So another big beast of the global economic commentariat has produced his assessment on Robert Gordon’s take on whether US growth is over. Martin Wolf at the FT joins Paul Krugman in finding Gordon’s hypothesis plausible. Some more sceptical reactions were John Cochrane’s and Bill Easterley, who tweeted it was nonsensical.
At a personal level I was already predisposed to find Gordon’s hypothesis persuasive, having been won over by Tyler Cowen’s Great Stagnation argument. Given that I enjoy a good historical read I am now also looking forward to Gordon’s Beyond the Rainbow: The American Standard of Living since the Civil War. However, I am a bit surprised at how little scrutiny Gordon’s handling of the data has received. The core chart he presents is this one:
Now, I have tried to replicate his nice steps using Maddison’s data. It is difficult. He certainly must have selected his year brackets very carefully to get such a neat downward step pattern. Even after trying to smooth the data by taking the rolling 20-year average per capita growth rate I get the following much messier picture:
A number of additional thoughts emerging from the data and my chart:
(1) Overall, despite the messiness of the data Gordon’s hypothesis of declining trend growth still seems plausible.
(2) The UK was certainly not the frontier economy from 1300. The charge was first led by the Italian city states and the baton is then taken by the Netherlands.
(3) Not shown in the chart is the fact that, according to Maddison’s data, growth between 1300 and 1750 was already a significant uptick from the previous 1000 years.