How many people are necessary to maintain modern civilisation?

Question

Modern Earth; a series of connected events (natural disasters, mass migration, drug-resistant pandemic, collapse in biodiversity, armed conflict) have led to a significant reduction in human population over a single generation.

Does "civilisation" continue? Societies could certainly adapt to short-term disruptions—but if the complex, global, inter-connected supply chains upon which they currently depend do not recover sufficiently quickly, would they not collapse entirely?

For example, a number of contemporary industries (unfortunately) depend upon oil: not only is most transportation fuelled by petroleum; but most fertilisers, pesticides, pharmaceuticals and plastics are also derived from it. If oil extraction, refinery and distribution were to suddenly cease for a sustained period (before such dependent industries have identified and adapted to alternatives), oil reserves would deplete and those industries themselves could collapse. That in turn could lead to a collapse of further dependent industries, including the manufacture of machines used in virtually any given supply chain.

Do we have any idea how many people are required to keep "it" (the modern industrial world) functioning?

Please give due consideration to the fact that disaster-struck areas may be uninhabitable, with their populations migrating en-masse and seeking refuge in other areas—thereby placing further strains on the system.

---

Clarification

There have been some very probing and helpful comments, which have prompted me to spend some considerable time reflecting on exactly what my question is and this edit is an attempt to elaborate. My thanks to all those commenters who contributed to this discussion.

By "significant reduction in human population over a single generation", let's assume the reduction is unforeseen, uniform across all demographics and complete within 10 years. To be clear, it is an "unmitigated apocalypse".

By "maintain the modern industrial world", I meant maintain our current technological capabilities. It strikes me that a good measure of this is the smartphone—it requires mining and refining many minerals, including some that are so rare that viable ore has only been found in a few locations on the planet; even once refined, maufacture of some parts (e.g. 7nm-process semiconductor fabrication) is so specialist that the number of capable facilities can be counted on one hand; even once assembled and distributed, their function depends on an infrastructure of components (antennæ, switches, routers, servers, etc) that themselves require such parts and maintenance thereof. Other yardsticks of technological capability might include space exploration, subatomic research (particle accelerators, quantum computing, etc), artificial intelligence, nanotechnology, ...

Answer 1

The New Scientist article "Why the demise of civilisation may be inevitable" from Issue 2650, which is based on the work of Jared Diamond, in particular Collapse suggests that killing as few as 2% of the workforce could initiate a cascade of failures in infrastructure and transportation systems sufficient to bring down modern western civilisation; if those who died were key staff with unique skillsets and/or knowledge in their vital workplaces. But you don't even have to kill them, prolonged (read two weeks or more) absenteeism of a larger fraction of the workforce (as little as 10% in some sectors, up to 30% in more robustly staffed organisations) would have the same effect.

As to the bare minimum population you would need to maintain what we have, that's not really about total numbers so much as population density. Our current society is pretty well as complex and advanced as it can be at current levels of population density, kind of by definition. Maximum population density is largely determined by the efficiency of the growing, storing, and transporting of food to urban populations. In order to maintain what the west currently has we actually need most of the world to remain intact and active in the world commodity trading market or the raw materials aren't available to the urban-industrial centres that need them.

As a note if we get a bit further with manufacturing miniaturisation this question becomes almost moot as manufacturing and mining facilities become one and the same and potentially you can put the basis of one in your backpack.

Answer 2

I'm going to respond to this with a framing challenge, because as it stands I think you're asking the wrong question.

The issue is that the modern industrial world is a product of population DENSITY rather than total population. This may seem semantic but in a post-apocalyptic environment it's very meaningful. Modern industrialism depends on and benefits from massive economies of scale that accrue from having lots of people all close enough together to quickly and easily exchange goods and services.

In a post-apocalyptic scenario, if you're envisioning two million survivors scattered across the entire planet, they're all going to be spending pretty much all of their time just keeping themselves fed and sheltered, and it's straight back to a hunter/gatherer economy worldwide.

