The graph above shows a startling feature about German data: a huge increase in GDP and employment in 1991. Anybody with a little background in history will immediately recognize that this is not due to an error in the data or a sudden increase in German productivity: It all has to do with the reunification of East and West Germany after the fall of the Berlin Wall. While not as spectacular as this one, other changes in the definition of data, including the redrawing of geographic boundaries, do happen and one needs to be careful interpreting the data in those cases. Sometimes, statistical offices go back to the underlying data and recompute the times series using the new definition. For the German case, we have a new graph below that uses data from the OECD that is recalculated all the way back to 1970. Now we do not see the huge jump; rather, we notice how the large increase in unemployment in the eastern parts of Germany after reunification had a significant downward impact on employment. Quite a different story from the first graph.
How these graphs were created: For the first graph, search for “Germany” and click on the “discontinued” tag in the side bar (you may have to expand the list). Check the two series you want and choose “Add to Graph.” Expand the menu for the second series, set the y-axis to the right side. Finally, add the vertical line by adding another series, selecting “Trend line” from the dropdown menu, setting both dates to 1991-01-01 and putting start and end values that make the line appropriately long. For the second graph, search for “Germany GDP,” select the annual index series for consistency with the previous graph. Then add the second series by searching for “German civilian employment.” Set the y-axis to the right and add the trend line again.
Suggested by Christian Zimmermann
View on FRED, series used in this post: