Is there a "great stagnation"?

"Don’t be distracted by the internet. It’s a wonderful development, but its mature existence has coincided with the worst macroeconomic decade since the 1930s."

That's like saying, in the 1930s, "Don't be distracted by electricity. It's a wonderful development, but its mature existence has coincided with the worst macroeconomic decade since the 1870s."

What's the actual complaint here? That the internet wasn't powerful enough to keep corporate profits (across all sectors) soaring forever? The fall is the fault of financial idiocy, not the fault of the internet as a technology.

The internet is a really big deal. (So are small, powerful computers, free software, miniature electronics, and genomics, with nanotechnology probably in there, too.)

This complaint that there's less innovation now I just find unfathomable. It seems to require willful blindness to what's actually happening in technology. Alternately, it seems to be tunnel vision on a limited economic perspective - to see that this technology lets us do far more, yet with fewer workers, and out of that to only see that there are fewer jobs.

I think Tyler Cowen is too quick to dismiss the promise of the internet, but overall I find the argument compelling (with critiques I've noted before).

What I think his argument rests upon, is the idea that innovation in what Mokyr calls macroinventions (I don't recall if Cowen uses the term) is slowing down. Computers were undoubtedly a macroinvention (defined by Mokyr as "dramatically reduced prices and created new products but also triggered a continuous flow of secondary and incremental inventions and improvements that made the new techniques work better...), but there don't seem to be as many as there used to be, and only a few fields offer the possibility of new ones, and who knows if these will play out (some bio-tech, nano-tech, AIs, etc. are potential candidates, but who knows if these will go the way of the ether or cold fusion today?).

My point of comparison comes from thinking of the industrial revolution and comparing it today. As Mokyr argues, much of the basic knowledge needed for the industrial revolution's innovations was in place by 1700 (with the notable lack of the theory needed for steam power to be effective). But, up until the 18th century, many of the ancillary inventions were lacking for all but short term take offs until then (in Europe Mokyr mentions the Italian city states, southern Germany, and the low countries as areas in Europe that headed in this direction and failed to take off, I'd add Song China).

What was missing was a set of cultural elements, and smaller inventions necessary for the macroinventions to make a splash. But once this advanced enough, we saw rapid advance for about 200 some years. One invention sparked others and inventions very frequently happened simultaneously in different places. Once the knowledge was there, the next step was easy. These inventions required basic science that was available in 1000AD, but 1000AD lacked the incremental advances so the macroinventions couldn't, and didn't, happen.

Later, by the end of the 19th century, a new cultural and institutional setting allowed for rapid change in basic science, which spurred its own set of macroinventions (for both periods, this included electricity, steam power, steel, germ theory, internal combustion, and a number of others). This carried us through up until the 70s, as Cowen claims. All of these advances relied on science that was fairly accessible to non-specialists, though it became less so as time passed. There were simply so many areas of knowledge that hadn't been tapped for new ideas yet that innovation came rapidly.

And this brings us back to Cowen, the low hanging fruit of the kind of innovations that could be brought about by individual technicians or individuals pursuing solitary work in a lab is gone. There just aren't huge unexplored bodies of knowledge that haven't been tapped for useful innovations. Most innovations today require pushing our basic understanding of science outward to allow a new set, we've already had all but the micro-inventions of our current state of scientific knowledge.

So this means slowing innovation. We can only get a new set of macro-inventions at the pace that our basic scientific knowledge expands, and this is becoming ever more expensive. This doesn't mean that macro-inventions won't happen, but the time between each one will lengthen. This will also mean slower growth, at least through the traditional economic channels since there will be less payoff for catching up with the best technical practices, too many changes are incremental.

I'm less sceptical than Cowen is about long term prospects for growth however. I think there are plenty of opportunities left, both from the incremental changes and from looking at what caused growth before the industrial revolution. Societies could, and did, raise their standard of living and finds new ways of growing. The industrial revolution simply reduced the importance of these factors. From here forward, I believe older patterns will reassert themselves and reward economists that understand their pre-1700 history as well as modern economics. My guess is that the next few decades will reward societies that find new ways of doing things and that focus more broadly on their societies and that societies that try to focus on the factors of growth that were most successful for the past 200 years will find themselves falling behind, much to the consternation of many economists.

Let's not discuss developmental economics at all. Let's all just rush to see who can commit the most post hoc fallacies the fastest! That's the way to do real science!

Does Cowen even mention Solow's growth model?

PS, I don't doubt that we're in a great stagnation. I just think most of the explanations are very silly and fallacious.

