Monday, May 13, 2019

Free Solo and Economic Growth

We recently watched "Free Solo", the great movie about Alex Honnold's free (no aids, no ropes) solo climb of El Capitan. Among many other things, it got me thinking about economic growth.

The abilities of modern day rock climbers are far beyond those of just a generation ago. The Wikipedia history of El Capitan starts with a 47 day climb in 1958, using pitons, ropes, and all sorts of equipment, and continues through development of routes and techniques to Alex's three hour romp up the face. 

Why wasn't it done long before? There is essentially no technology involved. Ok, a bit. Alex is wearing modern climbing boots, which have very sticky rubber. But that's it. And reasonably sticky rubber has been around for a few hundred years. There is nothing technological that stopped human beings from climbing much like this thousands of years ago. Alex, transported to 1890, might not have free soloed El Capitan without his current boots, but he would have climbed a lot more big walls than anyone else.

Clearly, there has been an explosion in human ability to climb rocks, just as there has been in human productivity, our knowledge of how to do things, in more prosaic and more economic activities. And, reading the history, the rate of improvement has grown over time. 

I think that in studying economic growth, we (and especially we in the Silicon Valley) focus way too much on gadgets, and too little on the simple fact of human knowledge of how to do things. Southwest Airlines' ability to turn an airplane around in 20 minutes, compared to the hour or so it took in the 1970s, and still does at many larger airlines, is just as much an increase in productivity as installing the latest gadget. Growth is about the knowledge of how to do things, only sometimes embodied in machines. Free solo is a great example of the pure advance of ability, from a pure advance of knowledge, completely untethered from machines.

And the same patterns emerge that growth theorists tell us about. 

Knowledge externalities When one person learns how to do something, and can and does communicate that knowledge to others, then the others can quickly benefit from knowledge and the group advances. 

Alex, like Newton, climbed from the shoulders of giants. Just how do you get up El Capitan? There are now many established routes. A "route" is, as the movie made clear, a succession of incredibly tiny holes cracks and ledges in a 3000' face of rock, that experienced climbers figure out how to stitch together. Alex didn't have to figure all that out, and chose an established route.

Likewise, nobody in 1958 had any idea that you could hang by your thumbs and fingers to exploit little pieces of rock. This knowledge, demonstrated in the movie, emerged from the community of rock climbers and boulderers over time. Alex is incredibly good at it, but he learned from others.

Knowledge transmission Everyone is all upset about intellectual property these days, but nobody patents anything in rock climbing. (There is some patentable technology in the devices people use to climb with ropes, and that has enabled free climbing, but it's really not central.) The knowledge gets produced, which is costly to the individual producing it, and then passed on, where it is much easier to learn than it is to innovate, and the whole group gets better. 

Once a piece of knowledge is produced it is in society's interest to pass it on as quickly as possible. The whole IP business trades a later reduction in growth -- slowing adoption while the innovator gets to earn some rents -- for the idea that these rents are vital to creating knowledge in the first place. But lots and lots of productivity-increasing knowledge -- most, I would hazard -- is created like new hand-holds or new routes, for free. There are other social institutions that promote the creation and dissemination of knowledge, and rock climbing is full of them. 

The size of the group and the cost of transmitting information The key insight of modern growth theory is that, as a result of the process described above, the larger the group studying any problem, the faster knowledge advances. If 1000 people are figuring out how to climb, and each of their good ideas disseminates through the group, each member of the group gets to use new ideas more quickly than if there are 100 people doing it. 

(I think our models don't pay enough attention to the dissemination question. Most new ideas are bad, so the process of sifting through new ideas, figuring out which are good and bad, refining them, is a lot of what a group does, and all that and learning takes time and effort. The world does not just have one individual innovating at great expense, then the rest learn for free. Academics, who spend a lot of time reading hard papers, writing referee reports and comments that distill the ideas, throwing most new ideas out, distilling again to teach, see that every day!)

The move makes clear, that the world of rock climbing has expanded vastly since the 1950s. Bouldering is a weekend recreation for thousands, unlike dedicated mountain climbing in the 1950s. No surprise then that the rate of knowledge creation is higher. 

