The Great Stagnation (Tyler Cowen)
The polymath creativity guru Edward De Bono prescribes what he calls ‘provocations’, in order to stimulate lateral thinking and out-of-the box problem solving. Provocations are unusual, sometimes obviously wrong conceptions or solutions to a problem that facilitate us in breaking out of conventional thinking. Tyler Cowen’s The Great Stagnation can be seen as such a provocation - a 90 page book initially only available in digital form, light on detailed policy prescriptions, heavy on new thinking. Cowen is a libertarian economics professor at George Mason University, who runs the wildly popular bog Marginal Revolution (and has posted on it daily since 2003), and the podcast Conversations With Tyler. I highly recommend both.
The central argument of The Great Stagnation, is that the expectations of growth in the United States are unduly high, as a result of special circumstances leading to higher growth in the 300 years prior to 1973. The consequences of these overly optimistic expectations are political dissatisfaction, and an overoptimism that led to the 2008 financial crisis. Of course, the implication of Cowen’s argument is that the 3% annual growth considered as merely average can longer be achieved.
Why was growth abnormally high in that period? According to Cowen, the U.S. was able to exploit low-hanging fruit in three forms, fruit which has now been all but consumed. The first of these is land. Since its inception America was blessed with vast tracts of fertile land (often stolen from Native Americans, it must be added), allowing for large and constant immigration, and exploitation of which led to significant growth. Critics of the book point out that ‘land’ hasn’t been a particularly significant factor in growth since the 19th century, and that the data shows much higher productivity in the cities than rural areas. This particular factor excludes from the narrative other countries at the technological frontier, to which the thrust of the argument can be applied; Britain, Germany, France etc. Despite these criticisms, comparisons with Europe in the same period certainly point to America’s resource advantage as a factor in its enormous economic success.
The second factor is education. At the start of the 20th century, almost no young Americans—a mere 6.4 percent—graduated from high school, and well under 1 percent went to college. In 2011 (when the book was published), three-quarters graduated from high school, and about 40 percent enrolled in college. This is the very definition of low hanging fruit. Millions of intelligent adults went uneducated, and changing that brought massive productivity gains. Once most adults are given between 10 and 20 years education, it is difficult to make meaningful gains. Of course, improvements can be made at the margin, but not of the magnitude of going from no education to universal education.
Thirdly, Cowen argues that new technologies and inventions, the main drivers behind economic growth, are no longer as productive as historically, particularly the second half of the 19th century. This is a more contentious assertion. The transformative technologies of the industrial revolution and early 20th century, from steam engines to cars to radio to electricity and more, clearly transformed the developed world; the question is whether such progress has stagnated. Cowen argues that those inventions occurred mainly when amateurs were able to invent paradigm-changing technologies, before the electrification and complexification of the modern world. He later shows that the area in which amateurs are still able to break ground is the internet, Mark Zucherberg being the quintessential example. This argument is a version of the one made by Peter Thiel, the libertarian tech-billionaire cited by Cowen as an influence. He makes the full argument in his book Zero To One (published three years later) but had made it in interviews and speeches prior to 2011. His argument is that great progress has been made in the world of bits (online by programmers, able to create programmes from anywhere), but not in the world of atoms - real, physical inventions. Thiel’s argument is nakedly libertarian - that government regulation is the barrier in the world of atoms, but the free for all of coding has shown what people can do if you leave them to create. Cowen avoids politicising his argument in this way. Instead he contends that we are simply less inventive, and that our progress has resulted in more private rather than public goods. Goods that benefit the individual but less so society.
Median family income, which more than doubled between 1947 and 1973, increased by less than one-quarter between 1973 and 2004. Cowen demonstrates how these figures may in fact be worse, since government spending has increased and is counted at cost price, unlike say, apples, who’s price will fall if quantity is increased. He shows how both education and health spending likely have lower productivity at the margin, but proportional spending has been counted in GDP figures at cost. The countervailing force in this distortion is the internet. Since its products are mostly free, much of the benefits are not counted in GDP figures. Someone may consume an hour of entertainment on Tik Tok with no transaction taking place, despite the benefit to the consumer.
Cowen dedicates a specific chapter to the internet, as the one exceptional breakthrough since 1973 that runs counter to his narrative. Cowen is aged 48 in 2011, and he says, the majority of the technology in his life is the same as when he was born (with the exception of the internet), something which his grandmother, born at the beginning of the 20th century cannot say. To what degree this is true is debatable, but even taking it at face value, ‘the internet’ is a meta-technology, akin to the wheel, spawning countless other inventions, ideas and businesses.
This chapter is perhaps where the logic of the argument is most powerful in 2011, and much less so 9 years later. The assertion that the internet benefits society immensely but is largely free and so doesn’t generate wealth, is not being used here as a criticism of the internet. Instead Cowen is making the case that the benefit to society, and the parallel shift toward meaningful activities and careers is a positive thing, but is ultimately expensive and doesn’t contribute to growth. Further, the fact that companies like Facebook and Google, some of the largest in America only employ a few thousand people shows how the internet is not the saviour technology, economically speaking. We will return to the latter point, but the former idea of the internet being free and not revenue generating is surely a transitory one, and already false. It appears to me that the book was written between the period of internet growth and the point at which companies have figured out workable business models, which now so many surely have. Amazon, Netflix, a plethora of video games, Uber, AirBnB, Deliveroo and most newspapers and magazines have transitioned into viable internet businesses.