If, however, you're envisioning two million survivors concentrated in a specific area that (for example) quarantined itself before anybody could be infected with the zombie virus, then you'd potentially have enough critical mass of different kinds of expertise to allow cooperative maintenance and task specialization: e.g. you have a bunch of people whose entire job is providing the food for everybody. You have another bunch whose entire job is maintaining power/water/electric. You have another bunch who kill the zombies before they swim across the river.

This is basically how urban societies developed in the first place, and the same rules would apply in an apocalypse.

To more directly answer your question: The larger an area your population of survivors is spread across, the more people you need. The more widely distributed the resources you require are (food, water, factories, sources of raw materials), the fewer people you need.

The number could potentially be as low as a few hundred in a commune out in the middle of Nowhere, Idaho that already has its own self-sufficient sources of food, water, and power, and a smattering of skilled electricians, mechanics, and engineers.

Answer 3

I've used a personal analogy for this, which I call "The brake pad problem." I estimate (without any serious research) that maybe a few hundred people around the world really know what it takes to manufacture an automobile brake pad. If you kill them all off, we're in serious trouble at least until a new process (or a reconstruction of the original) can be completed. The idea is that everyone is incredibly specialized, and our outputs are very interconnected. This makes the whole system unstable. My WAG, then, is less than 5% of the people in industrial countries/regions would cause a complete collapse.

You might want to read Lucifer's Hammer for more ideas.

Answer 4

A modest shock would be enough to prevent maintaining "our current industrial world".

7nm process semiconductors would become economically infeasible with a sustained 33% global depression, let alone megadeaths.

Note, however, that "rare earth minerals" are not rare; they are rare as in "rarer than iron" or carbon or silicon. Rare earth minerals are processed where they are cheap to process, not because they are there.

And it isn't that Lithium or whatever other element is only found in a few spots, it is just amazingly cheaper to get from those few spots because there is lots of it and it is easy to get. Gallium is produced as the by-product of purifying other ores. Etc.

Our current industrial world is set up for a constantly growing set of "global middle class" to be economically feasible. If we get a significant shock, some pieces will shut down.

Something as basic as Oil or Potassium will continue to flow unless we are talking only trace amounts of humanity left. Random complex doo dads? Demand will collapse; even if people knew how to make it and had supplies, the massive number of microwave ovens sitting idle and survival being more important would make making them not worth it.

Supercolliders, well, it depends on how France is doing. That is where the best-in-the-world Supercollider is. AI? If things are bad enough, you stop caring, and you repurpose people working on it for something more important. Quantum mechanics? Barring quantum computers, not an industrial-civilization scale effort. Nanotech? Much of it has to do with developing bleeding edge materials science. If you have, again, more important things to maintain, you stop doing it. Space exploration? Ditto.

These are economic questions more than anything. You can get to space in a whole pile of different ways. Do you think it is worth it? Then sure, you can keep at it. Heck, we could sent 1000 people to Jupiter in 2 years if we seriously thought we had to with our current level of technology (Orion) with a non-zero chance of success. It just isn't worth the costs.

Collapse of global and internal trade flows would cause economic damage faster than it would make things impossible. That economic damage would make things infeasible before they become impossible.

Imagine 99% of the world population dying off except for one small region caused by neutron bombardment (so just dead people/animals/plants, buildings etc are fine). That 1% region would (a) be economically savaged, and (b) have a massive ROI from resettling/expanding far beyond doing nanotech research. So even if they had a nanotech research lab there and everything they need, they'd stop doing nanotech research and they'd start farming and clearing roads and getting gas supplies etc.

Answer 5

It's a tough question.

Someone mentioned population density above, and that's definitely a major factor. But I'd also mention two more: logistics and specialisation.

Civilisation requires specialisation - if everyone's hunting or in the fields, they've not got the time or energy to focus on more abstract things. So you need a way to give people time off from basic survival activities and a way to recognise and reward their activities, ideally in a way that leaves them better off than if they were still out in the fields.

So at the very least, you need to make sure they have sufficient food and water. Which means you need more people who specialise in producing more food than they consume. And that requires large amounts of land.

(There's some interesting discussion here which suggests a square mile per 180 people in medieval times:

https://www.reddit.com/r/worldbuilding/comments/6qocm8/question_about_population_density_and_required/

Quite how much that remains true for modern civilisation is debatable; for all we've vastly increased yields and reduced manpower overheads, we also have much higher expectations and use a lot of resources (e.g. fertilisers, pesticides) which may not be readily available in the event of a population decimation...)

And that brings us to logistics, as loved by great generals everywhere. Because you need to have a way to quickly and efficiently transport said food. And all the other resources the city needs, many of which require large amounts of space and/or involve processes inimical to people's health.

Multiply all the above by a few thousand times to account for the complexities of modern civilisation and divide by a few hundred to account for the impact of modern technologies such as the internet. And that's what you need.

So I suspect the answer would be uncomfortably close to the populations we currently have, unless the decline in population was slow enough to be managed. Which in a way is what we're doing now, given the decline in Western birth rate...

It might actually be possible to quantify things by looking at recession and inflation data. After all, the former is when we have too much slack in the system and the latter is when we don't have enough; somewhere inbetween is the sweet spot for our current civilisation!

Answer 6

The problem is that most of the work today goes towards maintaining a society that supports this many people. You would not maintain everything we currently have. Would you field a single aircraft carrier? Would you keep one airline running or would you optimize to horses and electric cars.

You wouldn't need or even want most of the stuff you mentioned (if any)

What you would do is re-apply and re-focus our existing knowledge:

Create cars and machines with lower tolerances so that they can use 3d printed parts.

Focus our 3d printer technology on metal fabrication.

Improve solar, wind and battery technology so that what little oil we need would be focused on lubricating.

Without so many people, why would anyone choose plastic? Wouldn't paper, glass and metal work better in most cases if they weren't so much more expensive (rare?) In this world they would be so plentiful as to be virtually free!

Optimize nearly all mass fabrication like cars for maintainability and easier changes to production with much smaller volume (No need to manufacture thousands of the exact same car, so you don't have to make special machines JUST to stamp out left fenders for a two year run of a single make of car).

I think the things that we consider "Civilization" from a normal person's point of view wouldn't change much from a small town today--you'd still have:

• houses (Not as many high-rise apartments, if any)
• restaurants
• better food
• transportation (Simpler cars, horses, small planes)
• Lots of old movies
• New entertainment (Movies, plays, youtube-style stuff).
• Space flight (Probably, thanks Elon!)
• Some form of computers
• Huge amounts of excess electricity because of vastly reduced demand even if we lose the existing infrastructure.

We would not have:

• giant buildings/boats
• mass produced Salisbury steaks.
• giant food chains
• Space station would probably be unmaintainable
• Such powerful/dense computers (We would probably have to recreate most of the technology based on something easier to fabricate)
• the ability to provide much support (medical, emergency) outside of the civilized areas

With this in mind, civilization could probably scale down to a few thousand people pretty easily--it's just that the form of that civilization would change completely (and from the point of view of our future selves--nearly all for the better--they would likely call us completely uncivilized!)

Answer 7

At least 25 million people.

North Korea has a population of 25 million people and is somewhat able to maintain our current technological capabilities. It’s not top-notch and they do receive some imports from other countries. I imagine with better management they could do much better.

Answer 8

I'd like to suppliment all of the really good answers above, which collectively address the economic and industrial aspects of massive population decrease, and instead look at the psycological and sociological implications, because I think those put a much lower ceiling on the number of people you can lose and preserve civilization.

Simply put: when massive catastrophes happen, people panic. And when enough people panic, civilization falls apart.

(Warning: callous discussion of many people dying. Sorry, can't discuss this topic otherwise. Clearly the deaths of folks outlined here are of way more than academic interest to them and their loved ones.)

