Is U.S. Economic Growth Over? Faltering Innovation Confronts the Six Headwinds by Robert J. Gordon questions the belief that economic growth is a never ending phenomenon. He presents a history of economic growth and demonstrates how periods of expansion are driven by industrial revolutions. He states that the first industrial revolution started around 1750 with the development of steam power and the railroad. The second began in 1870 and was brought forth by electricity, the internal combustion engine, and running water. The third and final revolution was instigated by computers, cell phones, and the internet. He makes the point that before first industrial revolution, there was almost no worldwide economic growth, and that since that time, the global economy has expanded at a tremendous pace which causes many to believe that growth is the norm in a modern economy. Gordon however, ponders the idea that the past two to three hundred years have been the exception rather than the rule. He observes that the most recent industrial revolution inspired a much shorter period of growth than the first two. The first and second lasted for around a century while the most recent had an effect for only around thirty years. By inspiring growth, Gordon means that once a groundbreaking technology is developed, it is modified and applied to different areas until it its usefulness to the economy is exhausted. For example, the discovery of electricity led to lightbulbs, television, refrigeration, etc. Gordon believes that the third revolution’s time was so short because of the emphasis placed on entertainment rather than changes in productivity. Thus, he believes that the next revolution, whatever it may be, will also be short lived.
Gordon also presents six reasons why American growth will be difficult to achieve even if we experience long periods of innovation. These reasons include: the end of the demographic dividend (females are already integrated into labor markets and baby boomers are retiring), stagnating educational attainment, income inequality, globalization, environmental policy, and household and government debt.
My main concern with this article is the premise that “innovation does not have the same potential to create growth in the future as in the past.” This claim is based upon the observation that the third revolution did not lead to as long a period of growth as the first two. However, one occurrence is not a trend and trying to anticipate the scope of future innovation is an impossible task. Just as a person in the 1940s could not foresee the internet, readers of this paper in 2015 cannot predict what technology will be available in 2090. It could be a relatively minor development when compared to the invention of electricity or the railroad, or it could completely revolutionize the way the world works. We have no idea. That is why I think his “six headwinds” are a stronger rationale for why growth will be lower in the upcoming years than his comparison of the three revolutions. We can at least see that those problems exist.
http://www.nber.org/papers/w18315.pdf
Graph for comments:
For air travel CPI see series CWUR0000SETG01 vs all items CUUR0000SA0. The CPI has a longer series, but it moves the same.
4 Comments
I agree with George. From the blogpost it seems to me that Gordon merely discusses innovations that could be seen without talking about how information technology systems could promote economic growth. Despite the fact that plenty of innovations (such as cars and weapons) over the past few decades that we can see and touch, what really changed the world today are the innovations that are not visible, namely digital systems. Through digital systems, communications and productions have never been simplier and more efficient.
On top of that, Gordon states “…since then there has been no change in speed at all and in fact airplanes fly slower now than in 1958 because of the need to conserve fuel” in the paper as an evidence of a decrease in productivity as technology remains the same. But I think this is entirely false since speed of an aircraft is not the only factor that determines the productivity of air travels. For example, the cost of plane tickets have become significantly more affordable over the years. And also the increased efficiency at checking-in and security screening are indications of the improved productivity.
In 40 years of airtravel experience, it takes much longer to get from point A to point B, inclusive of parking, airport security and average delays. In the past I know someone who would routinely schedule an early morning meeting in one city and a late afternoon one in another. Fat chance for doing that today, given the additional hours of time in airports. In addition, there are fewer direct flight city pairs, so that travel from Roanoke is decidedly less convenient than 25 years ago.
In sum, the number of work-hours is Gordon’s focus, output per hour, and air travel is distinctly slower, not faster. Furthermore, a quick check of the PPI shows that while airfares have been stable during your flying experience, that’s not true over the long haul. (The CPI gives the same qualitative result.) So no matter how you cut it air travel in particular, and travel times in general, are slower, or at least no faster, for the US as a whole.
Let’s think harder about ongoing technical change. In particular, how about IT?
Well, as the marginal cost of calculations and data transmission of IT fall to zero, the marginal benefit of innovations that revolve around them also fall to zero. The time from order to confirmation of execution of NYSE stock trades was 90 seconds in the days of telegraph, in the 1880s. It’s a bit faster today (unless you’re a computer ‘bot) but not enough to lead to a significant improvement in pricing of stocks. It’s more convenient to call Port of Prince Trinidad than in 1980, but you could call in 1980 so the big productivity improvements were already had. In sum, IT suffers from diminishing returns just like everything else, and the rate at which returns have diminished over the past 30 years is greater than in the previous technology revolutions. After all, the big drop in communication times came with the telegraph. Telephones boosted that, because access became pervasive, and it was point-to-point. Since then? The benefit of each incremental improvement is very, very small and now brings almosts no increment to productivity.
For example, does Walmart have better inventory control than 30 years ago, in the mainframe computer era? No. They were the top dog in the US, so there was a diffusion process that meant benefits were spread over a few years. (Seven-Eleven Japan is arguably the best bricks-and-mortar retailer world-wide in its use of IT.)
More when we chat about what “technology” is.
Perhaps this is less economics related, but with the “singularity” in AI coming closer and closer, potentially just decades away, it seems premature to predict that growth as we have known it for the last few centuries is over. Advancements in computing will radically change our society and could potentially have effects on the manufacturing sector and even basic services similar to those of the industrial revolution. In short, I agree with George that it is impossible to predict what changes the next century hold, but with such advancements as the singularity already somewhat foreseeable, it seems unwise to predict future innovation can’t predict another revolution in growth.
Comments are closed.