German Hyperinflation Research Articles

A Multivariate I(2)cointegration analysis of German hyperinflation

This paper re-examines the Cagan model of German hyperinflation during the 1920s under the twin hypotheses that the system contains variables that are I(2) and that a linear trend is required in the cointegrating relations. Using the recently developed I(2) cointegration analysis developed by Johansen (1992, 1995, 1997) extended by Paruolo (1996) and Rahbek et al. (1999) we find that the linear trend hypothesis is rejected for the sample. However, we provide conclusive evidence that money supply and the price level have a common I(2) component. Then, the validity of Cagan's model is tested via a transformation of the I(2) to an I(1) model between real money balances and money growth or inflation. This transformation is not imposed on the data but it is shown to satisfy the statistical property of polynomial cointegration. Evidence is obtained in favour of cointegration between the two sets of variables which is however weakened by the sample dependence of the trace test that the application of the recursi...

A simple procedure for detecting periodically collapsing rational bubbles

This paper proposes a new procedure for detecting the presence of periodically collapsing rational bubbles via an analysis of the properties of the relevant observable time series. The procedure is based on random-coefficient autoregressive models. An empirical application of the procedure to German hyperinflation data is examined and discussed.

Stock returns during the German hyperinflation

We examine the relationship between stock returns and inflation during the German hyperinflation period. As Cagan (1956) argues, hyperinflation is close to a pure monetary phenomenon, the monetary sector is almost isolated from the real sector of the economy. We can thus study the impact of inflation on common stock returns directly, unaffected by the potential interactions of other economic variables such as expected real output. The negative relationship between inflation and stock returns widely reported in the literature may be spurious because they have not made such a distinction. By using forward exchange rate premium as a proxy for expected inflation, we find that stock returns, inflation, expected inflation, and unexpected inflation are cointegrated during the German hyperinflation period. The results also indicate that the fundamental relationship between stock returns and both realized and expected inflation is highly positive. JEL classification: E310; G120

A structural time series test of the monetary model of exchange rates under the German hyperinflation

Abstract In this paper some structural time series evidence is presented for the monetary model of exchange rates under the German hyperinflation. The results show that the money supply and the expected rate of change of the exchange rate played a dominant role in the determination of the exchange rate of the mark against the US dollar. The results also support the property of proportionality between the exchange rate and the money supply. Furthermore, evidence is found for the validity of the constituent components of the monetary model, PPP and the quantity theory.

Non‐Linear Dynamics of Inflation in High Inflation Economies

Attempts by governments to finance a substantial proportion of expenditure by seigniorage can lead to multiple inflationary equilibria. Theoretical models suggest that, in these circumstances, inflation follows a non‐linear process with up to three steady states and that the stability characteristics of these depend on the process by which expectations are formed. In this paper we show that the exponential smooth transition autoregression (ESTAR) model is capable of exhibiting the required characteristics and so provides a suitable vehicle for analysing inflation in high inflation economies. We estimate ESTAR models for three well‐known inflationary episodes—the German hyperinflation of the early 1920s and post‐Second World War inflations in Argentina and Brazil. Our results imply that, during the periods in question, each of these economies possessed a stable low‐level equilibrium rate of inflation but that the variances of inflation shocks were large enough to drive each economy into a high inflation state. For Brazil, this high inflation state is stable around a particular value but in the cases of Argentina and Germany the high inflation state is characterized by inflation cycles.

Testing the currency-substitution model under the German hyperinflation

In this paper the currency-substitution model is tested under the German hyperinflation using several expectations-formation mechanisms. The maximum-likelihood estimates of the currency-substitution model reveal that extrapolative and adaptive expectations seem to have been predominant and that there was a significant degree of currency substitution. The results also reveal that expectation was destabilizing and that it was not possible to distinguish between the effects of the expected change in the exchange rate and the expected inflation.

The statistical skeleton of the demand for real money balances during the German hyperinflation episode January 1921-November 1923

The aim of this study is to re-examine the common belief that the existence of the demand for real money balances during the German hyperinflation episode 1921:01–1923:11 is conditioned on truncating the data on June/July 1923 or on the omission of the first 18 months. It is demonstrated in this study that a stable relation can be found for the period 1921:03-1923:11 through taking into account the nonlinear functional relationship amongst the variables of the cointegrating regression. Using a variant of Phillips-Loretan ECM it is shown that, while the equilibrium error is forecastable on the basis of lagged change of real wage and the variability of inflation alone, the adjustment of actual to desired real money balances is, indeed, instantaneous as predicted by Cagan.

