now turn to a brief description of this view.

2.2 The Gold Standard

After World War I, the United States was on the gold standard. TheUnited States government stood ready to convert a dollar into a specifiedamount of gold on demand. To understate things, immediately after thewar, Hungary, Austria, Poland, and Germany were not on the goldstandard. In practice, their currencies were largely "fiat," or unbacked.The governments of these countries resorted to the printing of new


1O5

104

WHOLESALE PRICES

AUSTRIA

DECEMBER DECEMBER DECEMBER DECEMBER

JUNE JUNE JUNE

1921-1924

Fig. 2.1 Wholesale prices in Austria.

unbacked money to finance government deficits.5 This was done on such ascale that it led to a depreciation of the currencies of spectacular propor-tions. In the end, the German mark stabilized at 1 trillion (1012) papermarks to the prewar gold mark, the Polish mark at 1.8 million papermarks to the gold zloty, the Austrian crown at 14,400 paper crowns to theprewar Austro-Hungarian crown, and the Hungarian krone at 14,500paper crowns to the prewar Austro-Hungarian crown.6

This paper focuses on the deliberate changes in policy that each ofHungary, Austria, Poland, and Germany made to end its hyperinflation,and the deliberate choice of policy that Czechoslovakia made to avoidinflation in the first place. The hyperinflations were each ended byrestoring or virtually restoring convertibility to the dollar or equivalentlyto gold. For this reason it is good to keep in mind the nature of therestrictions that adherence to the gold standard imposed on a govern-ment. Under the gold standard, a government issued demand notes and


10 V

10"

WHOLESALE PRICES

6 HUNGARY

104

103JUNE JUNE JUNE JUNE

DECEMBER DECEMBER DECEMBER

1921-1924

Fig. 2.2 Wholesale prices in Hungary.

longer-term debt which it promised to convert into gold under certainspecified conditions, i.e. on demand, for notes. Presumably, people werewilling to hold these claims at full value if the government's promise topay were judged to be good. The government's promise to pay was"backed" only partially by its holding of gold reserves. More important inpractice, since usually a government did not hold 100% reserves of gold,a government's notes and debts were backed by the commitment of thegovernment to levy taxes in sufficient amounts, given its expenditures, tomake good on its debt. In effect, the notes were backed by the govern-ment's pursuit of an appropriate budget policy. During the 1920s, JohnMaynard Keynes emphasized that the size of a government's gold reservewas not the determinant of whether it could successfully maintain con-vertibility with gold: its fiscal policy was.7 According to this view, whatmattered was not the current government deficit but the present value ofcurrent and prospective future government deficits. The government was


10V

107

104

WHOLESALE PRICES

POLAND

JUNE JUNE JUNE JUNEDECEMBER DECEMBER DECEMBER

1921-1924

Fig. 2.3 Wholesale prices in Poland.

like a firm whose prospective receipts were its future tax collections. Thevalue of the government's debt was, to a first approximation, equal to thepresent value of current and future government surpluses. So under agold standard, a government must honor its debts and could not engage ininflationary finance. In order to assign a value to the government's debt,it was necessary to have a view about the fiscal policy regime in effect,that is, the rule determining the government deficit as a function of thestate of the economy now and in the future. The public's perception of thefiscal regime influenced the value of government debt through privateagents' expectations about the present value of the revenue streamsbacking that debt.8 It will be worthwhile to keep this view of the goldstandard in mind as we turn to review the events surrounding the ends ofthe four hyperinflations.9

However, it will be useful first to expand a little more generally on thedistinction between the effects of isolated actions taken within the context


1015

10PRICE LEVEL

GERMANY

I09

I06

DECEMBER DECEMBER DECEMBERDECEMBER DECEMBER DECEMBER

1919-1924

Fig. 2.4 Wholesale prices in Germany.

of a given general strategy, on the one hand, and the effects of choosingamong alternative general strategies or rules for repeatedly taking ac-tions, on the other. The latter choice I refer to as a choice of regime. Thevalues of government expenditures and tax rates for one particular quar-ter are examples of actions, while the rules, implicit or explicit, forrepeatedly selecting government expenditures and tax rates as functionsof the state of the economy are examples of regimes. Recent work indynamic macroeconomics has discovered the following general principle:whenever there is a change in the government strategy or regime, privateeconomic agents can be expected to change their strategies or rules forchoosing consumption rates, investment rates, portfolios, and so on.10

The reason is that private agents' behavior is selfish, or at least purpose-ful, so that when the government switches its strategy, private agentsusually find it in their best interests to change theirs. One by-product ofthis principle is that most of the empirical relations captured in standard


econometric models cannot be expected to remain constant across con-templated changes in government policy regimes. For this reason, predic-tions made under the assumption that such relations will remain constantacross regime changes ought not to be believed. The estimate that a 1%reduction in inflation would cost $220 billion GNP annually is one exampleof such a faulty prediction. When an important change in regime occurs,dynamic macroeconomics would predict that the entire pattern of cor-relations among variables will change in quantitatively important ways.

While the distinction between isolated actions and strategy regimes isclear in principle, it is an admittedly delicate task to interpret whetherparticular historical episodes reflect isolated actions within the same oldrules of the game or whether they reflect a new set of rules or governmentstrategies.11 All that we have to go on are the recorded actions actuallytaken, together with the pronouncements of public officials, laws, legisla-tive votes, and sometimes constitutional provisions. Out of this materialwe are to fashion a view about the government strategy being used.Common sense suggests and technical econometric considerationsconfirm the difficulties in making such interpretations in general. Havingsaid this, I believe that the examples discussed below are about as close tobeing laboratories for studying regime changes as history has provided.

2.3 Austria

At the end of World War I, the Austro-Hungarian empire dissolvedinto a number of successor states, of which Austria was one. From havingbeen the center of an empire of 625,000 square kilometers and 50 millioninhabitants, Austria was reduced to a mere 80,000 square kilometers and6.5 million inhabitants. Having suffered food scarcities during the warthat were produced by an effective Allied blockade, Austria found itselfconfronted with new national borders and trade barriers that cut it offfrom the food sources formerly within its empire. Further, the govern-ment of Austria reabsorbed a large number of Austrian imperial bureau-crats who were no longer welcome in the other successor states. Aus-trians also faced a large-scale unemployment problem stemming from theneed to reconvert the economy to peaceful activities and to adjust to thenew national borders. If this were not enough, as a loser of the warAustria owed the Reparation Commission sums that for a long time wereuncertain in amount but were presumed eventually to be substantial. TheReparation Commission, in effect, held a blanket mortgage against theassets and revenues of the Austrian government.

Austria responded to these pressing problems by making large expend-itures in the form of food relief and payments to the unemployed. Inaddition, the state railroads and monopolies ran deficits, as taxes andprices were kept relatively low. The government did not collect enough


taxes to cover expenditures and so ran very substantial deficits during theyears 1919-22 (see table Al). As table Al shows, in these years the deficitwas typically over 50 percent of the total government expenditures. Thegovernment financed these deficits by selling treasury bills to the Austriansection of the Austro-Hungarian bank. The result was a very rapidincrease in the volume of "high-powered" money, defined as the notesand demand deposit obligations of the central bank (see table A2). As thefigures in table A2 indicate, between March 1919 and August 1922 thetotal note circulation in Austria12 of the Austro-Hungarian bank in-creased by a factor of 288. This expansion of central bank notes stemmedmainly from the bank's policy of discounting treasury bills. However, italso resulted partly from the central bank's practice of making loans anddiscounts to private agents at nominal interest rates of between 6 and 9%per annum, rates which by any standard were far too low in view of theinflation rate, which averaged 10,000% per annum from January 1921 toAugust 1922 (table A3).13

In response to these government actions and what seemed like pros-pects for their indefinite continuation, the Austrian crown depreciatedinternationally and domestic prices rose rapidly (see tables A3 and A4).While between January 1921 and August 1922 the note circulation of thecentral bank increased by a factor of 39, the retail price index increasedby a factor of 110 (see tables A3 and A4) so that the real value of the notecirculation diminished during the currency depreciation.14 The "flightfrom the crown" occurred as people chose to hold less of their wealth inthe form of the rapidly depreciating crown, attempting instead to holdforeign currencies or real assets.15 From the viewpoint of financing itsdeficit, the government of Austria had an interest in resisting the flightfrom the crown, because this had the effect of diminishing the resourcesthat the government could command by printing money. Therefore thegovernment established a system of exchange controls administered byan agency called the Devisenzentrale. The essential function of this

Table Al Austrian Budgets, 1919-22 (in millions of paper crowns)

1 January-30 June 19191 July 1919-30 June 19201 July 1920-30 June 19211 January-31 December 1922

Receipts

1,3396,295

29,483209,763

Expen-ditures

4,04316,87370,601

347,533

Deficit

2,70410,57841,118

137,770

Percentage ofExpendituresCovered byNew Issues ofPaper Money

67635840

Source: Pasvolsky [25, p. 102].


Table A2

1919

1920

1921

1922

JanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecemberJanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecemberJanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberJanuaryFebruaryMarchApril

Total Note Circulation of Austrian(in thousands of crowns)

—4,687,0565,577,8515,960,0037,397,6928,391,4059,241,1359,781,112

10,819,31011,193,67012,134,47413,266,87814,292,80915,457,74915,523,83215,793,80516,971,34418,721,49520,050,28122,271,68625,120,38528,072,33130,645,65834,525,63438,352,64841,067,29945,036,72345,583,19449,685,14054,107,28158,533,76670,170,798

227,015,925259,931,138304,063,642346,697,776

1923

1924

1925

Crowns

MayJuneJulyAugustSeptemberOctoberNovemberDecemberJanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecemberJanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecemberJanuaryFebruaryMarchApril

397,829,313549,915,678786,225,601

1,353,403,6322,277,677,7382,970,916,6073,417,786,4984,080,177,2384,110,551,1634,207,991,7224,459,117,2164,577,382,3334,837,042,0815,432,619,3125,684,133,7215,894,786,3676,225,109,3526,607,839,1056,577,616,3417,125,755,1906,735,109,0007,364,441,0007,144,901,0007,135,471,0007,552,620,0007,774,958,0007,995,647,0005,894,786,3677,998,509,0008,213,003,0008,072,021,0008,387,767,0007,902,217,0007,957,242,0007,897,792,0007,976,420,000

Source: Young [36, vol. 2, p. 292].

agency was to increase the amount of Austrian crowns held by Austrians,which it accomplished by adopting measures making it difficult or illegalfor Austrians to hold foreign currencies and other substitutes for Aus-trian crowns.16 Despite these regulations, it is certain that Austriancitizens were holding large amounts of foreign currencies during 1921 and1922.