The kind of catastrophe you're talking about -- the loss of, say, 25% or 50% or 80% of the world population -- would be unprecendented in human history. Based on human history, lesser population losses have resulted in major or even total social breakdown. For example:

• The Syrian Civil War has "only" killed 2.3% of the original Syrian population. Yet the majority of Syrians are now refugees, and large parts of the country are lawless and without an industrial base.
• European diseases killed an estimated 25% of the native population of North America, which caused some tribes/nations to cease to exist (but, interestingly, not others), such as the cities of the Illinois tribes.
• Relatively limited (no solid figures, but not more than 10%) deaths due to famine in the Mayan civilization helped trigger a cascading collapse which wiped it out.

Now, you can find plenty of examples of civilizations that survived large population losses. Europe and the Black Death is probably the paramount example; despite losses of around 1/3 of the population in many areas, European civilization and technology didn't regress by more than a century -- despite profound and far-reaching effects on economics, war, and religion.

So losing a large percent of your population isn't a guarantee of panic, destruction, collapse, and regression to the stone age. But sometimes it is. The key questions tend to be:

1. How stable was your civilization in the first place?
2. What is the nature of the population reduction, and how do most people feel about, and react to, it?
3. Are there external or internal forces that would compound the effects of large-scale population loss?

If you're basing this on modern world civilization, there's ample cause to be pessimistic about (1) and (3), but I'll leave that up to you.

However, I cannot find a single example of a civilization that lost more than 60% of its population, over a generation or less, and continued to exist. So if you want to have an absolute numerical ceiling, I'd say that you need to retain at least 50% of your population.

Answer 9

Quick edited note here: My answer, as I understood the question, is aimed at what is required to maintain modern technological levels. I answered this way, because in my understanding, that is the required foundation for modern life and in my view civilization is too subjective and broad to be answered. For that, I could say US or Europe or Japan, with automatized colonies around the globe for rare resources.

What we DON'T need, to only maintain modern technology

Let's start the answer in reverse. Who we can dismiss? Who we don't need to maintain modern technological levels. A good part of humanity if I says so. No offense for anyone. I'm part of the "dismissed".

• Employees of huge companies who provide service to the masses? They can be significantly down-scaled. If functionality, the maintaining of technology instead of profit the purpose, only their engineering team. You can scrap their HR, sales, and so on departments.
• Most services and its workers too - you don't need 20 kind of chips and cars and so on, to maintain modern technology levels.
• Any kind of Media related profession - SCRAP!
• Governmental bodies - Significantly downscaled, based on the situation and type of government.
• Homeless, unemployed, and so on... they don't contribute to modern technology, so they too can be dismissed.

Other factors: Human intellect, productivity, automation

• Automation - we are heading towards a heavily automated world, most mundane task can be or will be replaced by robots.
• Human intellect - Automation means, you can dismiss almost all the heavy-lifters. You will however need highly educated and intelligent people to operate and maintain the machinery.
• Productivity - much depends on the disposition of the people. Are they lazy? Hard-workers? Do you need police to manage big masses of them and so on. This too will affect the final number.

The final factor to account in this scenario, do all human capable of reaching the same intellectual level? Do they born equal in that regard? Or for example every hundredth people has high enough intellect? I can't be trained to be a Quantum Physicist. People's view vary in this regard, so I will leave this as a simple multiplier.

---

They are pointers for a reasonably close estimate. My very rough guess, a few hundred thousand or most a million can maintain an inherited and hastily retrofitted modern technological world.

NOTE: In my view, they have to be highly specialized in various technological fields, and preferable the cream of the crop in intellectual ability, among other things. Mostly, because this will lower the necessary number. Not ALL of them have to be like that, but a significant percentage. Depending on the level of automation and mechanization, this percentage may vary. (Look at Factorio) Also, my assumption is that a hard-laborer can't be retrained (in reasonable time and capacity, views may vary) to be a rocket scientist, nuclear reactor operator and so on...

Answer 10

This answer might sound mean, but most people can be killed off without affecting the industries, if you choose the right people. You only need the people directly employed in the industries, the people supporting the cities and governments where those people live, the people providing supplies to those. Currently that's basically the West, factory places in Asia and some others scattered around the world. Probably around a billion people needed to keep everything running as it is, but could be less.

