Even by the standards of a field known as the dismal science, Northwestern University economist Robert Gordon is a remarkably gloomy thinker. This summer, while most of us were busy fretting about the tepid U.S. recovery, he managed to up the ante with a paper that looked 90 years down the line and asked, "Is U.S. Economic Growth Over?" As in, over for good.
His answer wasn't quite a straightforward, "yes," but it was nearly as bleak. Gordon predicted that a mix of technological stagnation and economic headwinds could feasibly slow the economy down to a crawling, pre-industrial growth rate, as mapped out in the green line on his graph below. With the new year just hours away here in the U.S., I thought it would be a good time to revisit my nominee for the most depressing economic idea of 2012, along with excerpts from a conversation I had with Gordon about his work a few months back.
Gordon's argument has two distinct halves, which can be summed up as this: Our greatest innovations are behind us, and the United States will be weighed down by its own dysfunction. We'll tackle each part one at a time.
Not unlike Tyler Cowen in The Great Stagnation, Gordon argues that the world has hit a technological impasse. For the last 250 years, he says, worldwide growth has been powered by three separate industrial revolutions that gave us immense, one-time gains. Round one, from 1750 to 1830, saw the advent of the steam engine and the railroads, which made it possible to travel vast distances in relatively short periods of time. Round two lasted from 1870 to 1900 and gave us the internal combustion engine, indoor plumbing, electricity, and the other foundations of modern infrastructure. The spin-offs from those inventions, such as highways, air-conditioning, and efficient factories, kept the economy growing speedily for decades more. Finally, in the 1960s we entered the information age. Whereas thinkers like Cowen believe that the Internet and digital technology still have lots of untapped economic potential, Gordon argues their major contributions are pretty much spent. There will continue to be great inventions in the future, he says. But they aren't like to power growth the same way as, say, the advent of the automobile.
Then come the headwinds. Gordon identifies six challenges he thinks are going to sap U.S. growth in the years to come: our aging population, our faltering education system, income inequality, foreign competition, the inevitable impact of global warming, and the need to eventually pay down our debt. Subtract the impact of each from out average pre-recession growth rate, and suddenly we're back in the 18th century.
That "exercise in subtraction" isn't intended to be taken literally. Rather, Gordon uses it as a thought experiment to try and make his readers consider the unique hurdles ahead of the U.S. "My guess is that a Canadian or Swedish economist looking at the past and future of his or her country would not be nearly so alarmed," he writes.
Since his publication, Gordon's theory has drawn a heap of criticism those who believe he's underestimating technology's potential and America's ability to fix its own problems.* As the Economist noted, techno-pessimists have been around a long time, and they tend to be proven wrong.
Gordon, though, hasn't been dissuaded. This month, he answered many of his detractors in a Wall Street Journal op-ed. "In setting out the case for pessimism, I have been accused by some of a failure of imagination," he wrote. "New inventions always introduce new modes of growth, and history provides many examples of doubters who questioned future benefits. But I am not forecasting an end to innovation, just a decline in the usefulness of future inventions in comparison with the great inventions of the past."
I interviewed Gordon not long after his paper was released. Here are a few excerpts, edited for length and clarity.
Why, if you want to understand the today's stagnation, go watch old TV episodes.
Many years ago, the New Yorker commissioned somebody to sit at a television set and watch television for a week and write an essay on it. And one of the things he came back and said was, "I was so struck by the situation comedies from the 1950s, the reruns, how similar their lives seemed to today."
This idea, which I've always called the New Yorker game, is to go and do that repeatedly over 30 year intervals, going back, and see when you could NOT make that statement. If you were sitting in the middle of the 1950s, you would see a lot of differences with the 1920s. But the comparison that would be the most stark would be between 1925 and 30 years before that. Because that's when you got this enormous transition away from horses to clean streets to electric power.
Why inventions aren't what they used to be (or: Why the car is a much, much bigger deal than the iPhone.)
The automobile replacing the horse [made] it unnecessary to feed 25 million horses. That, combined with the tractor, simply freed up about a quarter of our agricultural land to feed people or to export to other countries. That's an example of a really big deal, a big invention.