Table A4 reveals that the Austrian crown abruptly stabilized in August1922, while table A3 indicates that prices abruptly stabilized a month


Table A3 Austrian Retail Prices, 1921-24

Retail PriceIndex, 52Commodities

1921

1922

1923

1924

JanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecemberJanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecemberJanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecemberJanuaryFebruaryMarchAprilMayJune

100114122116121150143167215333566942

1,1421,4281,4571,6192,0283,4314,830

11,04620,09018,56717,68117,40917,52617,85118,20519,42820,45020,48219,36818,51120,95521,16621,47921,84922,94123,33623,33623,36123,79724,267



Table A4

JanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecember

Exchange Rates, Austrian Crownsper United States Dollar, in

1919

17.0920.7225.8526.0324.7529.6337.2442.5068.5099.50

130.00155.00

1920

271.43250.00206.66200.00155.83145.00165.00237.14255.00358.33493.66659.40

1921

654.00722.50676.00661.00604.00720.00957.00

1,081.502,520.004,355.008,520.005,275.00

New York Market

1922

7,375.006,350.007,487.507,937.50

11,100.0018,900.0042,350.0077,300.0074,210.0073,550.0071,400.0070,925.00

1923

71,500.0071,150.0071,000.0070,850.0070,800.0070,800.0070,760.0070,760.0070,760.0070,760.0070,760.0070,760.00

1924

70,760.0070,760.0070,760.0070,760.0070,760.0070,760.0070,760.0070,760.0070,760.0070,760.0070,760.0070,760.00


later. This occurred despite the fact that the central bank's note circula-tion continued to increase rapidly, as table Al indicates. Furthermore,there occurred no change in currency units or "currency reform," at leastnot for another year and a half.

The depreciation of the Austrian crown was suddenly stopped by theintervention of the Council of the League of Nations and the resultingbinding commitment of the government of Austria to reorder Austrianfiscal and monetary strategies dramatically. After Austria's increasinglydesperate pleas to the Allied governments for international aid hadrepeatedly been rejected or only partially fulfilled, in late August 1922the Council of the League of Nations undertook to enter into seriousnegotiations to reconstruct the financial system of Austria. These nego-tiations led to the signing of three protocols on 2 October 1922 whichsuccessfully guided the financial reconstruction of Austria. It is remark-able that even before the precise details of the protocols were publiclyannounced, the fact of the serious deliberations of the Council broughtrelief to the situation. This can be seen in tables A3 and A4, and wasdescribed by Pasvolsky as follows:

The moment the Council of the League decided to take up in earnestthe question of Austrian reconstruction, there was immediately awidespread conviction that the solution of the problem was at hand.This conviction communicated itself first of all to that delicately ad-justed mechanism, the international exchange market. Nearly twoweeks before Chancellor Seipel officially laid the Austrian questionbefore the Council of the League, on August 25, the foreign exchangerate ceased to soar and began to decline, the internal price levelfollowing suit three weeks later. The printing presses in Austria were


still grinding out new currency; the various ministries were still dispers-ing this new currency through the country by means of continuingbudgetary deficits. Yet the rate of exchange was slowly declining. Thecrisis was checked.17

The first protocol was a declaration signed by Great Britain, France,Italy, Czechoslovakia, and Austria that reaffirmed the political inde-pendence and sovereignty of Austria.18 The second protocol providedconditions for an international loan of 650 million gold crowns to Austria.The third protocol was signed by Austria alone and laid out a plan forreconstruction of its fiscal and monetary affairs. The Austrian govern-ment promised to establish a new independent central bank, to ceaserunning large deficits, and to bind itself not to finance deficits withadvances of notes from the central bank. Further, the government ofAustria agreed to accept in Austria a commissioner general, appointed bythe Council of the League, who was to be responsible for monitoring thefulfillment of Austria's commitments. The government of Austria alsoagreed to furnish security to back the reconstruction loan. At the sametime, it was understood that the Reparation Commission would give upor modify its claim on the resources of the government of Austria.

The government of Austria and the League both moved swiftly toexecute the plan outlined in the protocols. In legislation of 14 November1922, the Austrian National Bank was formed to replace the old Austriansection of the Austro-Hungarian bank; it was to take over the assets andfunctions of the Devisenzentrale. The new bank began operations on 1January 1923 and was specifically forbidden from lending to the govern-ment except on the security of an equal amount of gold and foreign assets.The bank was also required to cover its note issues with certain minimalproportions of gold, foreign earning assets, and commercial bills. Fur-ther, once the government's debt to the bank had been reduced to 30million gold crowns, the bank was obligated to resume convertibility intogold.

The government moved to balance its budget by taking concrete stepsin several directions. Expenditures were reduced by dischargingthousands of government employees. Under the reconstruction scheme,the government promised gradually to discharge a total of 100,000 stateemployees. Deficits in government enterprises were reduced by raisingprices of government-sold goods and services. New taxes and moreefficient means of collecting tax and custom revenues were instituted.The results of these measures can be seen by comparing the figures intable A5 with those in table Al. Within two years the government wasable to balance the budget.

The stabilization of the Austrian crown was not achieved via a currencyreform. At the end of 1924 a new unit of currency was introduced, theschilling, equal to 10,000 paper crowns. The introduction of this new unit


Table A5 The Austrian Budget, 1923-25 (in millions of schillings)

Closed Accounts

Item

Total revenueCurrent expendituresDeficit ( - ) or surplus (+)Capital expendituresTotal balance

1923

697.4779.6

-82.276.0

-158.2

1924

900.6810.0

+ 90.6103.6

-13.0

1925

908.5741.4

+ 167.190.6

+ 76.5


Note: 1 schilling = 10,000 paper crowns.

of currency occurred long after the exchange rate had been stabilized andwas surely an incidental measure.19

Table A2 reveals that from August 1922, when the exchange ratesuddenly stabilized, to December 1924, the circulating notes of theAustrian central bank increased by a factor of over 6. The phenomenonof the achievement of price stability in the face of a sixfold increase in thestock of "high-powered" money was widely regarded by contemporariesas violating the quantity theory of money, and so it seems to do. How-ever, these observations are not at all paradoxical when interpreted in thelight of a view which distinguishes sharply between unbacked, or "out-side," money, on the one hand, and backed, or "inside," money, on theother hand. In particular, the balance sheet of the central bank and thenature of its open market operations changed dramatically after thecarrying out of the League's protocols, with the consequence that theproper intrepretation of the figures on the total note obligations of thecentral bank changes substantially. Before the protocols, the liabilities ofthe central bank were backed mainly by government treasury bills; thatis, they were not backed at all, since treasury bills signified no commit-ment to raise revenues through future tax collections. After the executionof the protocols, the liabilities of the central bank became backed bygold, foreign assets, and commercial paper, and ultimately by the powerof the government to collect taxes. At the margin, central bank liabilitieswere backed 100% by gold, foreign assets, and commercial paper asnotes and the deposits were created through open market operations inthose assets (see table A6). The value of the crown was backed by thecommitment of the government to run a fiscal policy compatible withmaintaining the convertibility of its liabilities into dollars. Given such afiscal regime, to a first approximation, the intermediating activities of thecentral bank did not affect the value of the crown so long as the assetspurchased by the bank were sufficiently valuable. Thus the sixfold in-crease in the liabilities of the central bank after the protocols ought not tobe regarded as inflationary. The willingness of Austrians to convert


Table A6

1923:


1924:


1925:

JanuaryFebruaryMarchApril

Austrian National Bank Balance Sheet(end of month,

Gold

49,30483,43886,09773,27073,39173,39173,39173,39173,39162,11762,11783,177

91,274105,536106,663107,059107,443107,762108,342108,256108,950109,327110,643110,890

111,314111,474111,649112,168

ForeignExchangeandCurrency

1,058,2441,029,1341,336,3851,439,9991,682,2092,532,3162,947,2163,050,0853,126,5993,356,2323,504,6523,832,132

3,811,1483,921,5943,953,8723,669,3333,344,3373,178,3393,254,4773,453,1773,724,9164,032,4854,312,3554,770,548

3,337,9113,310,0323,202,8023,474,672

in millions of crowns)

LoansandDiscounts

731,046728,884821,397741,858875,942730,848658,966647,936863,317

1,069,3401,094,6201,325,380

1,253,1101,737,3341,733,4002,131,9842,660,4493,092,4703,304,8763,226,9622,852,6882,379,7001,945,6271,881,593

1,545,2951,285,1581,047,7191,059,069

TreasuryBills

2,556,8482,552,6822,550,1592,550,1592,550,1592,547,2122,539,7772,538,7192,537,6612,536,6042,535,5472,534,490

2,533,4342,532,3792,295,4282,294,4712,259,8392,237,7942,231,1732,219,4592,210,5272,202,1062,196,1812,178,185

2,172,4912,150,1512,107,9492,088,777

Notes inCirculation

4,110,5514,207,9924,459,1174,577,3824,837,0425,432,6195,684,1345,894,7866,225,1096,607,8396,577,6167,125,755

6,735,1097,364,4417,144,9017,315,4717,554,6207,774,9587,995,6478,002,1427,998,5098,213,0038,072,0218,387,767

7,902,2177,957,2427,897,7927,976,420

Deposits

279,092178,752329,109226,273343,339362,237535,121413,383373,673414,882617,321649,424

536,982558,800752,814696,141641,001741,400896,032997,677890,537502,579484,750533,450

438,390315,771295,498236,957


hoards of foreign exchange into crowns which is reflected in table A6 isnot surprising since the stabilization of the crown made it a much moredesirable asset to hold relative to foreign exchange.20

The available figures on unemployment indicate that the stabilizationof the crown was attended by a substantial increase in the unemploymentrate, though unemployment had begun to climb well before stabilizationwas achieved (see table A7). The number of recipients of state unemploy-


Table A7 Number of Austrian Unemployedin Receipt of Relief (in thousands)

Beginning of 1922

January 17April 42July 33October 38

1923

1171539379

1924

981076478

1925

154176118119

1926

208202151148

Source: League of Nations [14, p. 87].

ment benefits gradually climbed from a low of 8,700 in December 1921 to83,000 in December 1922. It climbed to 167,000 by March 1923, and thenreceded to 76,000 in November 1923.21 How much of this unemploymentwas due to the achievement of currency stabilization and how much wasdue to the real dislocations affecting the Austrian economy cannot bedetermined. However, it is true that currency stabilization was achievedin Austria very suddenly, and with a cost in increased unemployment andforegone output that was minor compared with the $220 billion GNP thatsome current analysts estimate would be lost in the United States per onepercentage point inflation reduction.

2.4 Hungary

Like its old partner in the Hapsburg monarchy, Hungary emerged fromWorld War I as a country much reduced in land, population, and power.It retained only 25% of its territory (down from 325,000 square kilome-ters to 92,000) and only 37% of its population (down from 21 million toabout 8 million). Its financial and economic life was disrupted by thenewly drawn national borders separating it from peoples and economicinstitutions formerly within the domain of the Hapsburg monarchy.

At the end of the war, Hungary experienced political turmoil as theHapsburg King Charles was replaced by the government of PrinceKarolyi. In March 1919, the Karolyi government was overthrown by theBolsheviks under Bela Kun. The regime of Bela Kun lasted only fourmonths, as Romania invaded Hungary, occupied it for a few weeks, andthen withdrew. A new repressive right wing regime under AdmiralHorthy then took power. The "white terror" against leftists carried outby supporters of Horthy took even more lives than the "red terror" thathad occurred under Bela Kun.

At the end of the war, the currency of Hungary consisted of the notes ofthe Austro-Hungarian bank. By the provisions of the peace treaties ofTrianon and St. Germain, the successor states to the Austro-Hungarianempire were required to stamp the notes of the Austro-Hungarian bankthat were held by their residents, in effect, thereby recognizing those


notes as debts of the respective new states. Before Hungary executed thisprovision of the Treaty of Trianon, the currency situation grew morecomplicated, for the Bolshevik regime had access to the plates for print-ing one- and two-crown Austro-Hungarian bank notes and it used themto print more notes. The Bolshevik government also issued new so-calledwhite notes. Each of these Bolshevik-issued currencies was honored bythe subsequent government.

