Monday, December 2, 2013

The Comprehensive Revision of Real GDP


Graph #1 shows the most recent pre-revision figures for real GDP, as a percent of the most recent post-revision figures. The pre-revision numbers have been rebased to 2009 to match the post-revision values:

Graph #1: The Old Measure of Real GDP as a Percent of the New Measure
Starts out near 100%, In other words, there is almost no difference between pre- and post-revision numbers, for the late 1940s. But by the recent years, the difference is more than four percentage points of GDP.

But I think the graph has things backwards. Let's not look at the old numbers as a percent of the new numbers. Let's look at the new numbers as a percent of the old. This way, we can think of the old numbers as the "given" and we can see the effect of the changes brought by the revision.

Graph #2: The New Measure of Real GDP as a Percent of the Old Measure

Starts out just about equal, as before. But there is a growing difference between the old and new, and a persistent increase in the "gain" resulting from the Comprehensive Revision of 2013. Recent numbers show the gain reaching nearly 5%, over the old series, and trending up rapidly. The trend suggests that the new numbers may soon be 6 or 7 or 8% more than what the old measure would have shown.

The difference becomes significant.

2 comments:

jerry said...

Here's maybe another incoming revision!
http://www.bbc.co.uk/news/business-23501037

The Arthurian said...

Thanks for the link.

When people have an agenda (like wanting GDP to look bigger) change of any kind should be immediately suspect.

One "advantage" of changes like revising GDP is that after a few decades they can look back and say "Oh the old numbers are not compatible" and then they can just stop looking at numbers from before the global financial crisis. Sort of like FRED looks at almost nothing before 1947.

//

From the article: "Drug patents will now be included in the GDP calculations"

Yeah... But are we not told that the reason for the high price of medicines in the US is all of the R&D that goes into developing the new medicines?

In other words, the cost of the R&D is included in the final price paid for the medicines.

So, counting the R&D in GDP is double-counting. So it is wrong.

When people have an agenda (like wanting GDP to look bigger) change of any kind should be immediately suspect.

Thanks for the link.