“The Great Stagnation” may have a flawed premise

Tyler Cowen’s e-bookette, The Great Stagnation is being debated around the various blogospheres, even by people who haven’t read it. I do dig the way it exploits the format of ebooks, being allowed to be longer than an essay, but not padded out into a book. A huge number of best sellers based on essays (Like The End of History) would have benefited from being able to do the same.

But on the substance of his argument, I have a question about the premise.

In short, Cowen’s argument is that for the past 30 or so years there has been little progress in the US and other rich economies. He also uses the kitchen test (noting the lack of progress in the average kitchen over the past decades compared to progress in earlier decades) that Robert Merkel in the link above describes as bunk but, as Labor Outsider in the comments protests, he does prefer median income and other statistical measurements. He attributes this slowdown in progress to the absence of low hanging fruit. Innovation is harder (you can’t invent nanotechnology in a garage) these days, we’re already educating everyone who can be educated etc. Countries that are progressing faster are running in the slip stream of advanced countries by adopting their ideas. When they catch up they too will slow down, just like Japan.

The does seem plausible. But then again, there’s a large number of places in history where I may have thought it plausible. There’s the infamous quote by Duell in 1899, “Everything than can be invented has been invented” and the lower classes always seemed like clay unsuitable to be be moulded by education until they were educated. And as Robert discusses, the Kitchen Test, if not bunk, is rather shoddy.

I have a large question about the data. Cowen says he is talking about a slowdown compared to the growth that followed the Industrial Revolution up to the early 1970s. But this is not the data he shows. He actually compares the past 37 years to the period 1947 to 1975. There’s good reason to do this because the data is better. But I think it is misleading. China’s growth is faster than the US because of catch up, but so was the post war boom.

This is somewhat similar to the logic of Friedman’s plucking model¹. The magnitude of a boom is determined by the magnitude of the bust that precedes it. The reason being the capacity of the economy has stayed the same. You’re merely catching up to where you were. Now if innovation and progress were continuing even whilst the bust meant there was no growth (like we assume with Okun’s Law), this means that not only can you catch up to where you were, but the new capacity is even greater than when you went into the bust. You grow rapidly by picking the fruit that had grown in recession, but which you were prevented from picking.

And what happened before 1947? The Great Depression and World War II. Imagine how much fruit we had left unpicked during 18 years of distraction. We had a very large way to catch up. Cowen has remarked a few times on his blog that the 30s was still an innovative time, and his essay quotes a Charles I Jones who states that most of the growth in that period was from “previously discovered ideas”. He should be aware this is a question that needs to be addressed.

It implies that the slow down he is attempting to explain may be of little consequence in technological terms, since the post war boom was macroeconomic noise over a long term trend. On the data he uses, we don’t really have a case that long term growth in the capacity of progress has actually decreased. The “Great Stagnation” might just be a period where yearly growth stayed fairly close to trend.

I’ll also note that the end of this period in the 1973 should have seen a large bunch of low hanging fruit emerge. This was the era when previously unexploited labour was entering in the workforce; Women and Baby Boomers. With these people no longer dependent (in terms of paid labour), the low hanging fruit hypothesis should have expected a boom in output per capita or median income.

I thought I might be able to do a comparison with what Cowen would consider an era of low hanging fruit for a country not playing catch up (and where easily stolen land wasn’t available). I took data from here for Britain in the 19th century. This is an era where Britain was certainly the leader in progress, and this was the heyday of the industrial revolution. If a country was picking low hanging fruit, it was here.

But the growth is relatively modest. I used GDP per capita (median income, Cowen’s preferred measure is not available, but since gains skew towards the wealthy, GDP per capita will be overstated in comparison). I took two random 30 year periods, 1830 to 1860 and 1870 to 1900 (these also happen to be periods of peace).

From 1830 to 1860 the mean annualised growth rate in GDP per capita was 1.65% and varied between 5.7 %and -3.1%.

From 1860 to 1890 the mean annualised growth rate in GDP per capita was 1.17% and varied between 5.7% and -3.2%.

The last three to four decades in America no longer look so bad!

Cowen does make a plausible case that the figures still overstate recent progress, but I think this is still something he needs to address.

¹ I say similar because this is something I’ve heard about rather than actually read, but the logic stands on it’s own.

1. fn1[↩][↩]