The Austro-Hungarian bank was liquidated at the end of 1919, and itwas replaced by an Austrian section and a Hungarian section. Thefunctions of the Hungarian section of the old bank were assumed inAugust 1921 by a State Note Institute, which was under the control of theminister of finance. In August 1921, the Note Institute issued its ownnotes, the Hungarian krone, in exchange for Hungarian stamped notes ofthe Austro-Hungarian bank and several other classes of notes, includingthose that had been issued by the Bolshevik regime.

As a loser of the war, Hungary owed reparations according to theTreaty of Trianon. The Reparation Commission had a lien on the re-sources of the government of Hungary. However, neither the totalamount owed nor a schedule of payments was fixed for many years afterthe war. This circumstance alone created serious obstacles in terms ofachieving a stable value for Hungary's currency and other debts, since theunclear reparations obligations made uncertain the nature of the re-sources which backed those debts.

From 1919 until 1924 the government of Hungary ran substantialbudget deficits. The government's budget estimates in table HI arereported by Pasvolsky substantially to understate the size of the deficits.22

These deficits were financed by borrowing from the State Note Institute,and were a major cause of a rapid increase in the note and depositliabilities of the institute. An additional cause of the increase in liabilitiesof the institute was the increasing volume of loans and discounts that itmade to private agents (see table H2). These loans were made at a very

Table HI

1920-211921-221922-231923-24

Hungarian Budget Estimates,(in millions of paper crowns)

Revenue

10,52020,296

152,8022,168,140

Expen-ditures

20,21026,764

193,4553,307,099

1920-24

Deficit

9,6906,468

40,6531,138,959

Percentage ofExpendituresCovered byIssues ofPaper Money

47.924.121.034.4


09

U

U <

t/j -_* r*

£ 3 oo .is «Z U iS

< 2

iS5 S

nc• I 3 5 «

| 1K u

I I2 c ^2 pa

in vcT irf T-T c-T (NT

H O • * « *

8

g g3 b t - ' C s ^ y > , o o - t ; 3 > o § 2 | a ^

X)

3 «

3° a

H rt (S N

o

t i o n © oa s i n m o o - ^ - 0 0

r oN PI

t-- ocT vo r im i n r

m in oo o ^H ^H<N • * in vo" o ooo in in in m m

M M M (N M N NOs 00 Osvo" ^H" ON*

in in in in

uX)ooO

6uoZ

6<uu<uQ

rn

x>(U 0>

o O u § § S UVH X) X)

2 *g § |3 U "2 <U

O < - "

3 3 §o .b TSZU «

o -

O C/3

S « 3

I S Q S

11£ U

930

co"ON

163

COr-°\

809

ONVO°i.

987

oo"vOON*

Uu

n

CNTD

c(Ua.o

c

pres

•™*DT3caEo

r o O

CN CN CN CN

§ 2C JOl!

oo

f I

00-^

c co _QJ O

o c c

.2 "7? «C o ooa °° a

i Ifc «23 •t -5


low interest rate, in view of the rapid rate of price appreciation, and to alarge extent amounted to simple gifts from the Note Institute to thoselucky enough to receive loans on such generous terms. These privateloans account for a much larger increase in high-powered money in theHungarian than in the other three hyperinflations we shall study.

As table H3 shows, the Hungarian krone depreciated rapidly on for-eign exchange markets, and domestic prices rose rapidly. Between Janu-ary 1922 and April 1924, the price index increased by a factor of 263. Inthe same period, the total notes and deposit liabilities of the NoteInstitute increased by a factor of 85, so that the real value of its liabilitiesdecreased substantially. As in the case of Austria, this decrease wassymptomatic of a "flight from the krone," as residents of Hungaryattempted to economize on their holdings of krones and instead to holdassets denominated in more stable currencies. As in the case of Austria,the government of Hungary resisted this trend by establishing in August1922 a Hungarian Devisenzentrale within the State Note Institute.

Table H3 indicates that in March 1924, the rise in prices and thedepreciation of the krone internationally both abruptly halted. The stabi-lization occurred in the face of continued expansion in the liabilities of thecentral bank, which increased by a factor of 3.15 between March 1924 andJanuary 1925 (see table H2). This pattern parallels what occurred inAustria and has a similar explanation.

As in Austria, the financial reconstruction of Hungary was accom-plished with the intervention of the League of Nations. Together with theReparation Commission and the government of Hungary, the Leaguedevised a plan which reduced and clarified the reparations commitmentof Hungary, arranged for an international loan that would help financegovernment expenditures, and committed Hungary to establish a bal-anced budget and a central bank legally bound to refuse any governmentdemand for unbacked credit. On 21 February 1924, the ReparationCommission agreed to give up its lien on Hungary's resources so thatthese could be used to secure a reconstruction loan. A variety of Westernnations also agreed to give up their liens on Hungary so that the new loancould successfully be floated.

The League's reconstruction plan was embodied in two protocols. Thefirst was signed by Great Britain, France, Italy, Czechoslovakia, Ru-mania, and Hungary, and guaranteed the "political independence, terri-torial integrity, and sovereignty of Hungary." The second protocol out-lined the terms of the reconstruction plan, and committed Hungary tobalance its budget and form a central bank truly independent of theFinance Ministry. The government was also obligated to accept in Hun-gary a commissioner general, responsible to the League, to monitor andsupervise the government's fulfillment of its commitment to fiscal andmonetary reform.


Table H3 Hungarian Price and Exchange Rate

HungarianIndexof Prices1

Centsper Crownin New York

1921:

JulyAugustSeptemberOctoberNovemberDecember

1922:


1923:


1924:

JanuaryFebruaryMarchAprilMayJuneJulyAugustSeptember

4,2005,4006,2506,7508,3008,250

8,1008,5009,900

10,75011,00012,90017,40021,40026,60032,90032,60033,400

38,50041,80066,00083,50094,000

144,500286,000462,500554,000587,000635,000714,000

1,026,0001,839,1002,076,7002,134,6002,269,6002,207,8002,294,5002,242,0002,236,600

0.3323.2629.1944.1432.1078.1512

.1525

.1497

.1256

.1258

.1261

.1079

.0760

.0595

.0423

.0402

.0413

.0430

.0392

.0395

.0289

.0217

.0191

.0140

.0097

.0056

.0055

.0054

.0054

.0052

.0039

.0033

.0015

.0014

.0012

.0011

.0012

.0013

.0013


Table H3 (continued)

OctoberNovemberDecember

1925:

JanuaryFebruaryMarch

HungarianIndexof Prices1

2,285,2002,309,5002,346,600

2,307,5002,218,7002,117,800

Centsper Crownin New York

.0013

.0013

.0013

.0014

.0014

.0014


'From July 1921 through November 1923, the index numbers represent retail prices and arebased on 60 commodities with July 1914 = 100. From December 1923 through March 1925,the figures are based on wholesale prices computed by the Hungarian Central StatisticalOffice. They refer to the prices of 52 commodities on the last day of the month with 1913 =100.

A reconstruction loan of 250 million gold krones was successfullyplaced abroad in July 1924. The loan was secured by receipts fromcustoms duties and sugar taxes, and revenues from the salt and tobaccomonopolies. The purpose of the loan was to give the government aconcrete means of converting future promises to tax into current re-sources while avoiding the need to place its debt domestically.

By a law of 26 April 1924, the Hungarian National Bank was estab-lished, and it began operations on 24 June. The bank assumed the assetsand liabilities of the State Note Institute and took over the functions ofthe foreign exchange control office, the Devisenzentrale. The bank wasprohibited from making any additional loans or advances to the govern-ment, except upon full security of gold or foreign bills. The bank was alsorequired to hold gold reserves of certain specified percentages behind itsliabilities.

The government of Hungary also tried to establish a balanced budget.Both by cutting expenditures and raising tax collections, the governmentwas successful in moving quickly to a balanced budget (see table H4).Indeed, the proceeds of the reconstruction loan were used perceptiblymore slowly than had been anticipated in the reconstruction plan.

As table H2 confirms, the stabilization of the krone was accompaniedby a substantial increase in the total liabilities of the central bank. But aswith Austria, the drastic shift in the fiscal policy regime that occasionedthe stabilization also changed the appropriate interpretation of thesefigures. As table H2 indicates and as the regulations governing the bankrequired, after the League's intervention the note and deposit liabilities


Table H4

Period

Jul.-Dec. 1924Jan.-Jun. 1925

Fiscal year 1924-25

Hungarian Budget, 1924-25 (in millions of crowns)

Preliminary TreasuryAccounts

Re- Expen-ceipts ditures

208.0 205.9245.1 216.9

453.1 422.8

Surplus( + )orDeficit( - )

+ 2.1+ 28.2

+ 30.3

Reconstruction

Re-ceipts

143.8150.0

293.8

Scheme

Expen-ditures

186.3207.6

393.9

Surplus( + )orDeficit( - )

-42.5-57.6

-100.1


of the central bank became backed, 100% at the margin, by holdings ofgold, foreign exchange, and commercial paper. In effect, the centralbank's liabilities represented "fiat money" before the League's plan wasin effect; after that plan was in effect, they represented more or lessbacked claims on British sterling,23 the foreign currency to which Hungarypegged its exchange as a condition for British participation in the recon-struction loan.

Figures on unemployment in Hungary are reported in table H5, andunfortunately begin only immediately after the price stabilization hadalready occurred. All that can be inferred from these figures is thatimmediately after the stabilization, unemployment was not any higherthan it was one or two years later. This is consistent either with thehypothesis that the stabilization process had little adverse effect on

Table H5 Number of Unemployed in Hungary (figures relate only to membersof Union of Socialist Workers, in thousands of workers)

End of 1924 1925 1926


———222325313020303133

373737363034322725232627

282929262826

Source: League of Nations [15, p. 50].


unemployment or with the hypothesis that the adverse effect was solong-lasting that no recovery occurred within the time span of the figuresrecorded. The former hypothesis seems more plausible to me.

2.5 Poland

The new nation of Poland came into existence at the end of World WarI, and was formed from territories formerly belonging to Germany,Austro-Hungary, and Russia. At the time of its formation, Poland pos-sessed a varied currency consisting of Russian rubles, crowns of theAustro-Hungarian bank, German marks, and Polish marks issued by thePolish State Loan Bank, which had been established by Germany tocontrol the currency in the part of Poland occupied by Germany duringthe war. For Poland, the armistice of 1918 did not bring peace, a costlywar with Soviet Russia being waged until the fall of 1920. Poland wasdevastated by the fighting and by Germany's practice of stripping it of itsmachinery and materials during World War I.24

The new government of Poland ran very large deficits up to 1924 (seetable PI). These deficits were financed by government borrowing fromthe Polish State Loan Bank, which the new government had taken overfrom the Germans. From January 1922 to December 1923, the outstand-ing notes of the Polish State Loan Bureau increased by a factor of 523(table P2). Over the same period, the price index increased by a factor of

Table PI Polish Receipts and Expenditures (in thousands of zloty)

Receipts:

AdministrationState EnterprisesMonopolies

Total

Expenditures:

AdministrationState EnterprisesMonopolies

TotalDeficitSurplus

1921

261,67611,41372,222

345,311

765,263115,589

880,852535,541

1922

467,97914,55647,893

530,428

734,310145,003

879,313348,885

1923

—

426,000

—

1,119,800692,000

1924

—

1,703,000

—

1,629,000

74,000

1925

1,491,743133,530356,611

1,981,884

1,830,231106,34345,019

1,981,593

251

Source: Young [36, vol. 2, p. 183].Note: Conversion from marks to zloty was made on the following basis: 1921, 1 zloty =303.75 marks. First quarter 1922, 1 zloty = 513.52 marks; second quarter, 691.49 marks;third quarter, 1,024.97 marks; and fourth quarter, 1,933.87 marks.

