Working Paper: CEPR ID: DP473
Authors: Mark P. Taylor
Abstract: In this paper we propose a test of the hyperinflation model of money demand, which is valid under any assumption concerning agents' expectations, subject only to the restriction that forecasting errors are stationary. It is also demonstrated that highly efficient estimates of the model can be obtained, and restrictions on them tested, under the same weak assumption. Finally, it is shown of rational expectation. The arguments are illustrated by analysis of the classic data on European hyperflations previously analysed by Cagan (1956), Barro (1970) and Abel, Dornbusch, Huizinga and Marcus (1979).
Keywords: money demand; hyperinflation; expectations; cointegration
JEL Codes: 044; 211; 311
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| real money balances (E41) | current inflation (E31) |
| inflation expectations (E31) | money demand (E41) |
| cointegration (between real money balances and current inflation) (E31) | validity of hyperinflation model (E31) |
| failure to find cointegration (C22) | rejection of Cagan model (E19) |
| historical data (Y10) | support for hyperinflation model (E31) |
| expansion of money supply (E51) | inflation tax revenue maximization (H29) |
| rational expectations hypothesis (for Russia and Hungary) (D84) | rejection (Y60) |
| rational expectations hypothesis (for Poland) (D84) | acceptance (Y20) |