Why do some people claim that the end to economic growth is over or very near? One popular explanation goes this way. Economic growth requires the use of natural resources. On a limited planet, resources are limited. Therefore, when we have used up all our natural resources, economic growth will come to an end. QED.
That explanation belongs in primary school children’s books with illustrations that include smiley faces, but is not suitable for children past their 10th year.
Though quite popular among many biologists (Paul Ehrlich), politicians (Al Gore), gurus (Sadhguru Jaggi), engineers (pick your favorite), and other assorted crazies, that explanation is wrong. The idiotic notions underlying the false explanation were laid to rest by Julian Simon and others many decades ago. But few people have the patience, the capacity or the desire to read Simon.
If you’re a regular reader of this blog, you already know the answer. For the newbies, let me summarize the basic argument. But first, go read part 1 of this post if you haven’t read it yet.
There are no “natural” resources. All resources are artificial. We create resources from existing materials, which are too abundant to be ever exhausted, using our brains. Our ability to figure out how to get things done is what I define as technology.
Technology continually expands the resources available for our use. Even if the natural stuff were limited — which is definitely not the case — we can have indefinite growth. How? By transforming the available stuff into more resources. More resources means more potential for growth. If we wish to grow, there’s no limit to growth.
It’s useful to distinguish between extensive and intensive growth. If you use more land and more labor to produce more food, you have extensive growth in the production of food. When you run out of additional land and labor, you run out of your ability to grow. That’s a limit to growth.
But if you use better technology what increases the productivity of land and labor, you have intensive growth. The use of better fertilizers and more efficient farming tools and machines increases yields, and that’s growth that is not limited by land or labor. Machines are the important bit.
Machines are the product of technology. Humans have figured out how to make machines. We are a tool-making and tool-using species. We have invented lots of tools, and graduated to making machines that make machines. The lathe is probably the most important machine ever invented — because it is the great-great-great- … -great-grandfather of every machine tool today.

Here’s a picture of a direct descendant of the wooden lathes of hundreds of years ago — a precision CNC lathe.
The Archaic and the Modern Era
Human history can be neatly divided into two distinct eras. The division can be almost precisely specified. The Archaic World began with the birth of the human species around 300,000 years and ended around the year 1750 CE. The Modern World began just around 250 years ago and it may well last till the human species dies.
All species, just like all individuals, eventually die. About 99.9 percent of all species that have ever existed on earth have gone extinct.
The essential distinction between the Archaic and the Modern eras is that in the former, growth was limited to extensive growth alone, while in the latter, intensive growth became possible. Which meant, more output for the same material inputs.
The same material inputs (land and labor) but a lot more mental inputs.
In the Archaic World, for instance, one person would be able to spin a kilo of yarn a day. In the Modern World, one person working in a modern spinning factory can spin 4,000 kilos of yarn a day. Productivity increases of three orders of magnitude.
Today every human enterprise is orders of magnitude more productive than it was in the pre-1750s world. And never forget, today we have things that were not available at all at any price. Creature comforts we take for granted were unimaginable to even people 100 years ago, forget thousands of years ago.
All that was made possible by the First Industrial Revolution. It was a qualitative change, not just a quantitative change. Economic growth rates shot up from close to zero percent prior to the mid-1700s to average around 2 percent in the next couple of hundred years, and in the last few decades, some countries like China attained sustained double-digit growth rates. That lifted billions of people out of dire poverty.
“Yeah, yeah,” some says, “but you fail to see that this growth is unsustainable. You’re like Taleb’s turkey which thinks that life is fun with all the free food and free medical attention — till it all ends in early November.”
That kind of talk involves two kinds of misunderstanding. First is a misunderstanding of what Taleb’s point about “black swan” events — events that are rare (black swans are rare since nearly all swans are white), and which have major unanticipated negative consequences.
Taleb’s turkey did not anticipate the ax. Why? It does not understand the situation. Meaning its explanation why the farmer was feeding and caring for it was false. Turkeys are incapable of explanations, true or false. But we humans can come up with explanations, and even true explanations. How do we distinguish between true and false explanations? Through reason.
Which brings me to the second misunderstanding — the inability to understanding the logic of the argument that explains what economic growth is and what are its causes. If economic growth were only extensive, then of course indefinite growth is impossible on a foundation of limited resources.
But resources are not limited because resources can be indefinitely “manufactured.” But will they be indefinitely manufactured? Yes. Why? Because there will be more minds, and therefore there will be more ideas.
In the Modern Era, economic advancement is both due to extensive and intensive growth — the latter primarily due to the rapid development of technology.
Let’s quickly review three distinct revolutions. The first is the British Agricultural Revolution between the mid-17th century CE and last 19th century CE which increased labor and land productivity. This allowed the labor released from working in farms to be absorbed into industries that fueled the Industrial Revolution.
Gradually labor shifted from agriculture to industry. Then increases in industrial productivity released labor from manufacturing to the service industries. Labor participation grew in services, particularly in the research and development in the scientific, engineering and technological sectors.
Take the US for instance. In the year 1800, farm labor accounted for 60 percent of total labor; now it is less than 1%. What do the other 99% do? Some are in manufacturing (8%) and the majority in services (80%.) There are millions of people employed in the knowledge creation in various industries and educational institutions. They create technology. That technology then gets applied to increase the productivity of people and materials.
What does increase in the productivity of materials mean? It means we get more utility out of the same of material than before. The efficiency increases in, say, automobile and aircraft engines have allowed more miles per gallon of fuel. You have more computing power in your pocket than used to be available to giant corporations with their “super computers” just a generation ago. Allow me to quote myself —
Around 40 years ago, there used to be supercomputers at the most prestigious institutions in the most advanced industrialized nations. They were housed in huge buildings; they required an army of highly trained technicians to build, operate and maintain; they cost hundreds of millions of dollars; and the energy they used cost thousands of dollars per month.
If you had claimed in 1980 that in 2020 everyone would privately own a supercomputer, the first argument against you would be this: a supercomputer uses 30 metric tons of material. So 5 billion supercomputers would require 150 billion metric tons of material, and of course more than 60,000 times the energy to operate them than was being produced globally in 1980, and around 100 billion trained technicians (assuming only 20 of them per supercomputer) to operate them.
You would be asked: Are you stupid and retarded, or are you merely insane for suggesting that in just 40 years there would be 5 billion of 1980 supercomputers on earth?
In reality, humans figured out how to design and manufacture faster, cheaper, smaller integrated circuits that used orders of magnitude less energy than before, and packaged them into little portable devices we use without a thought. Now the equivalent computing power of those supercomputers is available in devices that cost a few hundred dollars, that use only a few cents worth of energy, that people carry in their pockets, and even people who cannot afford indoor plumbing routinely use “supercomputers” in their daily lives.
[Source: Finite Systems, Infinite Cycles.]
The basic misunderstanding people have is that they don’t understand the power of human ingenuity, the power to create. Humans have been inventing things and discovering how the world works for thousands of years. In the past, the pace of invention and discovery was very slow — there weren’t that many people and most of them used to work in farms.
But now there are billions of people, and hundreds of millions of people are engaged in invention and discovery. On top of that, technology growth is a positive function of the stock of technology: the more technology you have, the greater the pace of technology creation.
People have been predicting the end of technological advancement since forever. It is claimed that in 1889, Charles H. Duell who was the Commissioner of US Patent Office 1898 to 1901, said that the patent office should be shut down since “Everything that can be invented has been invented.” The wiki says that that’s untrue. In fact, in 1902, Duell said something that is consonant with my point of view:
“In my opinion, all previous advances in the various lines of invention will appear totally insignificant when compared with those which the present century will witness. I almost wish that I might live my life over again to see the wonders which are at the threshold.”
He was right. I believe that the advances of the past 250 years will pale into insignificance compared to what is going to happen in the next 25 years. Most of the readers of this blog will be around to marvel at that. I believe that the world is at the threshold of an Age of Infinite Abundance.
The magic word was “plastics” in the movie The Graduate.
I will tell you the magic word of the future. Are you listening? Energy. There’s a great future in energy.
In the next bit, I will discuss the “objects and ideas” gap. And also answer the questions that followed the last Ask Me Anything. Until then, thank you, good night and may your god go with you.
