Student Economic Review, Vol.

22, 2008

GERMAN INFLATION AND THE MONEY SUPPLY,

1919-1923

DERVLA BRENNAN

Senior Sophister

Hyperinflation haunts the dreams of many a central banker. The

German experience post World War I is perhaps the most frequently
cited example of this inflationary nightmare occurring in reality.
In this paper, Dervla Brennan provides some invaluable advice for
any burgeoning economist seeking to become the ‘next’ Alan
Greenspan. A number of theories of hyperinflation are extensively
discussed. As ever, the difficulty is in separating economic facts
from politics. The growth of the money supply is considered to have
been the driving force behind the inflationary crisis, but whether
the supply was endogenous or exogenous is a major point of
contention.

Introduction

‘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 this means when it can live by
no other’.1

According to Milton Friedman, ‘Inflation is always and everywhere a monetary

phenomenon’. Inflations and hyperinflations, by definition and historically, are
periods of rapid trend increases in price combined with a decreasing demand for
real money balances (Holtfrerich, 1986). In the 1920s the word inflation came to
signify money growth rather than a rise in prices (Webb, 1985). Indeed the
German inflation from 1919 to 1923 saw an immense increase in the circulation
of paper money. By the end of November 1923 there were 400,267,640,302
billion notes in circulation compared with 663,200 billion in August 1923
(Sargent, 1981). This was accompanied by a rapid depreciation of the mark. In

1
J.M. Keynes. A Tract on Monetary Reform (1923)

143
 GERMAN INFLATION AND THE MONEY SUPPLY

July 1919, one US dollar could be bought at 14 marks. By November 1923, the
exchange rate was 4,420,000,000,000 marks to the dollar (Berghahn, 1987). It
was claimed by the Allies that the Reichsbank was purposely expanding the
money supply and thus depreciating the mark in order to indicate their inability
to pay reparations. Meanwhile it was claimed in Germany that the currency
depreciation, arising from disequilibrium of the balance of payments, had caused
the expansion of the money supply and the inflation.
What was the primary determinant of Reichsbank money supply
policy? This paper will begin with an examination of the Reich’s fiscal policy
during and after the war. A summary will then be presented of two competing
theories that seek to explain the inflation that existed in Germany at the time: the
‘Balance of Payments’ theory and the ‘Quantity’ theory. The crux of this paper
concerns Cagan’s (1956) seminal work on the demand for money during the
hyperinflation in Germany. Cagan’s paper reopened the debate on what
determined the money supply process in Germany post World War I. Ultimately
this leads on to an examination of whether the money supply was exogenous as
determined by the Reichsbank, or endogenous as determined by the actions of
other agents in the economy, with the Reichsbank playing a passive role.

Reich Fiscal Policy, 1914-1923

Germany relied on printing money instead of levying taxation to finance their

participation in World War I. This is the most inflationary form of war finance
(Holtfrerich, 1986:107). However, inflation is a form of tax which acts
unbeknownst to the public. As Keynes remarks: ‘[inflation] is the form of
taxation which the public find hardest to evade and even the weakest government
can enforce, when it can enforce nothing else’ (Keynes, 1923: 37). The economic
cost of an inflationary tax is the weakening power of the currency to carry out its
functions (Holtfrerich, 1986: 121).
Following the war the main economic question facing the German
government was whether they should restore the mark to its pre-war value or
maintain the value it had obtained by the end of the war (ibid). Germany chose
the latter, continuing with a policy of price inflation and exchange rate
depreciation, and thus also the inflation tax. This particular policy option is
reflective of the centrality the budget deficit had come to hold in currency debates
after the war (ibid). Ideally a tax reform was needed in order to meet the Reich’s
expenses. However, the government was too weak to ask such sacrifices from its
people, especially after having lost the war. Increasing the money supply was the