The size of the group is limited also by its ability to communicate. I locate the beginning of growth and science with Gutenberg. (An idea also unpatented and quickly improved on and copied.) Printing means that if you run a costly experiment, then you can share that with a much larger group, and a much larger group can discuss and refine the idea. If you can only share it by word of mouth or handwritten note, few will learn of it and be able to use it. 

So, similarly, I would say in the end that rock climbing is much more advanced than before because of technology -- but the technology of communication. First, the technology of print and media -- notice the magazine covers in the movie. And now, the technology of the internet. Each new idea in rock climbing is accessible quickly all over the world. Without that large group of interested people, this communal knowledge would not have advanced so far. 

Which gives me hope, in the end, for growth. We just unleashed a reduction in the cost of communication larger than Gutenberg created.  The group of people studying any problem is much larger, and the number of problems that can be effectively studied by groups of efficient scale (1000 - 10000 seems to be the size of an academic field before it splinters into subfields, and the same seems to be true of recreation) has exploded, the fraction of the human population that can work together on any problem has exploded. At least the possibility is there. It still took 200 years from Gutenberg to the scientific revolution, and lots can go wrong along the way. 

The movie, of course, is about the psychology of extreme danger. But I'll leave that for another day. 


29 comments:

  1. The references to Gutenberg reminds me of a discussion in Simon Winchester's fascinating book "The Man Who Loved China". There the author points out that the first printed book occurred in China about 500 years before Gutenberg. The question posed by Winchester's "Man" (Joseph Needham) was why, with such a lead, Chinese science and technology did not advance as fast as in the West. Though Needham provides no clear answer, it appears that government's relationship to the market played an important role.

    ReplyDelete
    Replies
    1. Mr. Nicholson, I suspect it may have taken 500 years partly because of time and cost it took to transmit information. Given today's technology, information is transmitted virtually instantaneously. The Treaty of Paris ended the Revolutionary War, but some British troops continued to engage as that information took weeks to get to them.

      Delete
    2. Scott SomervilleMay 15, 2019 at 12:01 AM

      Wine! One of many reasons may be that the European taste for wine promoted much effort in glass making. Refined glass making skills led to reading glass technology, extending the productive life of the scientists at work.

      Delete
    3. One of China's problems was probably their writing system which was complex, expensive to learn, rigid and limited compared to alphabet systems.

      Delete
    4. Really? Chinese are pictographically all literate.The abacus still works fine

      Delete
  2. I find most modern growth literature pretty boring... all the insights are quite trivial. By the way, people today also have higher IQ, they run faster and so on. Part of this is because of better nutrition, so there is an effect for technology.

    ReplyDelete
  3. It is also possible that there were always good climbers, but they were busy working.

    ReplyDelete
  4. I have to admit, I mostly clicked on this because I couldn't understand the connection with economic growth. But you did it: I learned something useful from reading this. Thanks!

    ReplyDelete
  5. As both a climber and economist, I think this blog post is spot on. Free solos took place before I started climbing 52 years ago, and training methods were already evolving then. The massive increase in the population of climbers -- particularly those who climb indoors -- led to an accelerating dissemination of knowledge. In my experience, the biggest factor in getting up a route is knowing someone else who did it. We think, "if they can do it, so can I." Honnold's genius, like that of most economic and intellectual innovators, comes from his willingness to make his imagination real.

    ReplyDelete
  6. I think it is clear the "cumulative increase in compentency" is a natural thing. It is, indeed, the basic step in Evolution. The more complex structure builds itself from the less complex predecessor. And ideas move upward in the same way. "Human observe; human copy, and improve."

    ReplyDelete
  7. Matt Gelfand (Twitter: @matt_gelfand)May 14, 2019 at 8:48 AM

    At first I thought the headline was misspelled and meant to be "Free Solow and Economic Growth" that is, something about expanding Robert Solow's model of economic growth. Perhaps the blog is about the Solow Model after all.

    ReplyDelete
  8. John, you should have titled this post "Crescat Scientia; Vita Excolatur."

    ReplyDelete
    Replies
    1. Ah a member of the Chicago illuminati warning another not to spill the special beans?

      Delete
  9. professor. Cochrane. Paul Krugman keeps saying stupid things on Twitter and NYT Could you deal with him?

