Population growth or decline will have little impact on climate change

It'll be too slow and too late for the timeframes that we need to decarbonise.

Limiting or reducing the number of people in the world is often touted as a key climate solution. Sometimes it’s held up as the solution.

Will population growth (or shrinkage) have much of an impact on climate change at this point?¹

In my book Not the End of the World, I argued no. The main reason being that large demographic changes happen over longer timescales than we have to get emissions down. If we’re serious about getting anywhere close to net-zero by 2050, 2060 or 2070, then each of our per capita emissions will need to be pretty low within just a few decades.

It’s extremely difficult to dramatically shift demographic patterns within that timeframe without drastic (and inhumane) measures. Basically, killing large numbers of people [sorry for the bluntness]. Even a strict one-child policy everywhere (also not humane) wouldn’t change things a lot in that timeframe. For fun explainers on this, see some of the late Hans Rosling’s videos about population growth.

The global fertility rate — the average number of children per woman — has more than halved over the last half-century and is now just over two children. This is close to the “replacement rate” which is the level at which the population would stop growing.

Source: UN World Population Prospects (via Our World in Data)

As a result, the UN expects the global population to peak in the 2080s. Many demographers think that the peak could come much sooner since projections have consistently underestimated the pace at which fertility rates have declined. Regardless, towards the end of the century, the world population is expected to start shrinking.

This has led some to believe that the decline will offer us a much more optimistic climate future. But this conclusion is also wrong, for the same reasons I argued before: that changes in population are incompatible with the pace at which we need to decarbonise.

I saw that this point here was fleshed out in much more detail in a new paper by Mark Budolfson and colleagues: “Is Less Really More? Comparing the Climate and Productivity Impacts of a Shrinking Population”. In it, the researchers model the impact of different scenarios of population growth and the impact on climate change.

This was a follow-up to a previous paper of the same theme: “Population Decline: Too Small and Too Slow to Influence Climate Change”. They are pretty blunt with the conclusions:

“Low fertility is a false solution to climate change: the population impacts are too small and too slow.”

Let’s take a look at the data and what it means.

---

Billions more people might only increase global temperatures by a fraction of a degree

The authors model the climate impacts of two population scenarios:

• Depopulation: Perhaps surprising to some, this is the path we’re currently on. As we saw earlier, the UN Medium projections suggest a global population peak in the 2080s. In fact, we’re probably currently on an even steeper depopulation path because fertility rates are falling faster than expected, and the UN keeps moving the date of the peak further forward.

• Stabilisation: In this scenario, the global fertility rate does not fall below replacement levels, so the world population grows through to the end of the 22nd century.

You can see these two scenarios in the chart below. By 2100, there’s around a 2 billion people difference. This expands to a 6 billion difference by 2200.

Now, let’s take these two population scenarios and calculate their global emissions and the impact that would have on global temperatures.

Of course, to do this, we need to make assumptions about how good we are at decarbonising. We need to make some assumptions about what per capita emissions will look like over time. In their baseline scenario, the authors take a standard trajectory from where we are now, and assume that per capita emissions fall by 1.8% per year.

This is quite a pessimistic scenario because it assumes we make little improvement to accelerate decarbonisation rates. But let’s run with it. Soon, we’ll see what impact more optimistic or pessimistic decarbonisation rates would have.

In the chart, you can see the impact on global emissions (in grey) and global temperature change (in red). The solid line is the depopulation scenario (which we’re currently on). And the dashed line is the population scenario where we have a few billion more people by 2100, and six billion more by 2200.

Global emissions would peak somewhere in the 2040s, then slowly decline through the early 22nd century. What’s crucial is that the difference in global temperatures between the two scenarios is tiny: billions more people increase temperatures in 2200 by just 0.1°C.

Now, these results seem unintuitive and hard to believe. But by doing a bit of a sense-check, it becomes more obvious why it’s true.

Let’s think about it first in terms of per capita emissions. In the emissions curves above, global average per capita CO₂ emissions would still be around 4 to 4.5 tonnes per person in 2050. When they’re still at this “high emissions” point, the difference in the global population is still very small. That’s because it takes a while for changes in fertility rates to filter through to the global population. There are probably around 250 million extra people in the “high” population scenario in 2050. Together, their additional emissions would be around 1 billion tonnes per year.

That sounds like a lot, but it’s less than 2% of annual emissions, and a tiny amount of the total cumulative carbon emissions (which is really what matters for warming) that we’ve accumulated in the atmosphere over centuries. We’ve already added around 1.8 trillion tonnes of CO₂. Let’s keep things simple and say we have another 25 years — to 2050 — of emitting around 40 billion tonnes per year. That would mean adding another trillion tonnes: so, 2.8 trillion tonnes in total, by 2050.

These additional emissions from a larger population — 250 million tonnes in 2050 — would be just 0.04% of cumulative emissions. Even over an entire decade, it would be just 0.4%.

Of course, the difference in population size starts to grow throughout the second half of the 21st century and into the 22nd. But, what’s key is that during this time, per capita emissions start to become pretty small (even in this relatively pessimistic decarbonisation scenario).

So, just at the point where population size could make a difference, emissions become low enough to mean that they don’t. By 2100, the difference between the two scenarios is a few billion people (roughly 10 billion versus 12 billion). But, per capita emissions will probably be down to around 1 tonne per person. So the difference in total emissions is still in the range of around 1 to 2 billion tonnes per year. Again, this sounds big, but it is small compared to the over 3 trillion tonnes that will have been emitted by 2100.