144
 DERVLA BRENNAN

easiest means of meeting fiscal requirements (Bresciani-Turroni, 1937). Haller

remarks: ‘If the state did cause the inflation, it did so in self-defence’ (Holtfrerich,
1986: 137).
Furthermore, Germany’s total financial obligation under the Treaty of
Versailles was for a long time unknown, leaving fiscal policy uncertain, and this
uncertainty destabilized the value of the mark (Sargent, 1981). Threatened by
allied penalties on the one hand and by collapse of the Reich on the other, no
German government could have fulfilled the reparation demands in any way other
than by printing money (Holtfrerich, 1986). Germans have always cited the terms
of the Treaty of Versailles as the reason for the inflation. Indeed both ‘Balance of
Payments’ and ‘Quantity’ theorists saw reparations as the disequilibrating factor,
acting either on the balance of payments or on the fiscal balance, and thereby
hastening the inflation and currency depreciation (ibid). This prevalent belief
prevented any consideration of monetary reform, which may have impeded
progress in tackling the inflation.
In November 1923 prices suddenly stopped rising, the mark suddenly
stopped depreciating, the government put an end to borrowing from the central
bank, the budget swung into balance and consequently the inflation came to an
end. Sargent argues that a permanent change in the fiscal and monetary policy
regimes was what was required to end the inflation (Sargent, 1981). Indeed the
abrupt change outlined above was facilitated by two such simultaneous
measures. The monetary reform involved the creation of a new currency; the
retenmark,2 and a new note issuing bank; the Retenbank. This bank was restricted
both in the total volume of retenmarks it could circulate and, more importantly,
on the amount of credit it could extend to the government for financing the deficit.
The fiscal reform entailed implementing a new tax system and curtailing
expenditure. Under the Dawes plan Germany’s reparations payments were
reviewed and reduced to a more manageable sum (Sargent, 1981). Thus earlier
attempts to end the hyperinflation had failed because the government did not
revise their fiscal policy of financing the deficit through increments of an
unbacked money supply.

Balance of Payments Theory versus Quantity Theory

During the war itself, and especially in the aftermath, a debate broke out in
political and academic circles regarding two theories of inflation. ‘Balance of

2
1 retenmark = 1 trillion paper marks.

145
 GERMAN INFLATION AND THE MONEY SUPPLY

Payments’ theorists cited the passivity of the German external balance caused by
reparations and the other exactions under the Treaty of Versailles as the cause of
the depreciation of the mark and the ensuing inflation. The depreciation of the
mark was thus not the consequence but the cause of the budget deficit, money
supply expansion and the inflation (Bresciani-Turroni, 1937; Holtfrerich, 1986).
According to Helfferich:

‘First came the depreciation of the German currency by the

overburdening of Germany with international liabilities and by the
French policy of violence…inflation is not the cause of the rise in
prices and of the depreciated currency, but the latter is the cause of
the higher prices and of the greater volume in the issue of paper
money…’(Laidler and Stadler, 1998: 820).

‘Quantity’ theorists, on the other hand, saw the continued issues of paper
money to finance the budget deficit as the fundamental cause of the inflation and
currency depreciation (Bresciani-Turroni, 1937; Holtfrerich, 1986). Robinson
argues that theoretical discussion of the German inflation was for some time
clouded by political prejudices, with the German writers blaming reparations and
the collapse of the exchange rate, and the Allies blaming the budget deficit and
creation of money (Robinson, 1938). It is claimed that while the Reichsbank
publicly held the ‘Balance of Payments’ theory as the root cause of the inflation,
they essentially adhered to the ‘Quantity’ theory in their confidential
correspondence with the government.

Recent Theories of Hyperinflation

Cagan’s 1956 Model

‘Hyperinflation provides a unique opportunity to study monetary
phenomena’ (Cagan, 1956: 25).

In Germany, hyperinflation is defined as beginning in August 1922 and ending

in November 19233. Real cash balances fell over the whole period of
hyperinflation but they fluctuated drastically from month to month. Cagan’s paper
proposes and tests a theory that accounts for this erratic behaviour of real cash

3
Cagan defines hyperinflation as: ‘beginning in the month the rise in prices exceeds
fifty per cent and as ending in the month before the monthly rise in prices drops below
that amount and stays below for at least a year’(Cagan,1956: 25).

146
 DERVLA BRENNAN

balances (ibid). By allowing for lags4 and assuming an exogenous money supply,
Cagan hypothesizes that variations in the expected rate of inflation5 account for
variations in real cash balances during hyperinflation. This is the case only where
expectations are formed adaptively: that is by extrapolating past rates of inflation
into the future. This implies a dynamic process in which the hyperinflation of
prices was caused by both past and current changes in the quantity of money.
Cagan concludes that: ‘domestic monetary factors alone explain hyperinflation’
(ibid: 90). He attributes this tremendous increase in money and prices to the
fiscal needs of the government:

‘Issuing money was a method of raising revenue by a special kind

of tax – a tax on cash balances…All that is required is to spend
newly printed notes. The resulting inflation automatically imposes
a tax on cash balances by depreciating the value of money’ (ibid:
78).

However, Cagan’s model is incapable of explaining the final months of the

German hyperinflation, for which it predicts lower real money balances than those
actually observed from August through November 1923.6 He offers two
explanations that may account for these outlying observations. First, he
hypothesizes that rumours of currency reform led people to expect that prices
would stop rising after a certain number of months. This encouraged people to
hold higher real cash balances than they would otherwise have held in view of the
current expected rise in the price level. The second explanation Cagan offers is
that he had used the wrong functional form in deriving his money demand
function7 (ibid). Cagan’s seminal paper on the demand for money under
hyperinflation has thus triggered an investigation into the German
hyperinflation. In particular, various attempts have been made to explain the final
months of 1923 when, contrary to conventional theory, real money balances
increased despite accelerating inflation (Flood and Garber, 1980).

4
‘The large changes in the balances during hyperinflation correspond to large changes
in the rate of price change with some delay, not simultaneously.’ (Cagan, 1956: 88)
5
This is equivalent to the rate of depreciation in the value of money or a decline in the
purchasing power of nominal money balances.
6
The period of maximum inflation and maximum money creation.
7
Frenkel (1977) in his study of the German hyperinflation rejects the possibility that
Cagan had used the wrong functional form in deriving money demand.

147
 GERMAN INFLATION AND THE MONEY SUPPLY

Flood and Garber: Hyperinflation and Rational Expectations

Flood and Garber (1980) combine Cagan’s model of hyperinflation with rational
expectations instead of adaptive expectations in order to explain the final months
of the German hyperinflation. They extend the standard monetary theory and
include observations from August through November 1923. The purpose of this
was to test Cagan’s hypothesis that agents expected a monetary reform toward the
end of the hyperinflation and thus began to hold higher cash balances despite
increasing inflation. True to the spirit of rational expectations, Flood and Garber
argue that when people anticipated a permanent change in the money supply
regime, they changed their expectations of future inflation to reflect the change
in monetary policy. They calculated the probability of monetary reform with the
following result: ‘…our measure of the probability of monetary reform…hit its
peak during the very week [November 15, 1923] commonly asserted to mark the
beginning of the reform’ (Flood and Garber, 1980: 48). Thus they conclude that
expectations of monetary reform were the factor responsible for increasing real
cash balances.

Webb: Endogenous Money Supply

Webb (1985, 1986) asserts: ‘the primal cause of the money growth and the whole
inflation was the growth of the government debt’ (Webb, 1985: 490). In his
analysis, Webb treats the money supply as endogenous rather than exogenous.
He argues that one should focus on the determinants of money supply rather than
money demand. The money supply depended on the government debt and how
much of this debt the public decided to monetize. This fraction of the debt to be
monetized was in turn influenced by inflationary expectations and on public
confidence in German finances. Thus government debt and inflationary
expectations indicating an endogenous money supply explain the money supply
better than the assumption of exogeneity (Webb, 1986). Webb contends that it
was positive ‘fiscal news’ combined with rumors of monetary reform in August
1923 that increased the possibility that the government would cease to run
deficits. A new government came to power mid-August and succeeded in
increasing revenue despite the increasing inflation. In September, the
government declared an end to passive resistance which further lowered
government expenditures (ibid). In addition, from August 1923 it seemed likely
that the Reichsbank was planning to institute a new policy regime. The
Reichsbank’s refusal to discount commercial bills of more than a month’s
duration was combined with an obligation to repay in terms of gold. They also
sent a private memorandum to the government informing them that after the end
of the year, the Reichsbank would cease to monetize any more government debt.
This was the first time the national bank threatened to use its supposed autonomy

148
 DERVLA BRENNAN

from the government. While this announcement was not made public, Webb
contends that it must have been known in the upper circles of the business
community that sat on the Direktorium and Aufsichtsrat of the Reichsbank (ibid).
Thus both signals of fiscal reform and monetary reform played a role in
lowering inflationary expectations towards the end of the hyperinflation.

Tullio’s Dynamic Model

The debate at this point rests on the question of whether the money supply
during the inflation was exogenous or endogenous. Tullio (1995) develops a
dynamic model of the German hyperinflation, simultaneously explaining prices,
the exchange rate, and money supply. In his model the money supply is partially
endogenized, and expressed as a function of nominal income and of the deviation
of actual from potential output. The stability of the model indicates that the
German inflationary process was primarily caused by fiscal deficit and excessive
monetary growth (Tullio, 1995). Tullio argues that there is some evidence indi-
cating that the Reichsbank passively financed increases in prices at times of low
economic activity in order to avoid a recession, allowing for the partial
endogeneity of the money supply (ibid). This partial endogeneity implies the
cause of the inflation was at times from prices to money. These results support the
hypothesis that the Reichsbank followed this real-bills doctrine, passively
financing increased demands for money and credit. In 1991, Cagan himself
returned to the topic and concluded:

‘the money stock cannot be treated as exogenous. A plausible way

to endogenize the money stock is to model the revenue needs of the
government for an inflation tax’ (Laidler and Stadler, 1998: 822).

