"Infinite Growth is a Pipe Dream" by Vaclav Smil

Financial Times, 8 August 2019

There are only two groups of people who believe in infinite growth: mainstream economists and people awaiting the arrival of the Kurzweilian "singularity", when the merger of biological and nonbiological entities will create immortal software-based humans. 

Economists' expectations are more modest but most standard forecasts of gross domestic product do not see economies following the trajectories of all living organisms: rapid, then slower growth, followed by a plateau and eventually some sort of demise.

Kenneth Boulding, who headed the American Economic Association in 1968, offered a pungent take on this point of view: "Anyone who believes in indefinite growth in anything physical, on a physically finite planet, is either mad or an economist."

I would not presume to join any intra-economic disputes, I just want to call attention to the fact that many current growth forecasts contain a perfect example of what philosophers call a category error: when attributes belonging to a specific category are presented as if they apply more broadly.

I am speaking of Moore's law, which holds that the total number of components that can fit on a microchip is growing so quickly that overall processing power doubles rapidly. This extraordinary achievement has led us to assume that we can replicate such gains in other economic sectors and drive decarbonisation of energy, huge food production gains and a fourth industrial revolution.

This mistaken impression, which I call Moore's curse, is easily illustrated. In 1965, when Gordon Moore first projected the growth of "components per integrated function", he put the doubling period at just about one year. That rate slowed and processing power now doubles every two years, for annual exponential growth of about 35 per cent. This translates into improved performance and lower costs for products driven by microprocessors, be they computers or phones.

This leads to expectations that this growth can be matched in other sectors of the economy. That is a category error based on a misunderstanding of fundamental biophysical realities. Modern economies depend on an enormous range of inputs whose yields, performances and capabilities have been constantly improving; but only at rates an order of magnitude lower than the 30 per cent growth dictated by Moore's law.

Start with food. The introduction of new plants and increased use of fertilisers, herbicides and pesticides have boosted global crop yields since the 1960s by 3.2 per cent annually for wheat and 2.6 per cent for rice. Corn yields have risen 2 per cent annually since the 1950s. Chicken meat, the ascendant protein, has been produced with increasing efficiency, but since 1930 the annual gain in the US has averaged less than 1.4 per cent.

Looking at energy and materials, most of the world's electricity is produced by large steam turbo-generators. During the 20th century their efficiency improved by about 1.5 per cent annually. Since the 1880s, indoor electric lighting has been through several revolutions (incandescent, fluorescent, halogen, LED) but annual efficiency gains have averaged about 2.6 per cent. Steel remains the dominant metal. Gains in the efficiency of its production have averaged less than 2 per cent since 1950.

Turning to transportation, annual growth of the highest speeds of inter-city travel by rail tripled during the 20th century (to 300km/h), implying average annual gain of 1.1 per cent. Since the US introduced fuel efficiency rules in 1973, the average efficiency gains for cars (not counting SUVs and pick-up trucks) has risen 2.5 per cent. The speed of intercontinental travel shot up from less than 40km/h in 1900 (ocean liners) to 885km/h by 1958 (the first Boeing 707 to Paris). While that averages to an annual growth rate of about 5.5 per cent, the speed has not (Concorde aside) increased at all during the past 60 years.

Even these moderate rates of growth must eventually reach their specific plateaux. Dreams of singularity aside, we will not achieve infinite crop yields and we will not travel at the speed of light. 

Moderate growth, falling overwhelmingly between 1 and 3 per cent and mostly between 1.5 to 2.5 per cent annually, should be broadly reflected in the aggregates of economic outcomes. Indeed, it has been. Since 1950 the average growth of US GDP per capita (in real terms) has been about 2 per cent.

Keep that fundamental 1 to 3 per cent annual growth range firmly in mind the next time you hear how the latest innovation will mimic components crowded on a tiny silicon wafer: it will not, it simply cannot.

The author is an emeritus professor at the University of Manitoba and the author of the forthcoming book 'Growth: From Microorganisms to Megacities'.