Thursday, April 28, 2016

The Rise and Fall of American Growth

American productivity growth used to be a lot higher than it is now. In the century between 1870 and 1970, and especially in the fifty years between 1920 and 1970, the American increased his productivity at a pace he has not been able to match since. Robert Gordon examines the history and the potential causes of American growth over different periods in American history in The Rise and Fall of American Growth.

The history lessons were fantastic. Gordon describes the many innovations over the last couple of centuries that have sparked changes in how we live, and they are fascinating. Many of these changes made things we already do a lot easier, while others made possible things we could never do before. When the latter took place, the productivity and standard of living estimates don't even capture the benefits.

This leads me to my first quibble with the book. For his productivity measure, Gordon uses Total Factor Productivity (TFP), which is an estimate after subtracting productivity that is estimated from growth in capital and labour skills. So already TFP is a measure on which one can't count on precision, because there are a lot of assumptions that have already gone into its calculation. Then, since TFP is a already an underestimate of growth (because new things are not captured), I don't really know how one can base a whole lot of conclusions on a reliance on these numbers, even if they are supplemented by his supporting qualitative opinions that today's changes in how we live aren't as beneficial as past ones. This is especially true when year to year and decade to decade, productivity measures change rather rapidly.

Gordon then uses the trending history to make estimates of productivity growth in the future. So it felt to me like assumptions are being based on assumptions are being based on even more assumptions, which I found hard to take. Especially because his outlook is so pessimistic. He touches on a number of coming technologies (driverless cars, 3D printing etc) but doesn't see them as being as life-altering as I do.

I would have liked to see him take on some of the standard arguments for why productivity growth may have fallen as of late, if he's going to put forth this own theories. For example, productivity in manufacturing seems to be a lot easier to achieve, and so as manufacturing has fallen as a share of the economy, productivity gains are harder to come by. Also, demographics may play a role, as the elder boomers who are a big share of the population are at stages of their lives where they may no longer be able to improve their productivities like they could in their 20's, 30's and 40's. Neither of these subjects were touched on.

I also found some apparent contradictions that I think betray the author's biases. For example, he celebrates the productivity gains for society as a result of freed up workers who used to chop up ice for ice boxes before the invention of the refrigerator, but laments the loss of manufacturing jobs today as some sort of hollowing out of the middle class. (I really don't understand the obsession many people seem to have with manufacturing jobs. No one sheds a tear for the all the agricultural jobs that were lost as we've gone from 70% working in agriculture to less than 1%.)

It was a heavy read, too. The book clocked in at over 1000 pages, and it was at times a chore to read. I decided to slog through it, however, because the history lessons were worth it for me.


No comments: