This year’s Nobel for economics is a technology prize!
On Tuesday, the Swedish Academy awarded the 2018 Nobel Prize in economic sciences to two American economists, William Nordhaus of Yale and Paul Romer of New York University’s Stern School of Business. Romer is well known for his work on innovation, and although the committee focused on Nordhaus’s research on climate change, this year’s prize is really all about technology and its central role in economic growth.
Romer’s 1990 paper Endogenous Technological Change is one of the most famous and cited of the last several decades. Until then, the foundational theory of economic growth was Robert Solow’s model. It said growth was the result of varied quantities of capital and labor, which we could control, and a vague factor, known as the Residual, which included scientific knowledge and technology. The Residual exposed a big limitation of the Solow model. Capital and labor were supposedly the heart of the model, and yet technology accounted for the vast bulk of growth — something like 85 percent, compared to the relatively small contributions of capital and labor. Furthermore, technology was an “exogenous” factor, outside our control, which didn’t seem to explain the real world. If technology was a free-floating ever-present factor, equally available across the world, why did some nations or regions do far better or worse than others?
Romer’s key idea was that technology — or more generally, ideas — aren’t some automatic or magical force outside our control, as the Solow model assumed. No, technology is the result of focused effort, often by for-profit firms and entrepreneurs, operating in dynamic but not perfectly competitive markets. Technology isn’t some unseen governor on the game of growth. No, technology is the whole game. Unlike capital and labor, which exhibit diminishing returns, moreover, ideas could generate increasing returns. Or as David Warsh summarized: “The more you learn, the faster you learn new things.”
Romer wasn’t the first to emphasize the importance of technology, entrepreneurship, research and development (R&D), or even increasing returns. Adam Smith and Jean-Baptiste Say offered glimmers long ago, and Joseph Schumpeter had the basic idea with “creative destruction.” But Romer formalized the model of technological change in both broad outline and deep math.
Romer also emphasized and foretold the centrality of non-rival, partially excludable goods, such as software, which today make up so much of the economy. (Remember the matrix of rivalry and excludability.) If I sell you software or a pharmaceutical, I still have the code and can sell it to others at near zero marginal cost. It’s only partially excludable, however, because the code or formula can be copied. Intangible goods are thus easy to share but also prone to illegal copying.
In an excellent short video, Alex Tabarrock offers a concise summary: Romer, he narrates, “built a model based on ideas, privately produced by for-profit firms, in monopolistically competitive markets with spillovers.” Max Roser tweeted that Romer told him it was this chart, of the rapid rise of living standards associated with the Industrial Revolution after millennia of stagnation, that got him into economics. I mention Romer in my own take on “entropy economics,” and I loved his work on “charter cities” in part because I’d written a paper on the success of China’s special economic zones.
If you want the long version of Romer’s story, Warsh’s terrific 2006 book, “Knowledge and the Wealth of Nations,” is the most comprehensive popular treatment, where Romer is the star, situated within the history of research on economic growth. And for a characteristically good roundup of Romer and all this week’s news, see Tyler Cowen.
Professor Nordhaus’s great contribution to the environmental debate was to ground policy options within an economic framework of realistic trade-offs. He began the work in the 1970s, and if more people had listened, perhaps the climate debates over the last several decades would have been more productive. More light, less heat, as it were.
Technology is implicitly at the heart of Nordhaus’s environmental work. Nevertheless, my own favorites of Nordhaus are the explicitly technological papers.
The more famous is his chapter on the “History of Lighting,” from 1996, which examined the cost of varied technologies to generate light over millennia — in open fires, tallow candles, oil lamps, incandescent bulbs, etc.
Nordhaus found that the “true” price of light was very different from the “traditional” price of light, as measured by official government indexes. Piecing together various indices, he found the traditional price of light had risen by a factor of between 3 and 5 over the past 200 years, which represented far less inflation than most other products. By measuring the actual luminosity and actual prices of various technologies over the period, however, Nordhaus estimated the true price of light had fallen by a factor of around 800 — a gigantic difference.
In another paper, “Two Centuries of Productivity Growth in Computation,” Nordhaus sought a better understanding of the true price of information processing. Focusing on the performance of computing technologies and not merely the prices of the physical components, he found:
Depending upon the standard used, computer performance has improved since manual computing by a factor between 1.7 trillion and 76 trillion . . . [and that] the price declines using performance-based measures are markedly larger than those reported in the official statistics.
In my own report on the 50th anniversary of Moore’s law, using Nordhaus’s framework, I estimated that in the 50 years of Moore’s law to 2015, the labor cost of computing had dropped by a factor of around one trillion.
Each of these less-heralded Nordhaus papers, focusing on the real prices and real performance of real gadgets based on big ideas, pointed toward the truly radical nature of technology, which he summarized in the lighting paper:
Quantitative estimates of the growth of real wages or real output have an oft forgotten Achilles heel. While it is relatively easy to calculate nominal wages and outputs, conversion of these into real output or real wages requires calculation of price indexes for the various components of output. The estimates of real income are only as good as the price indexes are accurate.
During periods of major technological change, the construction of accurate price indexes that capture the impact of new technologies on living standards is beyond the practical capability of official statistical agencies. The essential difficulty arises for the obvious but usually overlooked reason that most of the goods we consume today were not produced a century ago.
Don Boudreaux has more on this Nordhaus view that we underestimate the power of technology, here discussing a kind of unmeasurable Schumpeterian innovation “surplus.” Here, once again, is Tyler Cowen’s roundup.
The technology link
Climate hawks correctly cite Nordhaus as one of their own. And yet if one looks closely at his latest and most detailed models, they suggest today’s aggressive policy proposals would be highly counterproductive. As Bjorn Lomborg wrote in the Wall Street Journal this week,
Mr. Nordhaus’s most recent estimate, published in August, is that the “optimal” outcome with a moderate carbon tax is a rise of about 6.3 degrees Fahrenheit by the end of the century. Reducing temperature rises by more would result in higher costs than benefits, potentially causing the world a $50 trillion loss. It’s past time to stop pushing so hard for carbon cuts before alternative energy sources are ready to take over.
The real solution, therefore, is not severe limitations on existing efficient energy sources and subsidies for today’s inefficient sources. It is instead the invention of new technologies, the economic potency of which, as Romer and Nordhaus brilliantly described, will likely surprise us.