I apologize for putting this unrelated comment in this post. There was no AMA in the recent past, and I was apprehensive whether you will notice a question in an old post. Hence this unrelated question:
I am confused about the correctness of government interference to break monopolies. Sometimes I think this is good. I do believe that a free market without competition is terrible. But if the government starts deciding what a “monopoly” is and what is not, we have just let in a thin wedge that can corrupt free markets beyond any limit. But then, if the government does not break monopolies, who will? Can the free-markets self-correct? Has it ever happened in practice?
Teacher Atanu, your thoughts, please.
LikeLike
Atanu, you seem to have lost touch with ground realities. Yield-per-hectare has already peaked and plateaued in the most agriculturally productive regions of the world: Japan for rice and France for wheat. Yes, growth in the yield-per-hectare is, by definition, intensive, but there is an ultimate, not very high, ceiling to this growth: photosynthetic efficiency is a mere 2% (for comparison, efficiency of current poly-crystalline-silicon solar panels is 11-15% ).
The fact that performance metrics of computing machinery have grown by several orders of magnitude in a short period of time, provides no hope that energy production can likewise be scaled up on similar time scales. Again, yes, in theory there is a lot of energy “out there” (the solar energy that falls on the surface of the earth per year is greater than the entire fossil energy used), but in practice, fossil fuels are the only abundant energy source with EROEI > 30. And if such a high quality energy source gave us 2-3% GDP growth historically, and 8-10% GDP growth in China during the last decade (driven by intensive usage of coal), it is anyone’s guess what the growth rate will be if EROEI falls below 20 and then 10!
Growth of monetary assets or the money supply without a corresponding increase in energy production will only result in inflation. This is already the case, with all of the extreme inflation funneled into financial assets and real estate, the price of food and energy deliberately kept low to avoid food/fuel riots. But, for how much longer?
As an economist, it is irresponsible for you to ignore ground realities and instead fantasize in pipe dreams of unending growth in energy production.
LikeLike
This is a classic case of focusing on what we can measure while ignoring what we can not.
Genetically modified plants might get better at photosynthesis. If not, we might have vertical farms that will not need any sunlight at all but will use LED lights with frequencies specifically targeted for those plants. Not just that but because now we have artificial sunlight even at night, we might be able to grow food twice as fast in a 100-story building.
Even without vertical farms a few innovations here and there can enable us to convert the Sahara desert into lush green farmland or the desert of Nevada into one of the most fertile regions in the world.
I do not know what other problems humanity will face down the line but my money is on the fact that food will not be one of them.
LikeLiked by 1 person
Regarding GM plants with better photosynthetic efficiency, is there even a lab demo for this? The dwarf-varieties of wheat and rice introduced by Norman Borlaug and others were already a huge step-up in energy conversion efficiency. It will be very hard to make further improvements over these. All the patented GM strains in recent years are for higher resistance to disease and herbicides, none for higher yields (see the Lester Brown article below for reference).
For cereal (energy giving) crops, with modern intensive techniques, the EROEI (energy returned on energy invested) is about 2-3. With traditional agriculture, nitrogen from recycling cow/buffalo manure, and labor energy from oxen/buffaloes, EROEI is about 5-6. So, it is very likely that EROEI for vertical, solar powered farms will be negative once the embedded energy and depreciation for the structures and the additional energy needed to pump water are factored in.
You might not need any new innovations, just a whole lot of pumped desalinated water. Again, EROEI could be negative or just break even.
Food has already become the first limit that we have hit, so be careful about what you bet your money on. The ground reality is bad. See for example Lester Brown’s summary here:
Could Food Shortages Bring Down Civilization by Lester Brown (2009)
LikeLike
Is there any evidence for this claim ? We are producing far more food than we need. Modern food is of higher quality and offers more choice to consumers than ever in history.
Even without any disruptive innovations we can grow far more food that our current capacity.
You are too fixated on EROEI.
I am not an expert in energy field but in general variables like these are not predictable over long time. Future gets impacted disproportionately more by unpredictable step changes in technology rather than predictable trends.
Nearly all doomsday predictions around oil and coal have turned out to be false over last 100 years.
LikeLike