    ReplyDelete
    Replies
    1. The best way to neutralize the impact of stupid arguments is to not pay attention to them.

      Delete
  10. The spread of knowledge, and its discovery, does face institutional resistance, backed up by law. Patents, and tariffs, are weapons used to limit the use, and reach, of discoveries, inventions, and operations. The likely less limiting, non-western, practices entertained, say, in China, does give that country advantages, say, in genetic research, in biology, electronics, and chemistry. I recall the west profiting from Chinese knowledge freely, or stealthily. The story of the silk worm being smuggled out of China by Marco Polo is entertaining. The use of gunpowder, rocketry, paper, mobile type, tea, etc.,without permission, is amusing to entertain. Then, of course is the plant selection, and breeding of new world inventors in the cultivation of the potato, corn, beans, yams, sugar, chocolate, quinine and on and on. Did the discoverer of penicillin make out like a bandit? The patenting of human genes is an interesting reach of "invention" limiting the speed, access, and benefits to humanity. Nature itself is evolving products, immune from legal limits, even hazards....take Ebola, and AIDs, and maybe climate change. Recall the Chinese saying, "We live in an interesting times." Like the Music Man said, "You have to know the territory..." to make a buck.

    ReplyDelete
  11. Kudos to the author for mentioning the term "PATENT" in an article about technology growth. How often have I read such an article without a single mention of this term. Whether you like patents or not, they are important, and VC types and investors often ask, 'where are your patents'?

    ReplyDelete
  12. Regarding the knowledge transmission point, there's something interesting afoot in the adventure cycling community. The short of it is that content creators, who go on these adventure rides and post online about them and how epic they were, realize that the route itself is a major part of the value they're creating/producing. They're effectively giving away this information for free.

    Frustrated by this, a few content creators have started selling detailed maps and descriptions of their routes. This is not without controversy, of course. For an example, see here: http://www.pathlesspedaled.com/routes/

    I don't see this possible within the climbing community, but it's fascinating to me to see entrepreneurs try to capture value where before they were giving it away for free. Will it work? Who knows?

    ReplyDelete
  13. Good post. I'm also a climber and economist, and am not sure how one would define "output" in climbing. Perhaps output is the difficulty of the hardest climbs, so output growth is the same as pushing grades, that is, climbing harder and harder routes. By that definition there is definitely "growth" as the Yosemite Decimal System initially went from 5.1 to 5.10 and now includes many grades above 5.10, and up to 5.15. Adam Ondra and Chris Sharma are largely responsible for this extension of the range of the possible, and thus are more likely candidates as representatives of growth in climbing. But by this definition and by his own admission Honnold's achievement does not represent growth. He "only" free solos 5.12/5.13. Many people can climb grades harder than the maximum grade Honnold can climb. What's new here is that he free solos harder grades, but that is not a function of his superior skill as a climber. The proper interpretation of his achievement is that he is a climber drawn from the upper tail of the distribution of boldness. This is not something that he had to accumulate "knowledge" to do - just a certain kind of personality (or amygdala) that he was endowed with from the start. There is no technological recipe for boldness, like there is for climbing harder grades. I very much doubt that Honnold can "teach" someone to free solo El Cap. So the analogy in this respect is a bit contrived.

    ReplyDelete
  14. If Steve Jobs were a climber, he would have patented or copyrighted all routes he took, and demanded rents from anyone who used any part of his past route (aka sampling).

    Btw, Planet Money recent reran the interview with the founder of Patagonia which grew out of his love of rock climbing. He increased to price/cost of climbing equipment from that made in Europe. There routes were created permently so hardware was used once and left for future climbers. But the leave no trace ethos in the US required removing everything, and that grew to not causing damage from removing hardware. To remove pitons they needed to be better steel, and have different shapes, which increased costs 10x, but different hardware had to be invented like nuts and wedges that would cause very little damage, but be both safe and easily removed. But later, he started first importing clothes, then making clothes better suited to climbing, but in colors other than gray.

    Which was bought by non-climbers.

    Free climbers need more than shoes, but "profit" is made in the clothes which are both functional and stylish. Patagonia defines "profits" as paying jobs the workers want, not the money never paid to workers extracted from the workers who buy the clothes.