Conclusion

The growth of the money supply is a prominent feature in every account of the
German inflation. While money growth was frequently cited as a consequence
and sometimes as the cause of inflation, recent models of inflation, and in
particular hyperinflation, treat the growth of money as an immediate and often as
an exogenous cause of inflation. In these models inflationary expectations have
come to play an important intermediary role (Webb, 1985). Inherent in these
formulations is a denial of the possibility that inflation has a momentum of its
own. Rather it is the long-term government policy of persistent budget deficits
and high rates of money creation which gives momentum to the inflation rate
(Sargent, 1981). However, these models that accord a role to expectations in the

149
 GERMAN INFLATION AND THE MONEY SUPPLY

inflationary process are by no means perfect. There is, as yet, no ideal way to
model expectations. While the rational expectations formulation is certainly
useful, it is based on very strong assumptions and so comes with certain
limitations.
Kiguel (1989) notes that there was an element of bidirectional causation
during the inflation. High rates of inflation decreased the real value of government
revenues, thus increasing their financial requirements. The Government then
financed this deficit in revenue by increasing the money supply, reinforcing the
inflationary process (Kiguel, 1989). Thus at times the inflation did run from prices
to money. What is certain is that the inflation was caused by the acceleration of
money growth to finance the large accumulation of government debt. This
relentless printing of money was the only means of survival for a weak
republican government burdened with the charge of having accepted the
humiliating Treaty of Versailles. The exact nature of the money supply still
remains a topic of debate. However, contentions of an endogenous money
supply are dubious in the German case. The Weimar government made a
conscious decision to finance the budget deficit by printing money, thus
implying an exogenous money supply.

Bibliography

Berghahn, V. R. 1987 [1982]. Modern Germany: Society, Economy and Politics

in the Twentieth Century. Cambridge: Cambridge University Press.

Bresciani-Turroni, C. 2003 [1937]. The Economics of Inflation: a Study of

Currency Depreciation in post-war Germany. London: Routledge.

Cagan, P. 1956. ‘The Monetary Dynamics of Hyperinflation’ in M. Friedman

(ed.) Studies in the Quantity Theory of Money. Chicago: University of Chicago
Press.

Flood, R.P. and Garber, P.M. 1980. ‘An Economic Theory of Monetary
Reform’. The Journal of Political Economy 88:1:24-58.

Frenkel, J. A. 1977. ‘The Forward Exchange Rate, Expectations, and the

Demand for Money: The German Hyperinflation’. The American Economic
Review 67:4:653-670.

150
 DERVLA BRENNAN

Holtfrerich, C. 1986. The German Inflation 1914-1923: Causes and Effects in

International Perspective. Berlin: de Gruyter.

Keynes, J. M. 1971 [1923]. The Collected Writings of John Maynard Keynes

Vol.4: A Tract on Monetary Reform. London: Macmillan.

Kiguel, M. A. 1989. ‘Budget Deficits, Stability, and the Monetary Dynamics of

Hyperinflation’. Journal of Money, Credit and Banking 21:2:148-157.

Laidler, D.E.W. and Stadler, G.W. 1998. ‘Monetary Explanations of the

Weimar Republic’s Hyperinflation: Some Neglected Contributions in
Contemporary German Literature’. Journal of Money, Credit and Banking
30:4:816-831.

Robinson, J. 1938. ‘Review of Bresciani-Turroni’s The Economics of

Inflation’. Economic Journal 48:191:507-513.

Sargent, T. J. 1981. ‘The Ends of Four Big Inflations’ in R.E. Hall (ed.)
Inflation: Causes and Effects. Chicago; London: University of Chicago Press.

Tullio, G. 1995. ‘Inflation and Currency Depreciation in Germany, 1920-1923:

A Dynamic Model of Prices and the Exchange Rate’. Journal of Money, Credit
and Banking 27:2:350-362.

Webb, S. 1985. ‘Government Debt and Inflationary Expectations as Determi-

nants of the Money Supply in Germany 1919-1923’. Journal of Money, Credit
and Banking 17:4:479-492.

Webb, S. 1986. ‘Fiscal News and Inflationary Expectations in Germany After

World War I’. Journal of Economic History 46:3:769-794.

151