RDP 9203: Real Exchange Rates and the Globalisation of Financial Markets Charts

where I is total non-government investment, Y is GNP/GDP, S is equal to when S^T is total savings, S^G is government. saving and is an estimated Ricardian effect. Estimating this equation using only cross-section data on the countries in the sample, the United States, Japan, Germany, France, Italy, the United Kingdom, Canada, Australia and Switzerland does not yield a sufficient number of observations to provide reliable estimates of Φ. Thus pooled time-series-cross-section data are employed. A constant inventory of 40 observations (10 countries by 4 quarters) were maintained in the sample with quarter t's estimate of , obtained by adding the quarter' s values of the relevant variables for all the countries and deleting those for quarter t−4. The discontinuous lines represent two standard error estimates for the .