From 1950 to 2011, U.S. economic growth averaged 3.3 percent annually, divided roughly equally between average labor force increases of 1.5 percent annually and productivity gains of 1.8 percent. (Productivity — efficiency — generally reflects new technologies, better management and more skilled workers.) With the labor force increasing more slowly, the pace of potential U.S. economic expansion would drop to 2.3 percent annually, assuming that productivity gains stay the same. Unfortunately, that’s an iffy assumption.

In a fascinating paper, economist Robert Gordon of Northwestern University speculates that productivity increases have peaked. Per capita income gains may gradually slow to half or less of their historical rate. Most economists, he writes, believe “economic growth is a continuous process that will persist forever.” It may not, he argues.

Gordon identifies three industrial “revolutions.” The first began in England around 1750 and featured cotton spinning, the steam engine and railroads. The second, dating from 1870 to 1900, was the most significant and involved the harnessing of electricity, the invention of the internal combustion engine and the advent of indoor plumbing with running water. These, he contends, triggered other advances: appliances, highways, suburbs, airplanes, elevators and modern communications (telephones, televisions).