How Should We Measure Progress?



A few years ago, the American economist Robert Gordon wrote a must-read book called The Rise and Fall of American Growth. He pointed out that the first half of the 20th century was punctuated by steady improvements. We went from having little penetration of electricity to everyone having electricity; from no cars to cars everywhere; from no airplanes to routine jet travel at 600 mph; from not having air conditioning, to everything being air conditioned.


The second half, on the other hand, saw few such obvious improvements. We still travel at the same speeds; there haven’t been step-function improvements in those prior inventions; and new inventions such as the internet haven’t really moved the needle on measures such as GDP. I wrote a thread about the book here:


And yet, in many aspects of our lives, it certainly feels as if there has been tremendous progress. Is there a better way to measure it?

Last week I had a few magical experiences on Clubhouse, the new, audio-first social network. If you haven’t heard, Clubhouse allows anyone to start a room or a club, discuss any topic, and invite folks to join them in the conversation.



In this particular instance, I was having lunch when my iPhone notified me that Jason Crawford, the author of the excellent blog Roots of Progress, had just started a room. Since it’s a topic I’m interested in, I joined, and before I knew it, Jason had invited me to speak in a room that had grown to over 1,000 participants from all around the world. Like I said, it’s magical.


Jason posed a question: how should we measure progress?


It sounds like an easy question to answer. Yet if you invite ten economists to opine, you’ll get ten different answers. Measures such as GDP, or how much a country produces for every inhabitant, are an obvious starting point.


To cite a now-clichéd criticism of GDP, if I buy a set of Britannica books for $1,500 it gets counted in GDP. But if I don’t and use Wikipedia instead—which is better than Britannica along every dimension—it shrinks GDP.


There are folks trying to modify GDP to better take into account the digital goods that provide us with enormous consumer surplus. How exactly does one measure the benefit of talking with dozens of friends around the world in high-definition video, on my phone, for free, through Facebook?


Jason went around the room. A woman from Dubai suggested measuring access to healthcare and education, for example, as a measure of progress. An economist from Ireland talked about countries like Brunei, which are trying to measure progress through a broad happiness index.


When it was my turn to speak, I pointed out that we’ve made a lot progress in so many areas, but we’ve also gone backwards in many others.


One of my favorite examples comes from venture capitalist Peter Thiel in 2014: “I live in the Marina area in San Francisco. They built the Golden Gate Bridge in three and a half years in the 1930s, ‘33 to ‘36. They’re now building an access road to the bridge that’s taken eight years and possibly will end up costing more in inflation adjusted dollars than the whole bridge cost in the ‘30s.”


Another example is from an excellent 2017 investigative piece in The New York Times called The Most Expensive Mile of Subway Track on Earth. The Times discovered that it costs seven times more than the world average to build a mile of subway in New York City. It also takes longer than average.


Credit: AFP/Getty Images


The reason for these failures is not technological. They are political; in both cases, it’s regulatory capture by vested interests (frequently, unions), and outdated regulations such as NEPA. As a country, we have grown so prosperous and have enjoyed such a long stretch of peace that we’ve forgotten how to hustle and get things done with scarce resources.


One example: the book Freedom’s Forge narrates how America accomplished incredible feats of manufacturing with scale and speed, only because we were trying to win the war against Hitler.


The only entrepreneur we have capable of working apparent miracles today is Elon Musk. Everything he does is huge and ambitious; the gigafactories, the Starship (largest rocket ever built), the Boring Company and Neuralink. Compare the time it took to approve that access ramp in front of Peter Thiel’s house with what Musk has accomplished with his tunnels. The only problem with Elon Musk is that we only have one Elon Musk.


In a recent podcast interview, technologist Balaji Srinivasan argued persuasively that not only is the United States losing “market share” of GDP (again, that imperfect measure) to Asian countries, but certain Asian countries like India have more state capacity than we do as measured by their ability to rollout innovations such as a unified payments system and a digital identity system, among other innovations, in record time and with record levels of consumer adoption.



How did the Indian state become so competent, so fast? Conversely, how did we become so incompetent?


The war was a forcing function for us to get things done, or else. How can we simulate that today, lest we continue down our path to mediocrity?

After listening to everyone’s points in our impromptu Clubhouse meeting, I realized that progress can only be defined narrowly. Yes, maybe we’re really slow at building subway miles, but look here at our incredible speed in creating a vaccine for Covid-19. Isn’t that progress?



It sure is. But Jason asked the group, “If you had a dashboard to track progress over time, what would you put on that dashboard?”


Here’s my answer: over the last 30 years, most components of the consumer price index (CPI) have mostly come down in price, except for three components that have persistently bucked that trend: education, healthcare, and housing. In fact, these three sectors have accounted for almost 90 percent of the rise in CPI over the last 30 years.



As someone astutely observed in our meeting, maybe that’s because of the Baumol cost blessing? The idea goes as follows: in the 1850s, it took four musicians to play a string quartet. In 2021, it also takes four musicians to play a string quartet. Unless we pay them a living wage comparable to what they could earn elsewhere, we won’t have any string quartet musicians.



This means that sectors with low labor productivity should exhibit inflation, and that is what we observe in education, healthcare and housing.


I find this answer deeply unsatisfying. It’s possible to learn nearly anything on Khan Academy for free, not to mention on YouTube, Coursera, Udemy and other similar sites. There are plenty of free textbooks online. Yet we have the persistent issue of accreditation and signaling, but hopefully that is starting to change. Why should education continue to inflate?


Healthcare is so deeply dysfunctional, a simple app that exploits the opaque contracts among pharmaceutical companies, PBMs and retailers can lower the cost of drugs for up to 80 percent, but few people—even doctors—know about it. Healthcare is yet another example of regulatory capture getting in the way of progress. There are thousands of startups trying to make a dent.




Housing has resisted the advancement of technology, but for how much longer? It seems inevitable that we should be able to churn out dwelling units like we do paper clips. But actually building in places people want to live is a tough problem because so much wealth is controlled by the landed gentry who don’t want development in their backyards. Again, a political, not technological problem.



But if Baumol is right, the correct answer is that these are not good measures of progress.


Maybe in the future, we will spend all our time learning with one-on-one tutors, supplemented by digital lectures and problem sets; we’ll have high-quality, labor-intensive healthcare (replete with human psychiatrists and massage therapists); and many of us will live in tiny apartments in dense urban areas.


That, though, doesn’t sound like progress to me. It sounds like stasis and mediocrity. Maybe the answer is we just need to expand into new frontiers to shake us out of complacency.



Star Trek's idea of San Francisco in the future


In 2011, Marc Andreessen famously wrote an op-ed explaining why software was eating the world. In 2020, he wrote another piece, this time exhorting us to build all the amazing things we can’t seem to:


The problem is desire. We need to *want* these things. The problem is inertia. We need to want these things more than we want to prevent these things. The problem is regulatory capture. We need to want new companies to build these things, even if incumbents don’t like it, even if only to force the incumbents to build these things. And the problem is will. We need to build these things.
[…]
It’s time for full-throated, unapologetic, uncompromised political support from the right for aggressive investment in new products, in new industries, in new factories, in new science, in big leaps forward.

This would certainly be progress on my dashboard.