CQ

S oO <U

U E

on

soEo

T3e

wV)

»

BaI-

o

M

J2 -Cc3 .•£

CQ ^

'do w "o

is-

)iir

Op t«

'§ "cO K

(JH ff

JO E

o ^ o o o o o o o o p o o

•—i en Tt ON

o

oD.

i- JO JOS 6 ft JO JO

a o <u u

,079

,560

,087

co

,755

oo

,575

co

,697

co

,242

CN

,734

r—1

,792

co

,777

o

,029

VD

,537

o

,275

©

,775

©

,775

o

,375

©

,375

o

,625

o

,375

o

,625

o,6

25©

,925

o

,625

©

,625

©539

o o o o o o o o o o o ou - i i r i i / S i o i / ^ i r i i o o Q O O OC N C N C N C N « N ( N C N © O © © ©vo *~~< vo VD VO VO *O ^D ^J in ^ în r-" co" ^D" t--" o" ©" oo" oo" oo" TJ-" r-T

ON

O

IO ON VO CO

T-H vd 00 CNON r- c-- CM

rf

r* T-* CN en r- ON CN in ON

h ^ N » N Tf o \ oo o \ r - i © v q v q

T — I T j - V O C N V O C O > r ) O N r H < — I

in in in in -<t © coVO 00 00 vo ON CO f-IT) vo vo in oo T-H CN

in vo oo © vo •© i^ © r- co ONCM •* ON 00 IT) rj-

I T I O N O N O N O N O N O N O O CN CN 00 CO T-H in © in ON in ONON ON ON © ©

CM CN CO CO CO •*

<D

!|ltB-8» E S E E E

<U 1) <U> o u*Hll Ml Q s X

UsZ U 5

O wU E

m

00 c»

•g -aO «*

CO

615

239209

247665

250

oo

553

093

i—i

001

276

i—i

101

300

so

426

see

787

385

o

706

463972

579092

661437

793

CO

160

909,

oo

300

sO

205

,841

00

396

,332794

,733

649

,566

o

709

,478

m

776

,871

so r-~ co co ON ON SO 00t~- co r- r- _O (N m (N ONs

SO

O N

O

,555

i—i

,451

00oo00so

o

,555

00

,629

(N SO^H SOSO sOsO CO

O

,093

ON

00

r-

i—i

,320

oo,400

T—I

,950

,610

oo

,657

,725

t--

,440

IT)

,339

oo ON

Oin

• ^ ri ON O CO SO00 Tt

f~- ON ON ON ON O

ON ts l l l f l l l l l• • llllllflllll CN)ON

3 23 3 5

!-?> I-T, %

oo <N so coCO

,197

, ,402

080'

COCN

,494

,217

,

CO

,955

,371

,12

5,

,830

,659

,

CO

00

CO"

Os835

,205

,244

,59

6,.5

50.6

97.

570.

oo - j- co r-~CN co co cot-H_ -^ ©_ ©_r-" ©" t--" oo"c-- TJ- so oo©^ m •*_ in^

CN" co" oo" o"CN

00 Os Os Tf

©^ r^ t- t-t-»" o" os' oTm T-H os os00 IT) Tf Tf

t H io vqO Os Os Os

CN CN co Os

o o o o

GO 1-4 1-4 1-4CO Os Os OsCN <N (N (N

SO

,019

.

©

t--

,712

,

soCO00_

Os

,781

.

_în

co"

oo

,396

.

mooo"

CN

,445

.

T-H

oo

in

in

,440

.

OsCN00^co"00

in

,988

CO

00

ooo©Os00in

s

CN Os rf soO 00 CO >n00 CN Os OsSO so Os 00H n tjtfOs CN so CN

in oo © in00 CN t-H CNO . t-j_ CN Os^

co" so" oo" o"co CN in Ttin so so co

00 00 CO t-H

vd ooso so

3 S H -C

$8Z&

m o o ©

mo

5 •* •* © oCO Os CO som m •* •*

VO t-H co © © CNo!

©oo

r-Os

,826

OsSO

,236

SO

,224

in00

,230

rj-CO

,374

SO

f-co ,8

97in oo oo os CN CN co

s O C O O s s o r - ^ t O s C O© • * s o c o r - ~ - o o o s mO s C N ^ O t - H O s O s O Oc N c N c o © o o - < a - © T j -c N ^ O t - H T - H o o i n ' O i ni n o o r - r - r - o i n o s

so so 00 CO CN 00 Tfco oo Os oo vo co r--co co oo >n in t~- Os

s

N N CM

o t in

,963

,684

,392

CO

oo

©CO

,683

,

COOs

,324

,288

,

ooOs

,015

,900

,

OsOs

in in in in

CN •<* OI t-H in co

CN^ CO^ so^

CN" CO" r»"

oo" in"

Os © 00 •<*•CN CO CO •*00 Os so Os

CO Os CN (No) CN so co•* <N m vo

(N CN CO (N

853'

«n

242,

,320

r~-ro

206,

:,9

02

CNOsCO

259,

:.6

2121

6.

,223

521,

SO

,828

951,

CN

869'

,543

,

CN

,749

,658

,

Si

t-H r--r^ osoo m^t-n" oo"

oo mr-" oo"CN CN

§

,358

.

00CN

ON CN Os 00Tf CO t-H CNCN O so Tt

>•. a g 2 H'c

CN «J

—: c

o S3 **- C

•• CS RS


2,402 while the dollar exchange rate decreased by a factor of 1,397 (seetables P3 and P4). As in the other inflations we have studied, the realvalue of the note circulation decreased as people engaged in a "flightfrom the mark." Extensive government exchange controls were imposedto resist this trend.

Tables P2 and P3 indicate that the rapid inflation and exchange depre-ciation both suddenly stopped in January 1924. Unlike the cases ofAustria and Hungary, in Poland the initial stabilization was achievedwithout foreign loans or intervention, although later in 1927, after cur-rency depreciation threatened to renew, a substantial foreign loan wasarranged.25 But in terms of the substantial fiscal and monetary regimechanges that accompanied the end of the inflation, there is much similar-ity to the Austrian and Hungarian experiences. The two interrelated

Table

Year

1921

1922

1923

P3

Month




Polish Index Numbers of Wholesale Prices, 1921-25

WholesalePrice Index1

25,13931,82732,88231,71032,63935,39245,65453,10060,20365,53958,58357,046

59,23163,44573,46575,10678,63487,694

101,587135,786152,365201,326275,647346,353

544,690859,110988,500

Year

1923

1924

1925

Month

AprilMayJuneJulyAugustSeptemberOctoberNovemberDecember


JanuaryFebruaryMarchAprilMay

WholesalePrice Index1

1,058,9201,125,3501,881,4103,069,9705,294,6807,302,200

27,380,68067,943,700

142,300,700

242,167,700248,429,600245,277,900242,321,800

Source: Young [36, vol. 2, p. 349].4914 = 100.


Table P4

Year

1919

1920

1921

1922

Month

JulyAugustSeptemberOctoberNovemberDecember




JanuaryFebruaryMarchAprilMayJuneJulyAugust

Polish Exchange Rates,

Cents perPolish Mark

6.885.633.883.081.881.29

.70

.68

.67

.60

.51

.59

.61

.47

.45

.37

.26

.16

.145

.130

.132

.130

.124

.082

.0516

.0489

.0256

.0212

.0290

.0313

.0327

.0286

.0236

.0262

.0249

.0237

.01850.0135

1919-25

Year

1922

1923

1924

1925

Month

SeptemberOctoberNovemberDecember

January

FebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecember


JuneJulyAugustSeptemberOctoberNovemberDecember

JanuaryFebruaryMarchAprilMayJune

Cents perPolish Mark

.0127

.0095

.0065

.0057

.0043

.0025

.0024

.0023

.0021

.0013

.0007

.0004

.00035

.0001113

.0000502

.0000234

.0000116

.0000109

.0000113

.0000114—

Cents perZloty

10.2919.2519.2319.2219.2219.2119.20

19.1819.1819.1819.1819.1819.18


changes were a dramatic move toward a balanced government budgetand the establishment of an independent central bank that was prohibitedfrom making additional unsecured loans to the government. In January1924, the minister of finance was granted broad powers to effect mone-


tary and fiscal reform. The minister immediately initiated the establish-ment of the Bank of Poland, which was to assume the functions of thePolish State Loan Bank. The eventual goal was to restore convertibilitywith gold. The bank was required to hold a 30% reserve behind its notes,to consist of gold and foreign paper assets denominated in stable curren-cies. Beyond this reserve, the bank's notes had to be secured by privatebills of exchange and silver. A maximum credit to the government of 50million zlotys was permitted. The government also moved swiftly tobalance the budget (see table PI).

In January 1924, a new currency unit became effective, the gold zloty,worth 1.8 million paper marks. The zloty was equal in gold content to19.29 cents.

Table P2 reveals that, from January 1924 to December 1924, the notecirculation of the central bank increased by a factor of 3.2, in the face ofrelative stability of the price level and the exchange rate (see tables P3and P4). This phenomenon matches what occurred in Austria and Hun-

Table P5 Polish Unemployed

1921:


1922:


74,00090,00080,00088,000

130,000115,00095,00065,00070,00078,000

120,000173,000

221,444206,442170,125148,625128,91698,58185,24069,69268,00061,00062,00075,000

1923:


1924:


81,184106,729114,576112,75593,73176,39764,56356,515

———

67,581

100,580110,737112,583109,00084,00097,870

149,097159,820155,245147,065150,180159,060

Source: Statistiches Jahrbuch fur das Deutsche Reich [33].


gary and has a similar explanation. As table P2 reveals, the increasednote circulation during this period was effectively backed 100% by gold,foreign exchange, and private paper.

The available figures on unemployment are summarized in table P5.The stabilization of the price level in January 1924 is accompanied by anabrupt rise in the number of unemployed. Another rise occurs in July of1924. While the figures indicate substantial unemployment in late 1924,unemployment is not an order of magnitude worse than before thestabilization, and certainly not anywhere nearly as bad as would bepredicted by application of the same method of analysis that was used tofabricate the prediction for the contemporary United States that eachpercentage point reduction in inflation would require a reduction of $220billion in real GNP.

The Polish zloty depreciated internationally from late 1925 onward butstabilized in autumn of 1926 at around 72% of its level of January 1924.At the same time, the domestic price level stabilized at about 50% aboveits level of January 1924. The threatened renewal of inflation has beenattributed to the government's premature relaxation of exchange con-trols and the tendency of the central bank to make private loans atinsufficient interest rates.26

2.6 Germany

After World War I, Germany owed staggering reparations to theAllied countries. This fact dominated Germany's public finance from1919 until 1923 and was a most important force for hyperinflation.

At the conclusion of the war, Germany experienced a political revolu-tion and established a republican government. The early postwar govern-ments were dominated by moderate Socialists, who for a variety ofreasons reached accommodations with centers of military and industrialpower of the prewar regime.27 These accommodations in effect under-mined the willingness and capability of the government to meet itsadmittedly staggering revenue needs through explicit taxation.

Of the four episodes that we have studied, Germany's hyperinflationwas the most spectacular, as the figures on wholesale prices and exchangerates in tables Gl and G2 reveal. The inflation became most severe afterthe military occupation of the Ruhr by the French in January 1923. TheGerman government was determined to fight the French occupation by apolicy of passive resistance, making direct payments to striking workerswhich were financed by discounting treasury bills with the Reichsbank.