The automobile invented a whole new nature of activity called personal travel. Prior to the automobile, people didn't have their own horse. It was too expensive and maintain and stable a horse, particularly in a city. So the automobile created a whole new activity called personal travel -- vacations, going out and driving through the countryside on the weekend -- that didn't exist before, and was responsible for a tremendous amount of economic growth from the 1920s through the 60s as everything from motels to resorts to Disneyland to supermarkets, all these subsidiary inventions to the great invention of the internal combustion engine, gradually made their way through the economy. That whole process, as I emphasized in the paper, took 100 years.
Now let's take the iPhone, the iPod, and the iPad -- they're the same general idea. The iPod replaced portable CD players. The quality of the earphones didn't change. In fact, the earphones that came with the iPod were clumsy little buds. The music didn't change in fidelity. What changed was the convenience -- both in buying the music, and in having this vast library of music in one tiny little device. Most of the things that people do with modern apps on iPhones can be done with a laptop. But it's obviously much more convenient to have the smaller package. Nobody's going to walk down the street with an open laptop. So we're getting the same information, but we're getting it more often.
Why energy efficiency could be the next great revolution.
Let's say that we replace all the incandescent bulbs with compact fluorescent bulbs. They last 10 times as long, they're three times as efficient, and that allows you to use less energy and shut down some coal plants. You can also re-divert the capital and the workers into some more productive line of work. I think that's actually very analogous, on a smaller scale, to the replacement of the horse by the tractor. So I would agree with the premise that energy efficiency is just like any other kind of efficiency in industrial history. It's part of raising the standard of living. But we're not on the cusp of any kind of revolution here. We're incrementally inching toward what Europe has already achieved with $8 gallons of gas and escalators that look like they're stopped but are actually activated by a footpad so they don't run 24 hours a day.
Why technology won't give us the same bang for our buck in the future.
Remember, I didn't say growth is going away. It's not going to be zero. There are going to be continuing innovations in medical care. In energy efficiency. Driverless cars. I was very impressed by Tom Friedman's column on the movement towards small, programmable robots, analogous to the PC vs. the mainframe. I can see that as having a lot of uses. There is going to be continuing technical change. But say you have 10 inventions. Let's say they're all equally important. Then to increase the power of the inventions by 10 percent, you just need one more. If you've got 100 inventions, to increase them by 10 percent, you need 10 more. And so forth and so on. You're facing a continuous upward climb to keep coming up with this multiplicity of new ideas to get this same percentage growth rate.
It's really an argument about big versus small and fast versus slow. I'm not trying to make an argument that there's not going to be no growth in the future. And my graph shows that the standard of living, even in my pessimistic prediction, that the standard of living will double by 2100, by the turn of the next century. And doubling over what we have now is nothing to sneeze at, though it would be nice to know whether it will be distributed as unequally as it is today.
Why everybody agrees with our biggest economic problems (even if they don't agree with the solutions).
Who does not agree that baby boomers are retiring, so we're getting less hours per person? Who does not agree that we have the problems with cost inflation in higher education and an achievement in secondary education and a lack of convergence of minority achievement? Who disagrees that we have rising inequality? Is there anybody who disagrees that there has been outsourcing and movement of jobs abroad, that multinational corporations increasingly are supplying these growing emerging markets that you mentioned by building plants that are closer to the demand in China and elsewhere. And does anyone disagree that at some point we're going to have to do something to tame the government's deficit?
These are my headwinds, and it's not very controversial that they're there. Solutions [are] where everybody stops agreeing.
Why it doesn't matter if you believe his exact math about America's future.
The exact numbers don't matter. But they're all in the same direction. And whether you're subtracting .1 or .2, you're still subtracting to a lower level that's a lower growth rate than we've ever had, certainly in the records of GDP statistics. I should have said this was spurious precision. I like to do that because it focuses people on saying, my god, this really does add up to something. How can I argue that it doesn't add up to anything?
This article available online at: