Pavel Ryska. PCPE, April 18, 2015

Transcription

1 Institute of Economic Studies Charles University Prague PCPE, April 18, 2015

2 Motivation: Deflation has a bad reputation Bernanke (2002): Sustained deflation can be highly destructive to a modern economy and should be strongly resisted. Probable reason Knowledge about deflation is mostly limited to 2 episodes: (1) : e.g. Bernanke (1995), Romer (1992) (2) Japan from 1990 onward: e.g. Krugman (1998) Natural question arises: are these episodes representative of all experience with deflation that we have?

3 Figure: US real GDP and Consumer Price Index (log of index where 1804 = 100) By brief inspection, real GDP and prices do not have a clear-cut relationship.

4 - Two approaches. Deflation as a symptom: 1. Of economic growth: in M v = P Y, with M and v constant, rising Y implies falling P 2. Of recessionary distressed selling and higher money demand Deflation can accompany both growth and recession, but is not their cause. No need to counteract it using monetary policy.

5 - Two approaches. Deflation as a cause of recession: 1. Delay of spending contraction of aggregate demand 2. Fisher s debt-deflation: real value of debt bankruptcies and distressed sales further deflation, etc. both aggregate demand and supply contract 3. Mundell-Tobin effect: in i = r + π, inflation affects r, not only i Extreme case: liquidity trap, where inflation gets fully reflected in r r aggregate demand Scope for monetary policy.

6 Data Output and prices for 19 countries, time span c Overall 2789 observations. Control variables mostly unavailable, only limited investment series. Attention restricted to inflation rates -20 % to 20 %. Model: Unobserved effects Y it = β 0 + β 1 Y it 1 + β 2 P it + β 3 P it 1 + a i + u it a i represents a time-constant unobserved effect for country i. So-called fixed-effect estimation allows any correlation between a i and regressors.

7 with and without control variable (investment). Unconditional Conditional on I Coefficient p-value Coefficient p-value Y t ** *** <0.001 P t P t ** I t *** <0.001 I t *** <0.001 Observations Adj.R Table: Regression of output growth on inflation: All data Inflation is both statistically and economically insignificant for GDP growth. There is practically no relationship between price movements and GDP growth.

8 The is unlike any other episode in the sample. Figure: Regression for (a) and (b) all data except

9 Relatively few annual data to draw conclusions. However, GDP and inflation charts do not show a deflation problem. Figure: Contemporary Japan (1991 = 100): (a) Output and prices, (b) Unemployment and prices

10 In a large dataset, there is no evidence of deflation-recession relationship. Deflation is perfectly compatible with high rates of economic growth. is rather an exception than a rule, and can hardly be used as a standard for judging deflation. There were many special factors in the that we cannot empirically control for.

11 Cited literature Bernanke, B. (1995): The Macroeconomics of the Great Depression: A Comparative Approach. Journal of Money, Credit and Banking, Vol. 27, No. 1. Bernanke, B. (2002): Deflation Making sure it doesnt happen here. Speech before the National Economists Club, Washington. Krugman, P. (1998): Its Baaack: Japans Slump and the Return of the Liquidity Trap. Brookings Papers on Economic Activity, No. 2. Romer, C. (1992): What Ended the? Journal of Economic History, No. 4.