Table G3 estimates the budget of Germany for 1920 to 1923.28 Thetable reveals that, except for 1923, the budget would not have been badlyout of balance except for the massive reparations payments made. Thedisruption caused to Germany's finances by the reparations situation is


Table Gl German Wholesale Prices, 1914-24

Year

1914

1915

1916

1917

Month

JanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecemberJanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecemberJanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecemberJanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecember

PriceIndex

96969695979999

109111118123125126133139142139139150146145147147148150151148149151152161159154153151151156158159163163165172203199201203203

Year

1918

1919

1920

1921

Month

JanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecemberJanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecemberJanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecemberJanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecember

PriceIndex

204198198204203209208235230234234245262270274286297308339422493562678803

1,2601,6901,7101,5701,5101,3801,3701,4501,5001,4701,5101,4401,4401,3801,3401,3301,3101,3701,4301,9202,0702,4603,4203,490


Table Gl (continued)

Year Month

1922 JanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecember

1923 JanuaryFebruaryMarchAprilMayJune

PriceIndex

3,6704,1005,4306,3606,4607,030

10,16019,20028,70056,600

115,100147,480278,500588,500488,800521,200817,000

1,938,500

Year Month

1923 JulyAugustSeptemberOctoberNovemberDecember

1924 JanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecember

PriceIndex

7,478,70094,404,100

2,394,889,300709,480,000,000

72,570,000,000,000126,160,000,000,000117,320,000,000,000116,170,000,000,000120,670,000,000,000124,050,000,000,000122,460,000,000,000115,900,000,000,000

1151

1201

1271

1311

1291

1311

Source: Young [36, vol. 1, p. 530].'On basis of prices in reichsmarks. (1 reichsmark = 1 trillion [1012] former marks.)

surely understated by the reparations figures given in table G3. For onething, considerably larger sums were initially expected of Germany thanit ever was eventually able to pay. For another thing, the extent ofGermany's total obligation and the required schedule of payments wasfor a long time uncertain and under negotiation. From the viewpoint thatthe value of a state's currency and other debt depends intimately on thefiscal policy it intends to run, the uncertainty about the reparations owedby the German government necessarily cast a long shadow over itsprospects for a stable currency.

As table G4 reveals, the note circulation of the Reichsbank increaseddramatically from 1921 to 1923, especially in the several months beforeNovember 1923. As pointed out by Young [36], at the end of October1923, over 99% of outstanding Reichsbank notes had been placed incirculation within the previous 30 days.29 Table G4 reveals the extent towhich the Reichsbank note circulation was backed by discounted treasurybills. During 1923, the Reichsbank also began discounting large volumesof commercial bills. Since these loans were made at nominal rates ofinterest far below the rate of inflation, they amounted virtually to govern-ment transfer payments to the recipients of the loans.

Especially during the great inflation of 1923, a force came into playwhich was also present in the other hyperinflations we have studied.


Table G2

Year

1920

1921

1922

Month



JanuaryFebruaryMarchAprilMayJuneJuly

German Exchange Rates, 1914-25

Cents per Mark

1.691.051.261.672.192.562.532.101.721.481.321.37

1.601.641.601.571.631.441.301.19

.96

.68

.39

.53

.52

.48

.36

.35

.34

.32

.20

Year

1922

1923

1924

19251

Month

AugustSeptemberOctoberNovemberDecember


MayJuneJulyAugustSeptemberOctoberNovemberDecember


JuneJulyAugustSeptemberOctoberNovemberDecember

January

Cents per Mark

.10

.07

.03

.01

.01

.007

.004

.005

.004

.002

.001

.000,3

.000,033,9

.000,001,88

.000,000,068

.000,000,000,043

.000,000,000,022,7

22.621.822.022.0

22.323.423.923.823.823.823.823.8

23.8


'Cents per rentenmark and (after October 1924) per reichsmark. 1 rentenmark is equivalentto 1 reichsmark or 1 billion former paper marks. The reichsmark is the equivalent of the goldmark worth 23.82 cents.

c .yon -o

3 O tou i u<U C >- 3

.3 -s £ '!

•§.§ o00 13 2

JCOa,u cu

E

c'+3CO

_ oo b Q

00

m oON Hr»^ ooCD

Q

g § gU x H

3 'CC/5 T3

T-i r- TJ-•* (N 00O NO rn

00 h rH >rH

i JHON ON ON ON

VO

-a 5 c5

c i_ <u3 3 C

<S p ww •— ii O r3

H D h U H

ooj

CQ

O w

U £

NO <—i ON f - ON

oo ^ H CM ON

N O O N O N O N O N O N O O O i n C M• ^ • c o c o c M T t O N O N O o i n T i -O N c O N o r - ~ i n < - i - < s - r o r o c M

in in TJ- ro

o o ^ T f o o H i n o O ' t t^HTJ-CMNOCMONON'—I NOoor--ONinin-^cM<-ôCM" OO" ri O" O" r-" o" o" •*"m o o o c M r - - t — oot--or i—i o o -"' oo ON TJ- ©t--" NO" OO" OO" I-T O " in" r-T o"c M r o ^ t i n r ^ O N T - H t ^ oT - H T — I T — ( ^ — l T - H < - H C M C M T l *

r~- in ON NO ON r- 00 TJ- ( T -CM ON H r- NO inin rr CM •<*• NO r-

c M O o o c M O N i n i n c M c MT j - N O N o i n O N O c o O T j -N t ^ N O 0 0 I f l r - ( O r HCM" CM" CM" CM" r-T ^H" ,-*" T-H" ^H"

N \O H f<1 \O H (V)ON ID in © f- m CMin oo rH TJ- ro r- i—i

O CM ^t ©• * H O\ nin TJ- oo NO

r- ^H t~- oo f -NO ON ro NO i—iON 00 ^H r- TJ-

i- o fi H n IN. Hos ON ro Tt m © ooin \o in oo ON NO ON

CO OCM 00

O ro

c M o o r ^ ^ i - c M t - - c M c o ©© © T r c M c M - ^ r o r - i nTtoocMinONr-cMr~^No^©" ^4" T* r-T co" in" oo" m" ON"NOinco^HOscMinNONor—ICMinNOt^r^0Or~-rHO ^^ ^D Vi r**~ O r**** ^^ ON

M f O i " ' / ) l « O X O i 1 ' N tT-H T-H T-H f—i i—i I-H cN r^ ro

a p % B

oo in r-i^HrOr-le O O N O

,0ONNO

NO oo CN oo i n ON Tt NO enNO" ^H~ in" r-Tcn O ON ON

ON, c i cii i n eni eni in NO oo

oo

en

624030

425

!n

974

773

552

NO

cninON

282

422

<N

906

CNcn

CN)

948

466

ofN

322

030

NO

574

548

oooo

597

in

NOCNmCN

522

00ON

532

787637

in

263

599

CN530

819

m

O O C N r f ^ C N° \ "*„ °V ^ Ô\ -^r irT en" oo'i n i n m

NOr-

356

578

793

170

919

655

699

,606

545

297

,306

074

705

,776

418

cnr-H

,924

072

016

,211

560

598

,036

630

198

,252

066

CN

863

in

00

en

876

00

794

780

,636

441

045

enen

026

368

inr—

986948

,724

NO ON CM i n

ON cn ON O ON f-• * • * NO c n m ONNO ^ t ON ON ON O

t 00 * ^ NO00 NO ON NO r-t

CN fN CN CN CN

267

155

©

965

948

246

296

235

422

424

216

697

080

341

628

757

©

372

CN

724

NO

986

CM

720

693

014

O "tf- •«*

NO en NO00 00 rj-ON T—I \OT—1 H

914

NO

094

099

en

855

129

058

254

577

529

Osen

776

037

©cn

347

986948

724

322

494

201

477

g

222

672

359

464

s

CN

00

o

609

994

363

947

CN

661

o

552

348

899

224

NO

840

904

021

oo

338

00

752

cnin

ON

fN

987224

NO

CN

in

939

650

578

NO

r-oo

468

iON00

250

330

874

o

O

X

ioZ

X)

<L>{_)<UQ

enCMON

;>C03

C

*5 < June

July

Augi

Sept<

<uX3E

1- Oju >X O

Octo

15 N

u

10

Z©

<DX)

s

Dece

1924

cd 3 "^ —.

Janu

Febr

Marc

Apri

Mav

3 l 3 § !x a1 av

2 S E D

Q £

T3 i2

- § ' |o u <uH Q Q

oC "3

o .bZ U

833,

m

357, i

00CM

855,

oCM

093,

392,

oCs)

824,

m

649,

CO

980,

Os

302,

00

CO

CO

inCM

905,

CMro

CMCMCMOOOinsDrJ-r--i n r - c o s o o s T j - T - H i n r - -t q_ w H oy t-^ CM ^ nr-T so" oo" so" t--" co" so" • * " CM"c M c M i n - — i c M r - r - c M r - ^r f i n c O s O ' — i H o \ i-• O

o so CM *-i <—i co o so oo m os os O c O O s r ^ - O O C O m O c O O s c O© , O^ CM CM CM OS_ i-j_ t-^ CM O^ rf_ rôo" vd" vo" o " <ri" •*" •*" oo' CM" CT OO" - t "r - ~ v o m > O T t T t ^ H O s v o s c i s O t - (t-^ q oo CM ir)_ r^ o^ r~- r* o^ r- mT-H" O " CM" ON" O " •*" rH oo" in" r " o" irT» - H ^ H T - H T — I T — ( '—I T-HT—I CM CM

502

inCM

00

oin

Os

r-oCM

597

CM

306

038

CM

H 00

so r-

33CM CM

489

778

oCO

416

890

978

Osr-

CM CO SO

os r-- oo

o c M i n c M O s f - r ^ c o o s• — t~- co CM T—i o

CM 00 t— CM CO>o"so"co"r-^'-în"rôcr-t"o o O ' * r - T — I C M O S O C OcMoor~ in r~T - iT -H r~ -o

f- in "-H c<-f

rr Os oo ir>O in CM CMoo os CM r^

m co T-H so os CMos in CM oo r-- coo oo r- CM so so

O s O t - ~ O s O O T - H O C M T l - t - - C OC M C M T - H C O c O C M O s r - ' O O C M - ^ -

s o r j - T f o o o o c o c o o c o i n o s

SOr-~CMt--OsCMCMOOss o o o i n m r ~ - o s c M s o o sr~-cocoO'-Hr~-r~-'-Ht~-in so IH o" o\ r i ^t Is-" oCr~-cMr- -cM'* '—IOSTJ-SOn O s O t ^ N h H o oin"cTcTcrf-ros'os"o6"so"T - i c M c o T j - i n s o o o c O T - H1 1 T-H T—I T-H T-H T—t T-H CM CO

S 3 - f i - ts sX)

ts a3 U

S O Tt NO© oo in

oo oo NO • *

259

779459

969

240

Z.83

526

530

229

264

763

036

ooON

582

584

083

272

793

274

854

en

231

070

063

>n

666

NO

952

ON

857

CM

080

591

r-NO

966

NO

703

580

sooen

417

597

552

129

559

740

976

373

005

r-ON

024

548

00 00 IT)rJ m NOr~- ON oo

e n e N r H O o o o o e n ©r ^ T t N o r ~ - f N © © r s i

- . i n NO oo r- r- oo

en "* NOen rH osen in CM

8 of

r- oo m © ONin in eo rH Tt© NO en f- OSCM © in oo Tf

Tt -rj- rH Tt Ô rnCN oo oo CM NO o

en r~- - * en oo

g s S SrH rH en

) rHNO

°ien"

,095,

t-»m

,497,

enmCM

906'

CM

936

270

en

945

353

©in

ONON

oON

enin

537

058

r-

266

mON

253

293

550

368

079

403

454

366

575

359

327

rH

1,648

863

235

3,779

168

903

467

186

00

673

660

606

513

230

2,457

906

936

0,478

004

560

en"

tN (N (N00

oo

456

469

ON

o

086

754

831

094

,280

369

496

,984

777

787

,512

651

,517

355

984

,545

470

749

,563

rH IT,ON ONCM in

£3

O mCM 00

NO CMNO <N

00CM

909

822

,496

743

720

,844

302

640

,267

772

424

,507

en in NO oo

e n r o o o Ô O O O O O f - C N O N

f o r O O

oNOcTenO N t N O N O O N O - *

r o o o

e E

O Z Q

b S

X) X)

V <U X)Q O 2 fii||

I o o 2 6: £ O O Si H


Given the method of assessing taxes in nominal terms, lags between thetime when taxes were levied and the time when they were collected led toreduced revenues as the government evidently repeatedly underesti-mated the prospective rate of inflation and as the rapid inflation gavepeople a large incentive to delay paying their taxes. This effect probablypartially accounts for the reduced tax revenues collected during the firstnine months of 1923. The French occupation of the Ruhr also helpsexplain it.

