학술논문
Long-run purchasing power parity in the 1920s: the Greek experience.
Document Type
Article
Source
Subject
*Purchasing power
*Purchasing power parity
*Foreign exchange rates
*Consumption (Economics)
*Economic indicators
Drachma
*
*
*
*
Language
ISSN
0003-6846
Abstract
A test is made for the existence of long-run purchasing power parity (PPP) for three drachma bilateral exchange rates during the 1920s. Maximum likelihood estimation techniques of cointegration vectors are employed and they produce strong evidence in favour of PPP.
The experience with flexible exchange rates during the 1920s has been proved to be a fertile ground for testing the validity of the purchasing power parity (PPP) doctrine. However, the accumulated evidence from three bilateral exchange rates -- the dollar/pound, the franc/dollar and the franc/pound -- has failed to end the controversy surrounding the PPP theory. On the one hand, studies by Frenkel (1976, 1978, 1980) and Taylor and McMahon (1988) have given support to the PPP hypothesis for the 1920s; on the other hand, studies by Edison (1985), Ardeni and Lubian (1989) and Ahking (1990) have been rather unfavourable to it.
The two-step testing procedure for cointegration, suggested by Engle and Granger (1987), has come under criticism for one main reason. In the multivariate case this procedure may be used to test the null of non-cointegration against one cointegration relation only. However, in this case more than one relation may be present and, hence, another testing strategy should be applied. The long-run PPP hypothesis is tested within the framework provided by the maximum likelihood estimation technique of cointegrating vectors suggested by Johansen (1988). [ABSTRACT FROM AUTHOR]
The experience with flexible exchange rates during the 1920s has been proved to be a fertile ground for testing the validity of the purchasing power parity (PPP) doctrine. However, the accumulated evidence from three bilateral exchange rates -- the dollar/pound, the franc/dollar and the franc/pound -- has failed to end the controversy surrounding the PPP theory. On the one hand, studies by Frenkel (1976, 1978, 1980) and Taylor and McMahon (1988) have given support to the PPP hypothesis for the 1920s; on the other hand, studies by Edison (1985), Ardeni and Lubian (1989) and Ahking (1990) have been rather unfavourable to it.
The two-step testing procedure for cointegration, suggested by Engle and Granger (1987), has come under criticism for one main reason. In the multivariate case this procedure may be used to test the null of non-cointegration against one cointegration relation only. However, in this case more than one relation may be present and, hence, another testing strategy should be applied. The long-run PPP hypothesis is tested within the framework provided by the maximum likelihood estimation technique of cointegrating vectors suggested by Johansen (1988). [ABSTRACT FROM AUTHOR]