Can we Evaluate Multi-dimensional Problems Accurately?

22 August 2022

I was reading this thread by Tim Urban the other day where it showed a world not overpopulated but quite the contrary. An almost empty world. How can this be? How do the numbers add up?

In Tim’s thread, we see the following counter-intuitive image. If we take Manhattan’s population density and transfer it to New Zealand we could fit all of the world’s population in it.

World Population could fit in New Zealand

And another one where a specific area outnumbers the rest of the world in terms of population.

Lots of people in China and India

The math does work and it’s pretty easy to verify for yourself. Although this is more of a thought experiment and not something to be taken literally — some places in New Zealand are not livable plus there are rivers, lakes, wildlife, etc. But it does illustrate a point where it gives you a sense of proportionality.

Manhattan has an average population density of 28,154/km2 making it the city with the highest one in the US. Land area is around 59,13km2 which means 59,13 times 28,154 you get about the total population of the city(1,694,251). New Zealand is about 268,021km2 big so if you take that, times 28,154 you get pretty much the 7,5 billion — close enough(!).

Manhattan is probably not the best example of an ideal and sustainable place to live though so let’s open things up a bit.

Australia is a big place with around 90% of the land being classed as uninhabitable, so that leaves us with about 10% of habitable land which is approximately 768,685km2. This translates into fitting roughly 2.8 times the entire earth’s population, 2.8 times less densely than Manhattan.

Following this kind of logic seems that it’s more of a distribution problem than a number of people problem.

Just like being skinny-fat. You don’t need to lose weight per se, is just that the weight is not evenly distributed across your body. You look fat in a particular area due to a high percentage of body fat while skinny in another due to a low amount of muscle.

But many go beyond this simplistic description above and rightly so. The main concern is resources, like food, water, energy, etc. These are what make the problem multi-dimensional. So what about them?

The overpopulation issue has been drilled down our heads so much that we reactively think it’s a settled one, making us impulsively view images and analogies like the above to be phony, child-like, or a product of junk science. The main argument on the table is that more people means more resources consumed thus more emissions resulting in more pollution. So the solution is to cut back on people to a more manageable amount.

(One possible reason why such narratives seem compelling, is due to how we romanticize nature and envision living in the good old days where we gathered around fires, living a simpler life, without stress and in harmony with nature. It is not uncommon for the same group of people that hold such views to demonize growth, wealth, capitalism, and the free market perceiving them as inherently evil and as an interruption to the simple life, dismissing humanity’s past hardships or simply being unaware of them)

Many go beyond this simplistic view here as well because it smells like flawed thinking too. Why would you want to minimize the number of people in particular, as a variable and not tackle emissions or resources instead? What makes you think this will have the most effect?

It turns out that people always worried about population growth with too many people, in particular, being the problem. From Tertullian, a church father who lived around the year 200AD when population was 30 times less than today to Plato and Aristotle who lived 600 years before Tertullian when population was 60 times less, all of them were concerned for pretty much the same reason. Too many people.

Relatively closer to our time and probably one of the most familiar figures on the subject, is Thomas Malthus who raised concerns about over-population in his essay An Essay on the Principle of Population in 1798. Malthus calculated that human populations tend to grow exponentially, while the ability of humans to feed each other tends to grow more linearly so, at some point, one will outstrip the other.

On a similar note, Paul Erlich a biology professor from Stanford University wrote about overpopulation too. In his quite influential book The Population Bomb(1960) he quoted:

[…]In the 1970s hundreds of millions of people will starve to death in spite of any crash programs embarked upon now. At this late date, nothing prevents substantial increase in the world death rate.

Why did none of the above thinkers put their minds working towards a solution other than the reduction of humans? This is a process that takes many generations to observe any kind of meaningful change and it’s hard to think of a way to actualize it effectively without the use of some degree of force(see China’s one-child policy).

Erlich even suggested adding poison to tap water to make women infertile and some of his predecessors viewed famines and pandemics as a possible solution. Such extreme “solutions” are derived mostly from the certainty that one is on the right side of history. So right that the end justifies the means.


So food it is?

Most used food as their main metric to calculate the limit of population growth since it was the main factor of prosperity at the time with both Malthus and Erlich predicting mass starvations with a high degree of certainty.