In response to the inflationary public finance and despite the efforts ofthe government to impose exchange controls, there occurred a "flightfrom the German mark" in which the real value of Reichsmark notesdecreased dramatically. The figures in table Gl indicate that betweenJanuary 1922 and July 1923, wholesale prices increased by a factor of2,038 while Reichsbank notes increased by a factor of 378. BetweenJanuary 1922 and August 1923, wholesale prices increased by a factor of25,723 while Reichsbank notes circulating increased by a factor of 5,748.The fact that prices increased proportionately many times more than didthe Reichsbank note circulation is symptomatic of the efforts of Germansto economize on their holdings of rapidly depreciating German marks.Toward the end of the hyperinflation, Germans made every effort toavoid holding marks and held large quantities of foreign exchange forpurposes of conducting transaction. By October 1923, it has been roughlyestimated, the real value of foreign currencies circulating in Germany wasat least equal to and perhaps several times the real value of Reichsbanknotes circulating.30

The figures in tables Gl and G2 show that prices suddenly stoppedrising and the mark stopped depreciating in late November 1923. Theevent of stabilization was attended by a "monetary reform," in which on15 October 1923 a new currency unit called the Rentenmark was declaredequivalent to 1 trillion (1012) paper marks. While great psychological

Footnotes to Table G4

Source: Young [36, vol. 1, pp. 528-29].Note: End of month figures, in thousands of current marks; from January 1924 in thousandsof rentenmarks or reichsmarks. 1 rentenmark is equivalent to 1 reichsmark or 1 trillion (1012)former paper marks. The reichsmark is the equivalent of the gold mark worth 23.82 cents.'The large increase of advances at the close of November 1922 occurred because theReichsbank had to take over temporarily the financing of food supplies from the loanbureaus (Darlehuskassen), as the latter were unable to extend the needed accommodation,their outstanding notes having reached the maximum amount permitted by law.2In billions.3A decree of 15 November 1923 discontinued the discounting of treasury bills by theReichsbank.4See note above.5Date of first statement of reorganized Reichsbank.


significance has sometimes been assigned to this unit change, it is difficultto attribute any substantial effects to what was in itself only a cosmeticmeasure.31 The substantive aspect of the decree of 15 October was theestablishing of a Rentenbank to take over the note issue functions of theReichbank. The decree put binding limits upon both the total volume ofRentenmarks that could be issued, 3.2 billion marks, and the maximumamount that could be issued to the government, 1.2 billion marks. Thislimitation on the amount of credit that could be extended to the govern-ment was announced at a time when the government was financingvirtually 100% of its expenditures by means of note issue.32 In December1923, the management of the Rentenbank was tested by the governmentand effectively made clear its intent to meet its obligation to limit govern-ment borrowing to within the amount decreed.

Simultaneously and abruptly three things happened: additional gov-ernment borrowing from the central bank stopped, the governmentbudget swung into balance, and inflation stopped. Table G5 shows thedramatic progress toward a balanced budget that was made in the monthsafter the Rentenbank decree.

The government moved to balance the budget by taking a series ofdeliberate, permanent actions to raise taxes and eliminate expenditures.

Table G5 Ordinary Revenues and Expenditures of the German FederalGovernment (from Wirtschift and Statistik, issued bythe Statistisches Reichsamt, in millions of gold marks)

1923:

NovemberDecember

1924:

JanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctober

Ordinary

Total

68.1333.9

520.6445.0632.4579.5566.7529.7622.2618.2665.6714.3

Revenue

Of WhichTaxesYielded

63.2312.3

503.5418.0595.3523.8518.7472.3583.1592.0609.2686.7

OrdinaryExpen-ditures

668.7

396.5462.8498.6523.5459.1504.5535.1597.6581.6693.0

Excess of

JXCVCllUC

( + ) or Ex-penditure

( - )

-334.8

+ 124.1-17 .8133.8

+ 56.0+ 107.6

+ 25.2+ 86.9+ 20.6+ 84.0+ 21.3



Young reports that "by the personnel decree of October 27, 1923, thenumber of government employees was cut by 25 percent; all temporaryemployees were to be discharged; all above the age of 65 years were to beretired. An additional 10 percent of the civil servants were to be dis-charged by January 1924. The railways, overstaffed as a result of post-wardemobilization, discharged 120,000 men during 1923 and 60,000 moreduring 1924. The postal administration reduced its staff by 65,000 men;the Reichsbank itself which had increased the number of its employeesfrom 13,316 at the close of 1922 to 22,909 at the close of 1923, began thedischarge of its superfluous force in December, as soon as the effects ofstabilization became manifest."33

Substantially aiding the fiscal situation, Germany also obtained relieffrom her reparation obligations. Reparations payments were temporarilysuspended, and the Dawes plan assigned Germany a much more manage-able schedule of payments.

Table G4 documents a pattern that we have seen in the three otherhyperinflations: the substantial growth of central bank note and demanddeposit liabilities in the months after the currency was stabilized. As inthe other cases that we have studied, the best explanation for this is that atthe margin the postinflation increase in notes was no longer backed bygovernment debt. Instead, in the German case, it was largely backed bydiscounted commercial bills. The nature of the system of promises andclaims behind the central bank's liabilities changed when after the Ren-tenbank decree the central bank no longer offered additional credit to thegovernment. So once again the interpretation of the time series on centralbank notes and deposits must undergo a very substantial change.

By all available measures, the stabilization of the German mark wasaccompanied by increases in output and employment and decreases inunemployment.34 While 1924 was not a good year for German business, itwas much better than 1923. Table G6 is representative of the figuresassembled by Graham, and shows that 1924 suffers in comparison with1922 but that 1925 was a good year. In these figures one cannot find muchconvincing evidence of a favorable trade-off between inflation and out-

Table G6 Index of Physical Volumeof Production per Capita in Germany

Year

1920192119221923

Index ofProduction

61778654

Year

1924192519261927

Index ofProduction

779086111

Source: Graham [7, p. 287].


put, since the year of spectacular inflation, 1923 was a very bad year foremployment and physical production. Certainly a large part of the poorperformance of 1923 was due to the French occupation of the Ruhr andthe policy of passive resistance.

Despite the evident absence of a "Phillips curve" trade-off betweeninflation and real output in the figures in tables Gl and G6, there is ampleevidence that the German inflation was far from "neutral" and that therewere important "real effects." Graham [7] gives evidence that the infla-tion and the associated reduction in real rates of return to "high-powered" money and other government debt were accompanied by realoverinvestment in many kinds of capital goods.35 There is little doubt thatthe "irrational" structure of capital characterizing Germany after stabi-lization led to subsequent problems of adjustment in labor and othermarkets.

2.7 Czechoslovakia

After World War I, the new nation of Czechoslovakia was formed outof territories formerly belonging to Austria and Hungary. Under theleadership of a distinguished minister of finance, Dr. Alois Rasin, im-mediately after the war Czechoslovakia adopted the conservative fiscaland monetary policies its neighbors adopted only after their currencieshad depreciated radically. As a result, Czechoslovakia avoided thehyperinflation experienced by its neighbors.

Under Rasin's leadership, Czechoslovakia early on showed that it wasserious about attaining a stable currency. Even before the peace treatiesrequired it, Czechoslovakia stamped the Austro-Hungarian notes thencirculating within its border with the Czechoslovakian stamp, therebyrecognizing them as its own debt. There was considerable drama associ-ated with this event, as the National Assembly passed the plans forstamping in secret sessions on 25 February 1919. From 26 February to 9March, the frontiers of the country were unexpectedly closed and foreignmail service was closed. Only Austro-Hungarian notes circulating withinthe country could be presented for stamping. As part of the stampingprocess, the government retained part of the notes in the form of a forcedloan.36 About 8 billion crowns were stamped.

A banking office in the Ministry of Finance took over the affairs of theold Austro-Hungarian bank. Czechoslovakia moved quickly to limit bystatute the total government note circulation and to prevent inflationarygovernment finance. A law of 10 April 1919 strictly limited the fiduciaryor unbacked note circulation of the banking office to about 7 billioncrowns. This law was obeyed, and forced the government to finance itsexpenditures by levying taxes or else issuing debt, which, because of the


statutory restriction on government note issues, were interpreted aspromises to tax in the future.

From 1920 on, Czechoslovakia ran only modest deficits on currentaccount (see table Cl). Among other taxes, Czechoslovakia imposed aprogressive capital levy on property, which raised a cumulative amount ofabout 11 billion crowns by 1925. It also imposed an increment tax on theincreased wealth individuals had obtained during the war.

Table C2 shows the note and deposit liabilities of the banking office.The government's abstention from inflationary finance shows up in thesefigures.

Table Cl

Revenue:OrdinaryExtraordinary

Total

Expenditure:OrdinaryExtraordinary

Total

Deficit

Czechoslovakia, Receipts and Expenditures, 1919-25 (exclusive ofexpenditures for capital improvements covered by loans)

1919

Esti-mated

2,6141,096

3,710

2,6106,005

8,615

4,905

Ac-tual

—

—

—

7,450

1920

Esti-mated

7,9502,477

10,427

7,1758,103

15,278

4,851

Ac-tual

—

13,455

—

13,931

476

1922

Esti-mated

15,9231,376

17,299

10,6727,354

18,026

727

Ac-tual

—

21,894

—

18,558

1922

Esti-mated

17,2911,593

18,884

13,2896,524

19,813

929

Ac-tual

—

17,733

—

18,663

930Surplus — — — — 3,336 — —

1923 1924 1925

Revenue:OrdinaryExtraordinary

Total

Expenditure:OrdinaryExtraordinary

Total

DeficitSurplus

Esti-mated

17,961851

18,812

13,6055,773

19,378

565

Ac-tual

—

15,664

—

16,540

876

Esti-mated

15,987404

16,391

12,2004,703

16,993

603

Ac-tual

—

—

—

—

—

Esti-mated

—

15,702

—

15,974

272

Ac-tual

—

—

—

—

—



Table C3 shows the path of exchange rates and how, after declininguntil November 1921, the Czechoslovakian crown rapidly gained to about3 United States cents.