In favour of Cowen’s argument is the growth rate of countries who are not at the technological frontier. As India and China develop, they are able to skip several steps in development, moving immediately to the latest ideas in transport, communications, production and finance. The countries doing this have experienced spectacular growth rates by the standards of developed countries, China being the most obvious example. Even forcing structural changes, such as Stalin’s rapid transition to an industrial urban economy rather than a rural agrarian one during his Five-Year Plan(s) yields growth rates that are unachievable while an economy is in the process of discovering the best path to growth, as the developed nations always are. An idea as transformative and implementable as universal education is of course a rare occurrence.
That the over optimism stemming from historical growth led to the financial crisis is well argued. Mortgages may have been the first bubble to burst, but this was because that market was stretched thinnest, not because it was the only over inflated market. That banks fell into turmoil was a result of their spectacular levels of leverage, not because they were the only reckless actors in the economy. All likely true, and not contradictory to the obvious corruption and hubris of banks as a tributary factor. This same sentiment of expected long term growth is an explanation of the polarisation of politics, and the general dissatisfaction with democracy, argues Cowen. If 2011 Tyler Cowen could see the political landscape in America today, no doubt he would be astonished with how much further this polarisation still had to travel.
Is Cowen’s answer then, to simply accept the new gloomy paradigm? Not entirely. He does state that we should recognise the political malaise and not add to it. Japan, which has suffered economic stagnation since the early 1990s has certainly dealt with the problem culturally and politically far better than the U.S. (or indeed Europe), but prescribing stoicism where there is none is unlikely to be an effective solution.
On the positive side, China and India’s catch-up growth will both increase the quantity of people pushing the technological boundary, and provide a massive new market for anyone in the U.S. able to create valuable new innovations. The return on new technologies will be immense in a globally wealthy, open and connected world. Further, the internet will democratise new ideas by keeping isolated entrepreneurs at the forefront of new developments; one doesn’t need to attend Harvard or MIT in order to access the latest scientific papers.
In addition to not adding the political malaise, Cowen’s other suggested responses are to improve the standing in society of scientists, and to be wary of our response when the next low hanging fruit arrives. He cites Singapore as an example of a place where science is sufficiently fetishised, and identifies totalitarianism as the sort of response to low hanging fruit to which we ought to be wary.
The Great Stagnation is a great provocation. Tyler Cowen spends his days consuming and disseminating information, including astounding levels of preparation for his podcast guests. Such a tiny book throws out arguments with minimal unnecessary justification, making it more enjoyable, safe in the knowledge that the ideas are themselves well researched. The solutions are vague and platitudinous, and deliberately so. ‘We should increase the prestige of scientists’ is a truism Cowen hands to the reader with the unspoken adjoiner ‘good luck with that’.
My own response to Cowen’s supposition are as follows. Firstly, we have no idea whether technology and innovation will be better or worse in the next century in comparison to the previous three. Both innovation and its effects are non-linear, and that is true of even reasonably established ‘new’ technologies. Electricity didn’t immediately change 19th century Britain because mills were set up for water power. Only once new factories could accommodate many individual motors in each machine (rather than a single large power source) could the benefits to productivity be realised. Conversely, the power of the internet is being harassed more effectively today than 11 years ago, but in another 11 years that effectiveness could explode. We just don’t know.
Secondly I strongly believe that any and all innovation that improves efficiency is a good thing for growth, including those that make companies able to produce more with fewer workers. That these inefficiencies have a negative effect on growth by generating unemployment (thus lowering demand) is a genuine problem we must solve. The solution is not to stay less efficient, or to unnecessarily employ someone. As Cowen states, successful economies often rise to prosperity through exports, because “the external world market provides a real measure of value...not based on privilege, connections, corruption or fakery.” The ‘truth’ a real market exposes ideas to must always prevail.
The fact that Google and Facebook are the Fords of the modern era, but are unable (unlike Ford) to provide massive numbers of well paid jobs is a feature of digital technology and it’s inescapable tendency towards network effects and power laws.
So, we know that individual companies can create massive value whilst remaining small in terms of workers. We know that we wish to increase the levels of innovation amongst the population, and we know that traditional education is at saturation point. My solution would be to give the adult population both the time and willingness to take risks, using a Universal Basic Income of some sort. This would allow companies to push for maximum efficiency without fear of unemployment destroying the market for their product, and said efficiency wouldn’t be offset by the population being either unemployed or employed in unnecessary ‘workfare’ type jobs. Decoupling real efficiency and individual poverty is the solution to unlocking the new low hanging fruit that is the internet. Read The Great Stagnation for provocations of your own.