This turned out to never happen in the real world (Malthus couldn’t predict the Industrial Revolution and Erlich couldn’t foresee innovations in agriculture). Quite the contrary. Even with the rise of population in the 20th century, world hunger declined and continues to do so.

It is estimated that we produce 1.5 the amount of food we need, meaning we have the capacity to feed everyone — in theory. The problem is the distribution of resources and not a lack of them, and although hunger still exists in parts of the world, it is caused by a lack of proper infrastructure and economic growth.

Around 200 years ago, more than 90% of people worked in agriculture in order to feed the population so trying to produce more food required more people as labor. Within the 20th century though, with the help of synthetic fertilizer and other modern technologies, agricultural output tripled on only 30% more land over the same period.

How about water then?

Water follows a specific recycling lifecycle and its quantity has stayed pretty much the same. It is said that around 1% of all water is drinkable and much of that 1% is trapped in glaciers so the amount that is accessible by humans is even less. This sounds scary but it shouldn’t. Similarly to food, water availability is heavily affected by distribution bottlenecks due to poor infrastructure and not because we are running out of it. For example, in South Africa, billions of liters of clean water are lost annually due to neglected infrastructure and leaky pipes. In America, things are no better with household leaks reaching close to 1 trillion gallons of water per year.

How about energy and emissions then?

All we hear — being culturally acceptable and even applauded nowadays — is that we just selfishly consume resources and give nothing back to the environment.

We are only thought of as consumers but why not as creators as well? We live on this planet just like animals do and we need to create resources in order to survive, initially, and later thrive. You can’t stop that.

When we refine a process and improve the efficiency of anything that consumes energy we also minimize its emissions, use less land, and generally do more with less. Two centuries ago, a major part of deforestation was due to wood being the primary source of energy only to move towards fossil fuels later on which was an improvement at the time. And with nuclear being a future option things might improve a lot more, energy-wise and land-wise, where it will be orders of magnitude clearer.

This is a somewhat familiar pattern that nations undergo throughout their development where they initially use more land only for this to flip around and start giving land back to nature and move towards cleaner sources of energy. Some studies even suggest that the world is getting actually greener with an increase in the earth’s terrestrial carbon sink.

But does the increase in knowledge outperform the growth of population?

Difficult to calculate but roughly, more people implies more knowledge. If we didn’t have people we wouldn’t have resources either and without the necessary knowledge humans create, we couldn’t identify what could be used as a resource or not. Rocks are just rocks and the air is just air up to the point where we create knowledge and build houses and windmills. So there is not an actual limit to resources but rather a limit to what knowledge we have acquired thus far.

Shouldn’t we be afraid of running out? No. The lesson here is that the limiting factor is not resources — but knowledge. Knowledge of how to take the useless and turn it into the useful. To create solutions from the matter around us — to literally transform the physical reality in which we find ourselves into a home. — Brett Hall

On average with higher economic growth and improved living standards, population density tends to decline and people tend to care more about the environment and where they live. If you live in the Western world, for example, you are likely recycling — or at least have it as an option — whereas in developing countries this is not a priority for obvious reasons.

If we think of it in this way aren’t humans the ultimate resource from earth’s eyes? We create resources, increase the supply of other resources, discover alternatives, improve efficiency, and have the capacity to give land back to nature.

The opposite problem

At the current moment in the developed world, fertility rates are falling, so it wouldn’t make sense to alarm Americans or Europeans to have fewer children. They already do so. Due to this reality, many predict that the population with cap out at around 11 billion at around 2100 only to start declining. But it’s hard to tell for sure.

With this in mind, many even raise concerns about underpopulation with legitimate arguments. One of them is a phenomenon called japanification which refers to what Japan has experienced in the past decades — economic stagnation due to a disproportionate amount of young and old population.

Doom scenarios are prevalent here too, with a future with far less innovation, less economic growth, fewer minds trying to solve humanity’s problems, and labor shortages.

Why proponents of the underpopulation threat don’t use the same arguments one could use against proponents of the population bomb? Couldn’t we, with technology and automation, fill the gap for labor shortages? Couldn’t we leverage automation to sustain an aging population that can’t work? It seems challenging for intellectuals of today — and from any time really — not to fall into for doom and gloom scenarios that Malthus and Erlich fell for. To head towards the same direction in the optimist/pessimist spectrum but with a different thesis.

