Angus Maddison did a wonderful job recovering the evolution of real GDP per capita before 1950. Various historical estimates are available now at the website of Groningen University. The estimates after 1950 with the most recent updates are published by the Total Economy Database maintained by the Conference Board.
We have demonstrated real GDP per capita, G, in developed countries follows a linear trend in the long run:
G(t) = At + C (1)
where the slope A and free term C have to be determined by linear regression.
In other words, real GDP per capita has a constant annual increment over the previous 130 to 150 years at least. There is a break between 1940 and 1950. After the break, the annual increment jumped up by a factor of 10 in all developed countries. We have attributed this change to the revision of inflation definition around 1950. By extrapolating the linear regression line obtained before 1940, we have found that the true level of real GDP per capita in the U.S. should be around $11000. The official value is ~$31,000 (Geary–Khamis 1990 US dollars).
Here we apply the same procedure and estimate the true values of real GDP per capita in the U.K., France, Italy, Switzerland and Austria. In five Figures we depict both relevant segments, i.e. from 1870 to 1940 and from 1950 to 2011, and regression lines with their equations. We also reproduce one Figure from our previous post on the US. Table 1 lists the official and true estimates. The difference between these estimates is completely attributed to the mistakes in definition of price inflation.
Table 1. Official and true GDP per capita in GK dollars as of 2011
All in all, the biggest countries have dramatically biased estimates of real GDP per capita. The rate of price inflation (i.e. the GDP deflator) has been underestimated and has affected the perception of real growth.