The Monetary Model of Exchange Rates in the History of Economic Thought

This paper traces the origin of the monetary model of exchange rate determination to three controversies surrounding three events in the European economic history. These events are the 1745 shift by Sweden to currency inconvertibility, the suspension of the convertibility of the pound during the Napoleonic wars, and the German hyperinflation of the 20th century. While the first controversy gave rise to the linkage between some forms of the quantity theory of money and purchasing power parity, the second controversy led to the refinement and development of these concepts. The third controversy introduced the role of expectations in the monetary model.

Market Fundamentals versus Speculative Bubbles: A New Test Applied to the German Hyperinflation

We develop and apply a method of testing for speculative bubbles. The method is designed to overcome two well-known problems in the identification of bubble phenomena—the problem of distinguishing any type of bubble from an expected future change in market fundamentals and the problem of detecting a periodically-collapsing bubble when the residuals of the fundamentals regression are integrated. We propose the strategy of estimating a switching regime model of market prices, partialling out expected changes in fundamentals and carefully analysing the properties of the residuals. Extending our analysis, we also propose a more direct test for bubbles, based on the estimation of the general (fundamentals-plus-bubble) solution for market prices. We apply our methodology to the study of German hyper- inflation in the 1920s. We find evidence consistent with the existence of a bubble during that hyperinflation.

Hyperinflation, the exchange rate and endogenous money: post-World War I Germany revisited

We re-examine the interaction of domestic and external factors in the German hyperinflation by extending Frenkel's monetary model of the exchange rate to incorporate: (1) endogeneity of the money supply with respect to government debt, expected inflation and money wages; (2) the direct impact of exchange rate depreciation on the domestic price level; and (3) endogenous wage determination. Four-equation estimates support each extension of the monetary model. It appears that government deficit finance, the collapse of the exchange rate and the monetary transmission of the wage-price spiral were all crucial contributors to the price explosion.

The time series behaviour of spot exchange rates in the German hyper-inflation period: (Was the process chaotic?)

A discrete time approximation to the continuous time hyper-inflation model of Sargent and Wallace (1993) in which the authorities finance a given budget deficit by printing money appears to admit the possibility of chaotic solutions. In this paper we investigate the time series properties of daily observations on the Pound Reichsmark spot exchange rate in the inter-war hyper-inflation period. Our empirical analysis is suggestive that spot rates were generated by a non-linear possibly chaotic process.

Inflation and Currency Depreciation in Germany, 1920-1923: A Dynamic Model of Prices and the Exchange Rate

This paper presents a dynamic model of the German hyperinflation, explaining prices, the exchange rate and the money supply. In the model estimated simultaneously, the nominal exchange rate is allowed to diverge from Purchasing Power Parity in the short run, the monetary approach to exchange rate determination and the quantity theory of money are assumed to be valid in the long run, the dynamic adjustment of prices and the exchange rate to monetary disturbances is made explicit, and currency substitution between domestic and foreign currency is introduced explicitly. The problem of the autocorrelation of the residuals of the money demand function found in previous studies is resolved, the fit of inflation and exchange rate changes is very satisfactory, and the speed of adjustment of the exchange rate to monetary disturbances is found to be much higher than for prices. Copyright 1995 by Ohio State University Press.

A test for rational expectations when some variables are I(2)

This paper exploits the result that in the case of I(2) processes the precise dating of variables is crucial when empirically testing for cointegration. The data for the latter part of the German hyperinflation episode exhibit I(2) behaviour. We utilize these data to discriminate between alternative expectations hypotheses in the demand for money.