Tzimisces:

What are the macroinventions? I'd say fire, the wheel, bronze, iron, steel, chemistry, germs/medicine, steam, electricity, internal combustion, the assembly line, electronics, and computers. The internet probably doesn't make the cut, at least not yet. It might in the future, but I think it won't be *that* big. But nanotechnology might be, and genomics almost certainly will be.

So right now, we're perhaps in the waning days of electronics, in the middle of computers (that one isn't over yet by a long shot), getting started on genomics, and nanotechnology is just barely beginning.

Where were we, say, in the 1920s? Steel was pretty much plaed out, steam was fading, chemistry was going, electricity and the internal combustion engine were starting. By 1950, steam was dead, electricity was fading, computers were starting, and electronics wasn't really started yet.

What I think happens in this conversation is that people compare all that electricity meant for the last 100 years with what genomics has done in the last 10, and conclude that genomics isn't as big as electricity, and therefore that new frontiers are fewer now. That's very short-sighted. Electricity in 1910 didn't show nearly as much fruit as it eventually would.

This year already Intel has released its new "Sandy Bridge" line of processors; for $300 you can get the equivalent or better performance than previous generation "Extreme Edition" chips that cost $1,000. That is incredible innovation at an incredible price.

Once we can figure out how to get Farmville to grow real food, we're set.

rewt,

I'm kinda sceptical about genomics. I did my undergrad in biology at University of Toronto, while people were certainly interested in what was going on they seemed much less taken with the idea that there were going to be the kind of huge advances the media was playing up. This has rubbed off on me and I remain sceptical. Genomics will matter, but the big yield and nutritional benefits of GM crops will mostly involve marginal land which will matter a great deal to poor people and lower the need for capital investment in farming but I don't see this as anything other than an incremental change as far as the rich world will be concerned. Same goes for most of the potential medicine related stuff I remember reading about. It will matter, and probably do a lot to replace the jobs that a recent group of blog postings have claimed the robots are stealing from us, but I see more potential for incremental improvements and narrow advances than I do for the broad based changes that came with steel, chemistry, flight, computers (I see the internet as a subset of computer), etc.

Nanotech is similar, I think there may be more room here for some big advances, but I think they are rather far away and may be more niche too. Of course, I'm not an expert on this stuff, undergrad only goes so far with it, and I haven't really kept current aside from reading the odd article that particularly catches my attention.

Personally, how I view it is that what's different is that we are going off one or two rather than multiple macroinventions at this point in time. Cowen's book picked up on the subject, perhaps using a different terminology, I don't remember, but it showed a decline in the number of these kind of big advances. I generally agree with this, a lot was changing in the first half of the century. We've got computers still going strong, and to a certain degree materials science for plastics and composites, but in the first half of the 20th we had wireless, electricity, internal combustion, flight, refrigeration, etc. I do think this has slowed down, and we're experiencing less of this than we were 50 years ago. Rather than having a half dozen or so big complexes of inventions going, we're relying on one or two at a time. Certainly, this allows for much higher growth than the 17th century was capable of, but I don't think it allows for as much as the early 20th, at least without a lot more effort.

If the criterion is the extent of technological change, no generation will ever match that that went from the way things had pretty much always been to railroad, steamship and telegraph.

But what about organizational change ? Consider the immense increase in average wellbeing in America as the limited liability corporation became the standard form of business organization.

We are in the midst of another great organizational change, as gate-keepers and middle-men of every description disappear, allowing individual access and choice to explode in wonderfulness.

As a Sci-Fi fan, I can think of one macro-invention that needs to take place to advance into next phase of civilization:
We need to really understand GRAVITY. Ability to manipulate gravity will be a civilization-changer that is probably more fundamental than discovery of electricity or nuclear power.

I think the real lesson is that you get much more attention by writing something short, easy to read and controversial in the sense that you're arguing about things no one has an answer for so everyone has a quote to offer.

I'm sorry but I view this debate as similar to the one about the extent of language in chimps. Take a position, any position. Doesn't really matter.

Hal Varian is a great economist but his answer is falling a bit into the 'Whizz, bang, Apple!' category that I see Cowan trying to counter.

30 years ago he could have received a phone call from the Economist, picked up the print edition, and faxed or telegraphed or even posted a reply and waited to see it published in the next edition. This would have taken four or five days longer, but the end result would have been much the same, and I'm not sure I'd want to pay anymore for the speed.

Of course I couldn't have sent this reply for another week, and you might not have published it, but it's making me late for work and it's perhaps not of good enough quality anyway.