Studies are likely to change and be proven wrong with others being made with better methodologies and more accurate results. All the articles, stats, and studies I linked so far might in 80 years’ time seem just like how we view Malthus’s essays today. Understandably a good effort for its time but flawed and incomplete by today’s standards.

So what to believe? Both sides seem to have relatively good arguments on the surface and are able to back them up. Obviously, the number of people we can sustain should have some theoretically upper limit but having an exact number is hard — or impossible — to pinpoint. Turns out that all complex problems follow a similar trajectory on a macro level that doesn’t have to do exclusively with data and is certainly not something unique to the over/under population debate.


Viewing the Problem

I’m quite suspicious of 1D(one-dimensional) interpretations of problems and solution proposals, where we take only one variable and say if it’s up it’s bad so we have a problem, and if it’s down it’s good so problem solved. For example, if we assume that fewer people means good since we consume fewer resources, then wouldn’t it be even better if no humans existed at all thus all resources are available? If you follow any 1D solution for a problem and extend it to the extreme you get silly answers like the above. In order to avoid sounding fringe, you must have something to pull you back from the edges every time you are tempted or it’s convenient to do so. That is why a 2D and above analysis is necessary for nuanced problem solving, meaning we understand that every problem is multifaceted with many moving variables. In other words, it’s complicated.

We seem to have a weak strategy when it comes to action but an overreacting strategy when it comes to alarming. It seems that problems persist and no improvement is made. But this is not the case. This can be understood quite well by observing the problem sellers and problem solvers dynamic. Sellers tend to take a bunch of problems, mash them together and construct a huge one out of them ready to be served. Ironically the solutions they often propose are 1D simplistic ones. For example, tax the rich and problem solved, go all organic and problem solved, regulate crypto to death and problem solved. Solvers on the other hand, because they understand the complexity of the problem, realize the need to break it down into many, more manageable, pieces. They don’t promise simplistic solutions but do work on one specific chunk of the problem without losing sight of the dimensionality of the matter embracing trade-offs. Trade-offs are inevitable when it comes to choosing what to work on, and what chunk of the problem to tackle while consciously ignoring everything else. Without trade-offs, no progress is being made. You just have to choose the right ones at the time. After all, a mediocre win is better than a glorious failure.

As humans, we are not wired to understand counter-intuitive facts because they are not needed for survival in the wilderness. We are not good at all at visualizing the massiveness of objects, making any acute comparison between them, and having a sense of how they evolve through deep time without artificial assistance. We see a crowded city and say gee there are indeed many of us. We see a picture of a train that is covered with people, hanging out of its windows saying wow the same thing will happen to us.

Without the help of an artificial constraint that gives us perspective like statistical results or with an artificial augmentation of our senses like telescopes and the like, we are doomed to trick ourselves with such fallacies.

One could argue that if you get your facts straight you are covered. You might indeed be. But things are more complicated than that. Having an abundance of information and a plethora of studies adds one significant downside to the equation, which is an important dynamic in data analysis.

That is the Curse of Dimensionality which states: The more dimensions a problem or a set of data has the more sparse their connections become due to the increase of the graph’s volume. So obtaining reliable results becomes exponentially more difficult and any piece of data becomes less insightful, even after adding only one extra parameter.

What this means is that by having endless data at our disposal, any conclusion can be drawn and any narrative can be constructed. It is not to say that it’s futile but to simply be aware of it and turn down a notch our certainty levels.

I can’t debunk anything here — nor I try to do so — but I do wonder if we really have a solid sense of how to measure things, read data effectively, and agree on any meaningful action points. Plus we seem to be overconfident about how the future will play out which was always a long shot. Some problems are just too far in the future to make any kind of confident prediction.

We are trapped in multi-dimensional problems with our only way out is to harness the engine that brought us progress. Probably the only proven way that has ever worked. Human ingenuity and innovation, with optimism as fuel.

On what principle is it that with nothing but improvement behind us, we are to expect nothing but deterioration before us? — Thomas Babington Macaulay



Subscribe to Newsletter

Subscribe to the newsletter to get my latest content by email.