Comments: Rational bubbles during Poland's hyperinflation: Implications and empirical evidence by M. Funke, S. Hall and M. Sola

A number of papers have tested for rational bubbles in the post-World War I hyperinflation in Germany. The present paper henceforth FHS does so in Poland around the time of the ‘big bang’. It finds some evidence that ‘the main determination of the hyperinflation was a rational bubble’ but cautiously notes that conventional structural models might also explain the hyperinflation. I salute their technically sophisticated investigation of this interesting question, but in the end find their evidence distinctly less suggestive of bubbles than do they. This evidence comes from a nice extension of techniques suggested by Diba and Grossman (1984) and Hamilton and Whiteman (1985), and applied to the German hyperinflation by Hamilton and Whiteman (1985). Using the Cagan model, that paper noted that in the absence of bubbles, prices will have the same order of integration as does money; if money is, say, Z(2) so that what is stationary is changes in the growth rate of money but not the growth rates themselves then prices should be Z(2) as well. But bubbles cause explosive growth, so no level of differencing will induce stationarity in a variable driven wholly or even in part by a bubble. Hamilton and Whiteman (1985) found that prices and money appeared to be Z(2) in the German hyperinflation, and concluded that bubbles probably were not present. Evans (1991) in turn pointed out that if (a) bubbles periodically grow and collapse, and (b) there are not a whole lot of observations during which the bubble was growing, the Hamilton and Whiteman (1985) test will likely fail to find bubbles. And tests are likely to find that prices and money are

Was the money demand during the German hyperinflation time varying?

Two versions of Cagan's model are estimated to analyse whether or not the money supply was endogenous during the German hyperinflation. The first version is the original version of Cagan's model with rational expectations. The second version, which is called the alternative version, builds on Cagan's model by considering that the semi-elasticity of the demand for money is a time-varying parameter. The estimation results for both models support the hypothesis that the alternative version provides a better fit to the German hyperinflation data than the original version, and that the money supply was endogenous.

The German Hyperinflation and the Demand for Money Revisited

This paper shows that a remarkably well-defined demand for real balances exists for the German hyperinflation episode, including the final months which have previously been considered as outliers. The authors' econometric analysis exploits the theory of cointegration, given the obvious nonstationarity of the data. Further, they consider two likely sources of misspecification: income variability and the necessity to distinguish between the high inflation and hyperinflation episodes. Copyright 1994 by Economics Department of the University of Pennsylvania and the Osaka University Institute of Social and Economic Research Association.

A model of speculative efficiency with ‘news’ error components

This paper proposes a theoritical model to assess of ‘news’ on the spot-forward-exchange-rate relationship and thereby predicts of the impact of ‘news’on the exchange rate. Two types of ‘news’ are analysed: rational expectations type innovations and announcement effects form newspapers. By using historical data form a time period which includes the German hyperinflation of the early 1920s, an estimate from of the model is derived which is substantially simpler than the version requried using contemporary data. Preliminary results support the main theoretical predictions but also cast some doubt on a literal interpretation of the exchange rate model plus rational expectations.

Cointegration and Cagan's Model of Hyperinflation under Rational Expectations

When money and prices are integrated of order two, Philip Cagan's (1956) model of hyperinflation will give rise to two different levels of cointegration. In most of the literature on the German hyperinflation, one of these cointegrating relationships has been ruled out a priori by assuming that velocity shocks follow a random walk. Using data from the German hyperinflation, the author finds this assumption to be unjustified. Based on a straightforward extension of the cointegrated VAR-approach suggested by John Y. Campbell and Robert J. Shiller (1987), a method to evaluate the Cagan model under rational expectations and no bubbles is proposed and implemented to the German data. Copyright 1993 by Ohio State University Press.

REAL WAGES AND DISTRIBUTIONAL CONFLICT IN THE GERMAN HYPERINFLATION*

Real wages during the German hyperinflation are analyzed in terms of the conflict between workers' demands and the pricing and employment decisions of firms. This framework incorporates the pressure effects of expected inflation and the wage gap between workers' target real and its observed value. Both of these effects are conditioned by unemployment that limits workers' ability to obtain their desired money adjustments. Empirical estimation over 1920-23 provides strong support for the hypothesized interaction between unemployment and real growth over the hyperinflation period. Copyright 1993 by Blackwell Publishers Ltd/University of Adelaide and Flinders University of South Australia

Testing for sunspot equilibria in the German hyperinflation

Testing for sunspot equilibria in the German hyperinflation