Article
The most mocked prediction of the 1970s was right. A KPMG analyst proved it 50 years later — and nobody noticed.
In 1972, a team of MIT researchers published a book called The Limits to Growth. It ran a computer model — cutting-edge at the time — to simulate what happens when exponential population and economic growth meet a finite planet. The conclusion, in their words: "If the present growth trends in world population, industrialization, pollution, food production, and resource depletion continue unchanged, the limits to growth on this planet will be reached sometime within the next one hundred years. The most probable result will be a rather sudden and uncontrollable decline in both population and industrial capacity." [1]
The book sold 30 million copies — and was savaged by economists, politicians, and pundits as doomsday fantasy. Bjørn Lomborg consigned it to the "dustbin of history" in 2002. [2] Critics claimed it had predicted resource exhaustion by 1990 and been spectacularly wrong (it did not; they never predicted any specific date for resource exhaustion). The attack was so successful that even people who'd never read the book "knew" it was the one where the computer was wrong. [3]
Then, in 2021, a KPMG analyst named Gaya Herrington decided to check.
What the model actually said
The World3 model tracked five interconnected variables: population, industrial output, food production, pollution, and non-renewable resource consumption. [1] The authors ran multiple scenarios with different assumptions about human behavior:
- Business as Usual (BAU): no major policy changes; growth continues until resource constraints trigger collapse.
- Business as Usual 2 (BAU2): a revised 1992 version of the same.
- Comprehensive Technology (CT): assumes aggressive technological solutions — 4% annual efficiency gains, pollution reduction, yield improvements well above historic averages.
- Stabilized World (SW): adds social value shifts on top of technology — smaller desired family size, perfect birth control, deliberate limits on industrial output in favor of health and education. [4]
The collapse scenario (BAU) had industrial output peaking around 2015-2020, then declining. Population would peak around 2030 and fall by about half a billion per decade. [2]
What the data showed
Herrington, as part of her Master's thesis in Sustainability at Harvard Extension School, compared real-world data from the UN, UNESCO, FAO, NOAA, and BP Statistical Review against all four scenarios. [4]
The results, published in the Yale Journal of Industrial Ecology, were stark: real-world data tracked BAU2 and CT most closely through 2020. The Stabilized World scenario — the only one with a comfortable outcome — tracked least closely. [4]
Turner had found the same thing in 2008 and again in 2014: "The data doesn't match up with other scenarios" — only BAU. [2]
In other words: a peer-reviewed study by a Big Four consulting firm found that the most ridiculed prediction of the 20th century had been tracking reality for 50 years.
What falls apart, and when
According to the business-as-usual scenario, resource extraction gets more expensive as easily accessed deposits are exhausted. More and more capital goes toward digging, drilling, and refining — and less goes to everything else. Industrial output per capita starts falling around 2015. As pollution mounts and industrial inputs to agriculture decline, food per capita falls. Health and education services get cut. Death rates begin to rise around 2020. Population starts declining around 2030. [2]
The 2008 financial crisis and the persistent economic malaise that followed? That's consistent with what the model predicted would happen as resource constraints began to bite. The Guardian piece co-authored by Turner in 2014 argued that "the pursuit of material wealth contributed to unsustainable levels of debt, with suddenly higher prices for food and oil contributing to defaults — and the GFC." [2]
But isn't this the one that got resource prices wrong?
A common counter-argument: Limits to Growth predicted resource scarcity would drive prices ever upward, yet many commodity prices fell through the late 20th century. This is true — and the model did not fully anticipate how technology would unlock new reserves (fracking, deep-sea drilling, previously uneconomic ore bodies).
Herrington's paper addresses this head-on: the Comprehensive Technology scenario assumes 4% annual efficiency gains. Actual technological progress has been closer to 1.5%. We are running the CT scenario at less than half the required innovation rate. That is not a good place to be. [4]
The CT scenario still shows industrial decline — just later than BAU. And the technology assumptions needed for CT are sorely optimistic given actual rates. [4]
The 50-year verdict
Herrington's conclusion was not a prediction of certain doom. It was a quantification of the gap between where we are and where we need to be:
"A deliberate trajectory change brought about by society turning toward another goal than growth is still possible. The LtG work implies that this window of opportunity is closing fast. Human society still has the capacity to shape its own future — but that capacity has boundaries." [4]
What the 1972 researchers got right was not a specific date or a specific resource. What they got right was the shape of the problem: that exponential growth in a finite system produces overshoot, and that the longer you wait to correct course, the harder the correction becomes.
The most mocked prediction is not wrong. It is, 50 years later and multiple peer-reviewed studies deep, the closest thing to a verified warning that economic modeling has ever produced. That the public narrative never caught up with the science is, in some sense, its own kind of validation.
Sources:
- Wikipedia — The Limits to Growth: https://en.wikipedia.org/wiki/The_Limits_to_Growth
- The Guardian / Graham Turner & Cathy Alexander — "Limits to Growth was right. New research shows we're nearing collapse" (2014): https://www.theguardian.com/commentisfree/2014/sep/02/limits-to-growth-was-right-new-research-shows-were-nearing-collapse
- Resilience.org / Andrew Curry — "Revisiting The Limits to Growth" (2021), summarizing the Gaya Herrington KPMG paper: https://www.resilience.org/stories/2021-08-16/revisiting-the-limits-to-growth/
- Herrington, Gaya — "Update to limits to growth: Comparing the World3 model with empirical data" (2021, Journal of Industrial Ecology 25(3): 614–626; submitted-version PDF): https://canadiancor.com/wp-content/uploads/2021/10/Herrington-Limits-2021.pdf (peer-reviewed version: Wiley, doi 10.1111/jiec.13084)
- Wikipedia — World3 model: https://en.wikipedia.org/wiki/World3