Although it depends where and in how many places you want to see modern civilisation. In all the places it exists around the world? In a select few nations? Or is one city enough?

Answer 11

Almost all of them

There are two problems here, one is the population required to maintain the modern world, but the other is the population required to have reason to maintain the modern world. The world isn't a museum, much of the technology and infrastructure exists only because it's needed to maintain the global population and respond to its demands.

As already been mentioned, you need population density, as you're already aware you need population numbers, but you also need population spread.

• If everyone lives in the same city, is there any reason to maintain GPS?
• If everyone lives in the same country, is there any reason to maintain global shipping and logistics?
• For a mere 2 million people is there any reason maintain vast high production car manufacturing plants?
• Large numbers of people have to have reason to travel long distances to maintain the aircraft industry and related technologies including air traffic control.
• The need for high capacity mass transit systems is tied to the need for large numbers of people to travel.
• The need for vast orange juice plants is tied to the global distribution of the product.

As soon as you start cutting into the global population, you start cutting into the requirement to maintain large swathes of industry and their related technologies.

Sooner or later even the slush machine in your local cornershop is declared a redundant technology, there just isn't enough demand to maintain it over something smaller and simpler, or doing the job by hand.

Answer 12

Less than 500 million

The Georgia Guidestones were created by "a small group of loyal Americans" "which had been planning the guidestones for 20 years" and was able to pay "several times higher than any project the [construction] company had taken" earlier. If this information is true, then we can assume they had hired actual economists to give an accurate answer.

At the time of construction (1979) the world population was 4+ billion, so they claim that less than 12.5% of the population would be needed. This obviously assumes that the population number was based on the requirements of a modern (1979 era) level of technology.

Answer 13

I feel that we need a collection of people skilled to connect technology and create a formidable ecosystem. For eg, To create a smartphone we need people trained and skilled in a lot of processes. So, we'll need a collection of differently skilled people in a singular place where they quarantine themselves which would involve making a constitutional structure to envision the armed forces, leaders to create law, economists, technicians and hunters and gathers and this ecosystem needs to adapt to changes in an exponentially fast manner and hence would require brainstorming and control and a very constituted system. We need to set up an economy parallel to every other process. This society and the ecosystem around would need to mature in a specifically designed manner as to support the sustainable development. I further deduce that we would need to be aware of the world atmosphere and habitants and geographical abnormalities and a satellite system to track target systems. Hence, it's only possible if we presently device such a system and follow log book instructions with certain situational innovations in case of such events. Our best bet would be the US military budget allocate for such situations where they run simulations and devise plans to counter them.

Answer 14

It is all on a sliding scale. Most of the basic knowledge to keep civilization alive will remain in the libraries of most major universities. If there are 10 plants than make computer chips and they are ALL gone it will take a while to build a new factory. If all the manufacturing plants are destroyed it will take a while to rebuild them. If all the refineries are gone they will need to be rebuilt. If shipping is mostly gone, getting the needed materials will be hard for many complex products.

I would expect the population to be in pockets where the survivors could defend themselves. Each pocket will rebuild as much as they can with what physically survived and the available expertise minus time and effort to stay alive. As the small pockets begin to interact the knowledge, expertise, products, etc. will become available to each other. I do not think population is as important as resources.

The population of Tonga is 108,000. They do not produce much and most modern technology would disappear if they were the only ones left. If there was a pocket of 20,000 professors and advanced students holed up at a major university they could do well for themselves.

One other factor. The knowledge is already out there. I have never built an airplane but know you need to deal with yaw, pitch, and roll and the shape of the wing adds lift and Jet engine thrust can exceed propeller thrust. I do not know how to make penicillin but know it was derived from green mold. I do not know how to make fertilizer but know that plants need nitrigen and tilled ground speeds root growth. I never built an engine but know basically how spark plugs ignite compressed gas and air to produce the power. Finding the specifics while knowing the basic concepts behind the inventions would make the inventions occure significantly faster than discovering them for the first time. What is missing would be fairly quickly recovered depending on the survival efforts of each pocket.