Table C4 shows the price levels. From 1922 to 1923, Czechoslovakiaactually experienced a deflation. Indeed, Rasin's initial plan had been to

Table C2 Note Issue of Banking Office of Czechoslovakia, 1919-24(in thousands of Czech crowns)

Year

1919

1920

1921

1922

Month





State Notesin Circulation

—

——

161,106664,997

1,443,5702,512,1993,513,4054,723,303

5,574,6886,462,8257,216,4387,216,4388,268,6959,729,2339,267,8749,814,920

10,310,22810,920,51410,946,65311,288,512

10,888,31910,914,78610,921,95610,928,56010,851,40311,167,51511,134,32711,455,17511,570,88112,327,15911,871,64712,129,573

11,230,06510,743,95810,323,06910,075,757

Year

1923

1924

1925

Month

MayJuneJulyAugustSeptemberOctoberNovemberDecember




State Notesin Circulation

9,717,7509,838,2059,916,077

10,171,38310,196,88010,139,3669,996,550

10,064,049

9,222,4348,947,9889,157,4079,567,3699,327,6769,375,9919,448,0869,218,4759,311,3789,278,9999,250,6889,598,903

8,820,0938,506,4678,280,3908,198,6539,078,4188,081,1068,090,0349,139,7928,222,6588,585,8478,500,9428,810,357

7,916,5407,727,8807,680,8677,525,934

Source: Young [36, vol. 2, pp. 305-6].

Table C3

Year

1919

1920

1921

1922

Month





Czechoslovakian Exchange Rates, 1919-24

Centsper Crown

—

——

6.135——

5.6254.5754.5753.1001.9501.900

1.425.975

1.2751.5302.1952.3352.1951.8101.5351.2451.1651.190

1.3001.2901.3071.3651.4601.4201.3121.2251.1601.0491.0381.249

1.7321.8551.733

Year

1922

1923

1924

1925

Month




JanuaryFebruaryMarchAprilMayJune

Centsper Crown

1.9601.9211.9242.1852.9023.2313.2853.1763.097

2.8562.9582.9692.9782.9792.9932.9972.9342.9952.9712.9062.925

2.8982.9022.9022.9572.9392.9362.9532.9792.9932.9812.9893.018

3.002.962.972.962.962.96



Table

Year

1922

1923

C4

Month


JanuaryFebruaryMarchAprilMayJuneJulyAugustSeptember

Czechoslovakian

WholesalePrice Index

1,6751,5201,5521,4911,4711,4711,4641,3861,1551,0591,017

999

1,0031,0191,0281,0311,0301,001

968958957

Wholesale Prices, 1922-24

Year Month

OctoberNovemberDecember

1924 JanuaryFebruary

MarchAprilMayJuneJulyAugust

SeptemberOctoberNovemberDecember

1925 JanuaryFebruary

MarchAprilMay

WholesalePrice Index

973965984

974999

1,0211,0081,015

981953986

982999

1,0131,024

1,0451,048

1,0341,0191,006


Note: July 1914 = 100.

restore the Czechoslovakia crown to the prewar gold par value of the oldAustro-Hungarian crown. Following Rasin's assassination, this plan wasabandoned and the crown was stabilized at about 2.96 cents.

2.8 Conclusion

The essential measures that ended hyperinflation in each of Germany,Austria, Hungary, and Poland were, first, the creation of an independentcentral bank that was legally committed to refuse the government'sdemand for additional unsecured credit and, second, a simultaneousalteration in the fiscal policy regime.37 These measures were interrelatedand coordinated. They had the effect of binding the government to placeits debt with private parties and foreign governments which would valuethat debt according to whether it was backed by sufficiently largeprospective taxes relative to public expenditures. In each case that wehave studied, once it became widely understood that the governmentwould not rely on the central bank for its finances, the inflation termi-nated and the exchanges stabilized. We have further seen that it was notsimply the increasing quantity of central bank notes that caused the


hyperinflation, since in each case the note circulation continued to growrapidly after the exchange rate and price level had been stabilized.Rather, it was the growth of fiat currency which was unbacked, or backedonly by government bills, which there never was a prospect to retirethrough taxation.

The changes that ended the hyperinflations were not isolated restric-tive actions within a given set of rules of the game or general policy.Earlier attempts to stabilize the exchanges in Hungary under Hegedus,38

and also in Germany, failed precisely because they did not change therules of the game under which fiscal policy had to be conducted.39

In discussing this subject with various people, I have encountered theview that the events described here are so extreme and bizarre that theydo not bear on the subject of inflation in the contemporary United States.On the contrary, it is precisely because the events were so extreme thatthey are relevant. The four incidents we have studied are akin to labora-tory experiments in which the elemental forces that cause and can be usedto stop inflation are easiest to spot. I believe that these incidents are full oflessons about our own, less drastic predicament with inflation, if only weinterpret them correctly.

Notes1. "Most economists believe that the underlying inflation rate—roughly defined as wage

costs less productivity gains—now stands at 9 to 10 percent, and that only a long period ofrestraint can reduce that rate significantly" (Newsweek, 19 May 1980, p. 59).

2. Paul Samuelson has aptly summarized the rational expectations view: "I should reportthat there is a new school, the so-called 'rational expectationists.' They are optimistic thatinflation can be wiped out with little pain if only the government makes credible itsdetermination to do so. But neither history nor reason tempt one to bet their way"(Newsweek, 28 April 1980). The second sentence of this quote is probably as shrewd asummary of the rational expectations view as can be made in a single sentence. However, itis difficult to agree with the third sentence: as for "reason," no one denies that logicallycoherent and well-reasoned models underlie the claims of the "rational expectationists"; asfor history, the evidence summarized in this paper is surely relevant.

3. There is actually no such thing as a "rational expectations school" in the sense of acollection of economists with an agreed upon model of the economy and view about optimalmonetary and fiscal policy. In fact, among economists who use the assumption of rationalexpectations there is wide disagreement about these matters. What characterizes adherentsof the notion of rational expectations is their intention to build models by assuming thatprivate agents understand the dynamic environment in which they operate approximately aswell as do government policymakers. Adherence to this notion leaves ample room forsubstantial diversity about the many other details of a model. For some examples of rationalexpectations models with diverse implications, see Lucas [21], Barro [2], Wallace [35],Townsend [34], and Sargent and Wallace [31]. Despite their diversity, it is true that all ofthese models impel us to think about optimal government policy in substantially differentways than were standard in macroeconomics before the advent of the doctrine of rationalexpectations in the early 1970s.


4. Bresciani-Turroni wrote: "Whoever studies the recent economic history of Europe isstruck by a most surprising fact: the rapid monetary restoration of some countries where forseveral years paper money had continually depreciated. In some cases the stabilization ofthe exchange was not obtained by a continuous effort, prolonged over a period of years,whose effects would show themselves slowly in the progressive economic and financialrestoration of the country, as occurred before the War in several well-known cases ofmonetary reform. Instead, the passing from a period of tempestuous depreciation of thecurrency to an almost complete stability of the exchange was very sudden" [3, p. 334].Compare these remarks with the opinion of Samuelson cited in note 2 above.

5. The notes were "backed" mainly by treasury bills which, in those times, could not beexpected to be paid off by levying taxes, but only by printing more notes or treasury bills.

6. League of Nations [13, p. 101].7. Keynes wrote: "It is not lack of gold but the absence of other internal adjustments

which prevents the leading European countries from returning to a pre-war gold standard.Most of them have plenty of gold for the purpose as soon as the other conditions favorable tothe restoration of a gold standard have returned" (Keynes [11, p. 132]). Writing aboutGermany in 1923, Keynes said: "The government cannot introduce a sound money,because, in the absence of other revenue, the printing of an unsound money is the only wayby which it can live" (Keynes [10, p. 67]).

8. This view can be expressed more precisely by referring to the technical literature ofoptimum economic growth. I am recommending that a good first model of the gold standardor other commodity money is a real equilibrium growth model in which a government issuesdebt, makes expenditures, and collects taxes. Examples of these models were studied byArrow and Kurz [1]. In such models, government debt is valued according to the sameeconomic considerations that give private debt value, namely, the prospective net revenuestream of the institution issuing the debt. A real equilibrium growth model of this kind canalso be used to provide a formal rationalization of my claim below that open marketoperations in private securities, foreign exchange, and gold should have no effect on theprice level, i.e. the value of government demand debt.

9. It is relatively straightforward to produce a variety of workable theoretical models of acommodity money or gold standard, along the lines of note 8. It is considerably moredifficult to produce a model of a fiat money, which is costless to produce, inconvertible, andof no utility except in exchange. Kareken and Wallace [9], Wallace [35], and Townsend [34]describe some of the ramifications of this observation. The workable models of fiat moneythat we do have—for example, those of Townsend [34] and Wallace [35]—immediately raisethe question of whether voluntarily held fiat money can continue to be valued at all in theface of substantial budget deficits of the order of magnitude studied in this paper. Suchmodels lead one to assign an important role to government restrictions, particularly onforeign exchange transactions, in maintaining a valued, if involuntarily held, fiat money.Keynes [10] and Nichols [24] also emphasized the role of such restrictions.

10. The sweeping implications of this principle for standard ways of formulating andusing econometric models were first described by Lucas [19]. The principle itself hasemerged in a variety of contexts involving economic dynamics. For some examples, seeLucas [20] and Sargent and Wallace [30].

11. Sargent and Wallace [32] describe a sense in which it might be difficult to imagine thata regime change can occur. As they discovered, thinking about regime changes in thecontext of rational expectations models soon leads one to issues of free will.

12. The Treaty of St. Germain, signed in September of 1919, required the successorstates of the Austro-Hungarian empire to stamp their share of the notes of the Austro-Hungarian bank. The stamp converted those notes to the currency, i.e. debt, of the newstates. The Austrian section of the old Austro-Hungarian bank functioned as the centralbank of Austria for several years after the war.

13. Needless to say, the central bank encountered a strong demand for loans at this rateand had to ration credit.


14. At the time, some commentators argued that since the real value of currency haddecreased and so in a sense currency was scarce, the increased note issue of the central bankwas not the prime cause of the inflation. Some even argued that money was "tight" and thatthe central bank was valiantly struggling to meet the shortage of currency by adding printingpresses and employees. This argument is now widely regarded as fallacious by macroecono-mists. Disturbingly, however, one hears the very same argument in the contemporaryUnited States.

15. "In Vienna, during the period of collapse, mushroom exchange banks sprang up atevery street corner, where you could change your krone into Zurich francs within a fewminutes of receiving them, and so avoid the risk of loss during the time it would take you toreach your usual bank. It became a reasonable criticism to allege that a prudent man at acafe ordering a bock of beer should order a second bock at the same time, even at theexpense of drinking it tepid, lest the price should rise meanwhile" (Keynes [10, p. 51]).

16. See Young [36, vol. 2, p. 16]. That a government might want to adopt such measuresif it were using inflationary finance was pointed out by Nichols [24].

17. Pasvolsky [25, p. 116].18. The content of this protocol is highly sensible when it is remembered that the value of

a state's currency and other debt, at least under the gold standard, is determined by itsability to back that debt with an appropriate fiscal policy. In this respect, its situation is nodifferent from that of a firm. In 1922, there was widespread concern within and withoutAustria that its sovereignty was at risk. (See the desperate note delivered by the Austrianminister to the Supreme Council of the Allied governments quoted by Pasvolsky [25, p.115]). The first protocol aimed to clarify the extent to which Austria remained a political andeconomic entity capable of backing its debts. A similar protocol was signed at the inceptionof Hungary's financial reconstruction.