    Also, "rock climbing" is now a demonstration Olympic sport using plastic rocks fixed to interior wood walls. Which greatly expands the supply of demand: people who want to climb with less investment, ie on lunch break or after work.

    ReplyDelete
  15. One quibble:

    There is nothing technological that stopped human beings from climbing much like this thousands of years ago. Alex, transported to 1890, might not have free soloed El Capitan without his current boots, but he would have climbed a lot more big walls than anyone else.

    Actually, there is a technological hurdle. Before Honnold did his free-solo ascent, he practiced all the pitches in the route many times using modern climbing equipment. It took a lot of trial-and-error to achieve a level of mastery on the most difficult sections such that he was finally willing to make the attempt. Without all the modern gear for practice, he could never have done the climb.

    But I do agree with the general point about the developing and sharing knowledge.

    ReplyDelete
  16. Not to quibble here because I realize this isn't the most important part of your article, but advances in climbing technology have actually had a significant effect on the grades people can climb. Modern ropes are vastly stronger and more elastic than earlier ropes, allowing climbers to safely take bigger falls and thus climb harder routes that accept less protection. The shift from hand-hammered pitons to nuts and cams dramatically sped up the process of placing protection, reduced the weight climbers had to carry, and made previously unprotectable routes safe. And climbing shoes with sticky rubber aren't marginally better than canvas sneakers with the insoles ripped out, or leather boots, they're vastly better. Honnold has a unique mind which is his main advantage, and that mind would likely be unique in any point in history, but he developed his skills using modern tools of the trade. There are plenty of climbers from the 50s and 60s who would be world class today if they were able to climb with modern equipment, and many of us who climb today would be deeply humbled if we were forced to use the equipment of the past.

    ReplyDelete
  17. It's mostly not a technology story, but:

    You're neglecting the technology of the climbing gym -- artificial holds which resemble outdoor rocks enough to keep people's interest and train their muscles, even in cold or wet weather. "Honnold started climbing in a climbing gym at the age of 5", says Wikipedia.

    More broadly, climbing gyms emerged after the cultural changes from air conditioning, better food storage, the pill, etc., and I don't think the gyms would have happened without those technologies.

    ReplyDelete
  18. Very nice argument. But I wonder how much of the improvement in maximal performance is simply because the world today has roughly twice as many people as in 1958, and so does the US? The tallest person amongst 300 million people will be taller than the tallest among 150 million, and the best performer will be better.

    ReplyDelete
  19. Growth as % of overall population who can climb mountains in the past compared to the present would not be much. Yes there are more people who can climb rocks but not sure if there is growth in % of people with the rock climbing abilities.

    Similarly though we reduced the cost of communication which you imply that increased our ability to create useful knowledge , it might not lead to that growth because that is an overall figure we might have so much negative externalities as much as positive and overall growth could be 0 or negative.

    ReplyDelete
    Replies
    1. Yes, but growth doesn't need all of us to get better at everything. We just need to get better or make things better in our sliver of the economy.

      Delete
  20. Remember the much maligned (on the right) "you didn't build that by yourself"? Or how about all of what you say here was discussed by Peter Kropotkin and other earlier anarchists? Every one of us, including the "giants" and "unicorns" inherit a body of socially (re)produced knowledge, infrastructure, routines, norms, etc. And the outsized share of it that so few expropriate while so many still suffer is morally obscene. Note, too, how the rock climbing community, and especially Honnold, is not at all motivated by anything approximating the profit motive. Cheers! --Enoch Lambert

    ReplyDelete
  21. Really interesting post. And thanks for the movie recommendation!

    If anyone is interested in fun and interactive simulations of knowledge transmission and similar network effect, go ahead and check this post out: https://www.meltingasphalt.com/interactive/going-critical/

    Like this one, it was featured on hacker news and is awesome.

    ReplyDelete
  22. The advance of rock climbing knowledge is probably just another example of Swanson's law which seems to be ubiquitous.

    When comparing modern advances to Gutenberg - Gutenberg's advances are probably more important because (1) what came before was so expensive and (2) Gutenberg would have been applied to the highest value uses first.

    Progress is ultimately constrained by the laws of thermodynamics.

    ReplyDelete

Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.