Will Long-Term Growth Slow Down? Becker

Sustained long-term economic growth beginning in the near future would help greatly toward overcoming two major problems confronting the United States (and Europe and Japan). One is the high ratio of government debt to GDP that resulted from budget deficits due to the rapid increase in government spending during the past several years. GDP that continues to grow faster than outstanding debt is the surest way to reduce the burden of the debt. Sustained long-term growth would also allay the fears of many parents that their children would not be any better off than they are.

Improvements in productivity due in large measure to new technologies have been the major source of long-term economic growth in per capita incomes. From 1880 or so to the beginning of the financial crisis, American productivity advanced on average at a rate of a little less than 2% per year. This helped, along with capital accumulation, to produce a long-term growth in American per capita incomes of about 2% per year. Since growth continued at such a steady rate for such a long time, one might reasonably expect that the US would resume growing at a similar rate once it gets beyond the effects of the financial crisis and the Great Recession.

However, a recent study by Robert Gordon of Northwestern, one of the leading experts on productivity, puts a damper on these expectations (see his “ Is U.S. Economic Growth Over? Faltering Innovation Confronts the Six Headwinds”, NBER working paper 18315, August 2012). Gordon argues that advances in productivity were slowing even before the financial crisis hit because the innovations of the past several decades, including computers and the Internet, were less important than those at the end of the 19th century and beginning of the 20th century. He also argues that future growth in the U.S. is likely to be even slower than in recent decades because of six “headwinds” that he believes will reduce growth. If Gordon is right, Americans face an unprecedented and dismal future of basically stagnating incomes.

It is common during a long and deep recession or depression for economists and others to become pessimistic about the economic future. For example, at the end of the Great Depression in 1939, a leading American economist of that time, Alvin Hansen of Harvard University, argued that the U.S. and Europe were in for long-term (secular) stagnation, partly because he believed that technological progress would be much slower in the future. Of course, he turned out to be completely wrong. However, Gordon’s forecasts abstract from the financial crisis and resulting recession, and he bases his pessimism on how the situation looked to him prior to the crisis.

Gordon follows an established approach by dividing the past 200 years into periods of three Industrial Revolutions. The initial one occurred during the last half of the 18th century and the first several decades of the 19th century, with the steam engine and railroad being examples of the major new technologies from that revolution. The second, and what he considers the most important, industrial revolution occurred between 1870 and 1900. This revolution gave us, among other inventions, electricity, the automobile, the airplane, and the small engine. The third revolution started around 1960, and encompasses computers, the Internet, and genomics and biotech.

Gordon’s main reason for pessimism about future growth is the evidence he presents that American labor productivity (measured by output per unit of labor input) advanced much more slowly after 1970 than it did between 1890 and 1970. He also points out that throughout most of history- that is, until the first industrial revolution- annual growth in world per capita income was close to zero. Perhaps, according to Gordon, we should think of the 3 industrial revolutions not as the norm for the future, but as “temporary” exceptions that will not be repeated in the future.

Gordon makes a thoughtful case for his conclusion that future long-term growth for the U.S. will be much slower than past growth. Still, I do not find the case convincing. While growth during the past two centuries was radically different from the slight annual growth during the prior two thousand years, the reason is not luck or accident, but in good part was due to the development of science, and especially to the application of science to industrial progress. Knowledge builds on knowledge, and the available evidence does not indicate that the accumulation of knowledge is subject to diminishing returns. This suggests that future knowledge could very well grow at a rate comparable to its growth during the past century and a half.

Another difference between the past two centuries and previous history is the emergence of economies that relied on competition and private enterprise. This was the economic system in Great Britain when it led the world in productivity advances, and it describes the U.S. economy after it took over leadership from Britain. Advances in technology and productivity are likely to continue at a good rate if the U.S. and other leading countries continue to emphasize competition and the private sector, and do not use governments to try to determine future technology “winners”.

I also believe that Gordon underestimates the full impact of the third revolution based on computers and other modern technologies. As he shows, the effects on productivity of the 2nd revolution lasted for close to 100 years, while the 3rd revolution has been going on for no more than about 50 years. It is extremely difficult even for the most informed individuals to predict the long-run effects on productivity of new technologies. Gordon quotes someone working in 1876 for Western Union, the major telegraph company, who claimed, “The telephone has too many shortcomings to be considered as a serious means of communication” and Bill Gates who stated, “640 kilobytes ought to be enough for anyone”.

I will say little about the six headwinds that Gordon believes will also slow down future growth since his arguments are not convincing. To take a few of his headwinds, I believe globalization will add to, not subtract from, U.S. growth and real per capita incomes, that the effects of having fewer young persons working relative to the number of retired older persons will be partially overcome by considerable extensions of the ages at which workers typically retire, and that inequality will likely begin to decrease. Along with Gordon I am concerned about the effects of global warming on the economy, but I expect new technologies to go a long way toward solving that considerable problem, just as technological discoveries overcame many challenges in the past.

I agree with Gordon that sizable future growth in per capita incomes in a leading economy like the American one will not come automatically just because past growth was considerable. However, I do believe that the dividends from the 3rd industrial revolution are far from exhausted, and that future growth can be robust given the right economic environment. What I mean by the “right environment” has several components, but number 1 would be a continued reliance on competition and the private sector as the principal way to organize the economy, and number 2 would be to improve investments in education and other human capital.

Comments

[Neilehat]

An empirical basis for the Systems and Economic Law of "Diminishing Returns" perhaps? Instead of the Theoretical? As for it's Proof, only time will tell... ;)

As for those standing in the unemployment line - hopefully not. Humanity and it's mind is the only anti-entropic principle truly in operation in any System...

[Jon Orloff]

I think Gordon's view that technology will not have a significant effect on productivity is dead wrong. I think we need to be a heck of a lot more concerned about what will happen to all those whose jobs will be done better, faster and more cheaply by robots over the next ten years.

[Cagomez]

Shouldn't technological innovation and therefore economic growth actually accelerate since the internet makes it easier for people to collaborate and share knowledge much more easily and cheaply than in any previous time in history?