By the middle of the 22nd century when the differences in population become really substantial, per capita emissions are close to zero, and therefore it has little impact on temperatures.

The final result is that by 2200, the difference in global temperatures between the scenarios is just 0.1°C. A final sense-check on this. Global temperatures are tightly correlated with cumulative emissions. Emitting around 200 billion tonnes of CO₂ is associated with a temperature increase of 0.1°C.²

Given that the difference in emissions between the scenarios in most years was around 1 to 2 billion tonnes (dropping much lower through the 22nd century), it seems very plausible that the total extra emissions over this period are around 200 billion tonnes (or slightly less).

---

The slower we decarbonise, the more population matters

The reason population matters less than many imagine is that by the time there is a large difference in global population, we’ve already decarbonised quite a lot. That means per capita emissions are low.

But, if we’re incredibly bad at decarbonising, then population does make more of a difference. Now, the scenario we just looked at was already pretty pessimistic in our progress on shifting away from fossil fuels. Let’s take an even more pessimistic scenario (which you can see in the chart below). Emissions continue to rise until the 2070s or 2080s, then very slowly decline. To be clear: this is highly implausible, but interesting to run the model on anyway.

By 2200, global temperature rise would be around 6°C (very scary indeed and a scenario where we might question whether a population of over 12 billion people was realistic). The difference in warming between the two scenarios was around 0.4°C. That’s a much bigger difference than the previous scenario we looked at, but I’d still argue that population growth was not the main — or even a big — driver of this change. In a 6.4°C warmer world, it would be our failure to decarbonise at any reasonable pace that is to blame, not the 0.4°C that came from having children.

In contrast, in a relatively optimistic scenario where we decarbonise quickly, the impact of population is tiny. That’s because per capita emissions in the second half of the 21st century — when there starts to be a meaningful difference in population numbers — are incredibly low. So the additional emissions from billions more people is insignificant.

In the real world, I’d expect that our decarbonisation pathway is somewhere between the main scenario and this rapid decarbonization one. That means the impact of higher population growth would probably be non-zero, but less than 0.1°C.

---

Population policy is not an effective climate policy

The key point here is that changes in population are not going to have a particularly meaningful impact on global temperatures. If we want to tackle climate change we need to decarbonise our economies; that’s really how we get out of this.

This point is true in both directions: people arguing that limiting population growth is key to tackling climate change and those expecting that a shrinking population will be a big boost for our climate goals. Both overstate the impact of population.

Now, some might argue that the scenarios above are not dramatic enough for what they’re proposing: that what they’re recommending is not for our current path — the depopulation scenario — to be the “low” population scenario, but for us to cut fertility rates and shrink the population much faster. But let me reiterate a few points again:

1. Fertility rates in many countries are already very low (and they’re falling rapidly in others), so you’d need pretty extreme (and politically unpopular) interventions to reduce them by a lot more.

2. The impact of these changes will be too slow to have a meaningful climate impact. It will take many decades for this to affect the size of the global population, which is too late given our climate goals of reducing emissions within decades.

As I’ve written about many times before, we should do many of the things that have been associated with declining fertility rates: reduce child mortality and poverty; give girls the opportunity to go to school, get an education, and employment; improve access to contraceptives; and empower women to make their own family planning decisions. These are all excellent things to do; I just don’t think climate change is the main reason to do so.³

As I wrote previously:

“My point is not that these things are not important. They are. It’s just that they are important enough in their own right. Girls should go to school because they deserve the right to go to school. Not so that they will have fewer children that will emit small amounts of CO₂ later in life.”

---

---

Some of the co-authors of this recent study — demographers Dean Spears and Michael Geruso — just published an excellent book called “After the Spike” which explores all of these population issues in more detail. As is quite rare for a book looking at birth choices and reproductive decisions, it’s compassionate rather than preachy or pushy.

It’s a contentious topic, so I’m sure readers will find lots to debate and argue over, but I think it’s a very good anchor to inform and open up these conversations:

https://www.penguin.co.uk/books/462440/after-the-spike-by-geruso-dean-spears-and-michael/9781847928351

1. I say “at this point” because there is no point going over implausible population dynamics from the past. Often people will say things like “well we wouldn’t have this problem if there were only one billion of us”.

But the reality is that there are eight billion of us (and most of us would not be alive if population growth stopped at one billion). Debating implausible scenarios of the past is not that helpful in moving us forward.

2. This is based on the Transient Climate Response to Cumulative CO₂ Emissions (TCRE) model, which is used in the IPCC AR6. It estimates that there is around ~0.45–0.6°C of warming per 1,000 GtCO₂. That means roughly 0.1°C for 200 GtCO₂

Matthews, H. D., Gillett, N. P., Stott, P. A., & Zickfeld, K. (2009). The proportionality of global warming to cumulative carbon emissions. Nature, 459(7248), 829-832.

3. The focus of this article is the impacts of population on climate change.

People will rightly point out that our environmental impacts go far beyond climate, and population growth continues to exacerbate those. In my book, I covered lots of other problems, including biodiversity loss, food systems, deforestation, overfishing, and I think it's possible to reduce our impacts there too, without reducing the number of people.

There's too much to go into here, but interventions such as dietary change would have a much, much bigger impact than small changes to fertility rates.