19. It should be noted that for two years the new bank vigorously exercised its authorityto control transactions in foreign currency. Only after March 1925 were restrictions ontrading foreign exchange removed.

20. This explanation is consistent with the argument advanced by Fama [6]. There is analternative explanation of these observations that neglects the distinction between insideand outside money, and that interprets the observations in terms of a demand function forthe total quantity of "money." For instance, Cagan [4] posited the demand schedule formoney to take the form

(1) Mt-Pt = *{E,Pt+x-Pt), a<0,

where P, is the logarithm of the price level, M, is the logarithm of the money supply, andEtPt+i is people's expectation of the log of price next period. There is always a problem indenning an empirical counterpart to A/,, but it is often taken to be the note and depositliabilities of the central bank or "high-powered" money. The money demand schedule or"portfolio balance" schedule incorporates the idea that people want to hold less wealth inthe form of real balances the faster the currency is expected to depreciate. Equation (1) canbe solved to give an expression for the equilibrium price level of the form

(2) Pt = J -1

where EtMt+i is what at time t people expect the money supply to be at time t + i.Consider the following two experiments. First, suppose that the government engages in a

policy, which everyone knows in advance, of making the money supply grow at the constanthigh rate (JL > 0 from time 0 to time T - 1, and then at the rate zero from time T onward. Inthis case, the inflation rate would follow the path depicted in figure 2.N.I.

For the second experiment, suppose that initially everyone expected the money supply toincrease at the constant rate |x forever but that at time Tit becomes known that henceforththe money supply will increase at the rate 0 forever. In this case, the inflation rate takes a


0

0

Fig. 2.N.1 Inflation path with an expected decrease in money supplygrowth from |x to 0 at time T.

sudden drop at time T, as shown by the path in figure 2.N.2. Now since the inflation and theexpected inflation rate experience a sudden drop at Tin this case, it follows from equation(1) that real balances must increase at T. This will require a sudden once and for all drop inthe price level at T.

This second example of a previously unexpected decrease in the inflation rate providesthe material for an explanation of the growth of the money supplies after currency stabiliza-tion. In the face of a previously unexpected, sudden, and permanent drop in the rate ofmoney creation, the only way to avoid a sudden drop in the price level would be to

t+1 t

0

0 T

Fig. 2.N.2 Inflation path with a previously unexpected decrease in moneysupply growth from |x to 0 at time T.


accompany the decrease in the rate of money creation with a once and for all increase in themoney supply. In order to stabilize the price level in the face of a decreased rate of change ofmoney, the level of the money supply must jump upward once and for all.

What actually occurred in the four countries studied here was not a once and for all jumpbut a gradual increase in the money supply over many months. This could be reconciled withthe observations within the model (1) if people were assumed only gradually to catch on tothe fact of stabilization and to decrease the rate of inflation that they expected as thecurrency stabilization continued to hold. I find this explanation hard to accept, but it is apossibility.

An alternative way to reconcile the preceding explanation with the gradual upwardmovement of "high-powered" money after the stabilizations is to add adjustment lags to theportfolio balance schedule (1). For example, consider replacing (1) with

(1') (M, ~ P.) = a(EtPt+1 - Pt) + X(M,_, - Pr_,)c t<O,O<\<l .

In this case, an abrupt stabilization of expected inflation induces only a gradual adjustmentof real balances upward at the rate of 1 - X. per period. My own preference at this point is foran explanation that stresses the distinction between backed and unbacked money.

21. See Pasvolsky [25, p. 161].22. See Pasvolsky [25, p. 298].23. Within a year and a half, these became a claim on gold as Britain returned to the gold

standard.24. Unlike Austria, Hungary, and Germany, Poland did not owe war reparations.25. League of Nations [13, p. 111].26. Ibid., p. 108.27. See the account in Paxton [26, pp. 146-50].28. Also see Graham [7, pp. 40-41].29. Keynes wrote: "A government can live for a long time, even the German government

or the Russian government, by printing paper money . . . A government can live by thismeans when it can live by no other" (Keynes [10, p. 47]).

30. See Young [36, vol. 1, p. 402] and Bresciani-Turroni [3, p. 345].31. After reading an earlier draft of this paper, John Kennan directed me to the following

passage in Constance Reid's biography of the mathematician Hilbert: "In 1923 the inflationended abruptly through the creation of a new unit of currency called the Rentenmark.Although Hilbert remarked sceptically, 'One cannot solve a problem by changing the nameof the independent variable,' the stability of conditions was gradually restored" (Reid [27,pp. 162-63]).

32. Young [36, vol. 1, p. 421].33. Young [36, vol. 1, p. 422].34. See Graham [7, chapter 12].35. Theoretical models of money along the lines proposed by Samuelson [29] predict that

too much capital will be accumulated when the government fiscal policy is so profligate thatmoney becomes valueless. See Samuelson [29] and Wallace [35].

36. The frontiers were closed to prevent notes from Austria and Hungary from enteringthe country. The Treaty of St. Germain, signed 10 September 1919, provided that thesuccessor states should stamp the Austro-Hungarian notes, signifying their assumption ofthe debt.

37. Of inflationary finance, Keynes wrote: "It is common to speak as though, when agovernment pays its way by inflation, the people of the country avoid taxation. We haveseen this is not so. What is raised by printing notes is just as much taken from the public as isbeer-duty or an income-tax. What a government spends the public pay for. There is no suchthing as an uncovered deficit. But in some countries it seems possible to please and contentthe public, for a time at least, by giving them, in return for the taxes they pay, finely


engraved acknowledgments on water-marked paper. The income tax receipts, which we inEngland receive from the surveyor, we throw into the wastepaper basket; in Germany theycall them bank-notes and put them into their pocketbooks; in France they are termedRentes and are locked up in the family safe" (Keynes [10, pp. 68-69]).

38. See Pasvolsky [25, pp. 304-7].39. A deep objection to the interpretation in this paragraph can be constructed along the

lines of Sargent and Wallace [30], who argue that for a single economy it is impossible toconceive of a rational expectations model in which there can occur a change in regime. Inparticular, the substantial changes in ways of formulating monetary and fiscal policyassociated with the ends of the four inflations studied here can themselves be considered tohave been caused by the economic events preceding them. On this interpretation, what wehave interpreted as changes in the regime were really only the realization of events andhuman responses under a single, more complicated regime. This more complicated regimewould have to be described in a considerably more involved and "state contingent" waythan the simple regimes we have described. I believe that the data of this paper could bedescribed using this view, but that it would substantially complicate the language andrequire extensive qualifications without altering the main practical implications.

References

1. Arrow, K. J., and Mordecai, K. 1970. Public investment, the rate ofreturn, and optimal fiscal policy. Baltimore and London: John Hop-kins University Press.

2. Barro, R. J. 1974. Are government bonds net wealth? Journal ofPolitical Economy 82, no. 6 (November/December): 1095-1118.

3. Bresciani-Turroni, C. 1937. The economics of inflation. London:George Allen & Unwin.

4. Cagan, P. 1956. The monetary dynamics of hyperinflation. In M.Friedman, ed., Studies in the quantity of money. Chicago: Universityof Chicago Press.

5. de Bordes, J. v. W. 1924. The Austrian crown. London: P. S. King &Son.

6. Fama, E. F. 1980. Banking in the theory of finance. Journal ofMonetary Economics 6, no. 1 (January): 39-58.

7. Graham, F. D. 1930. Exchange, prices, and production in hyperinfla-tion: Germany, 1920-23. New York: Russell & Russell.

8. Kareken, J. H., and Wallace, N. 1980. Models of monetary econom-ics. Minneapolis: Federal Reserve Bank of Minneapolis.

9. Kareken, J. H., and Wallace, N. 1980. Introduction to Kareken andWallace [8].

10. Keynes, J. M. 1924. Monetary reform. New York: Harcourt, Brace&Co.

11. Keynes, J. M. 1925. The United States and gold. In Young [36, vol.1, pp. 131-33].


12. Layton, W. T., and Rist, C. 1925. The economic situation of Austria.Report presented to the Council of the League of Nations, Geneva,19 August.

13. League of Nations. 1946. The course and control of inflation.Geneva.

14. League of Nations. 1926. The financial reconstruction of Austria.General Survey and Principal Documents, Geneva.

15. League of Nations. 1926. The financial reconstruction of Hungary.General Survey and Principal Documents. Geneva.

16. League of Nations. 1926. Memorandum on currency and centralbanks, 1913-1925, vol. 1. Geneva.

17. League of Nations. 1927. Memorandum on public finance (1922-1926). Geneva.

18. League of Nations. 1925. Memorandum on currency and centralbanks, 1913-1924, vol. 1. Geneva.

19. Lucas, R. E., Jr. 1976. Econometric policy evaluation: A critique. InK. Brunner and A. H. Meltzer, eds., The Phillips curve and labormarkets, pp. 19-46. Carnegie-Rochester Conference Series on Pub-lic Policy, Vol. 1. Amsterdam: North-Holland.

20. Lucas, R. E., Jr. 1972. Econometric testing of the natural ratehypothesis. In O. Eckstein, ed., The Econometrics of Price Deter-mination Conference. Washington: Board of Governors, FederalReserve System.

21. Lucas, R. E., Jr. 1980. Equilibrium in a pure currency economy. InKareken and Wallace [8].

22. Lucas, R. E., Jr., and Sargent, T. J. 1979. After Keynesian econom-ics. Federal Reserve Bank of Minneapolis, Quarterly Review, spring.

23. Lucas, R. E., Jr., and Sargent, T. J. 1981. Rational expectations andeconometric practice. Introductory essay to R. E. Lucas, Jr., andT. J. Sargent, eds., Rational expectations and econometric practice.Minneapolis: University of Minnesota Press.

24. Nichols, D. 1974. Some principles of inflationary finance. Journal ofPolitical Economy 82, no. 2, part 1 (March/April): 423-30.

25. Pasvolsky, L. 1928. Economic nationalism of the Danubian states.New York: Macmillan.

26. Paxton, R. O. 1975. Europe in the twentieth century. New York:Harcourt Brace Jovanovich.

27. Reid, C. 1979. Hilbert. New York: Springer-Verlag.28. Salemi, M. K. 1976. Hyperinflation, exchange depreciation, and the

demand for money in post World War I Germany. Ph.D. thesis,University of Minnesota.

29. Samuelson, P. A. 1958. An exact consumption-loan model of in-terest with or without the social contrivance of money. Journal ofPolitical Economy 66 (December): 467-82.


30. Sargent, T. J. 1980. Rational expectations and the reconstruction ofmacroeconomics. Federal Reserve Bank of Minneapolis, QuarterlyReview, summer.

31. Sargent, T. J., and Wallace, N. 1975. Rational expectations, theoptimal monetary instrument, and the optimal money supply rule.Journal of Political Economy 83, no. 2 (April).

32. Sargent, T. J., and Wallace, N. 1976. Rational expectations and thetheory of economic policy. Journal of Monetary Economics 2, no. 2(April): 169-83.

33. Statistiches Jahrbuch fur das Deutsche Reich. 1924/25.34. Townsend, R. M. 1980. Models of money with spatially separated

agents. In Kareken and Wallace [8].35. Wallace, N. 1980. The overlapping generations model of fiat money.

In Kareken and Wallace [8].36. Young, J. P. 1925. European currency and finance, vols. 1 and 2.

(Commission of Gold and Silver Inquiry, United States Senate, serial9.) Washington: Government Printing Office.