Increased population good or bad for average welfare?
Is increasing the number of people good or bad for average welfare? It decreases the resources available per person, but increases the number of people generating and sharing new ideas.
My current best guess is that increasing the population uniformly by 10% is bad for average welfare in the short term but becomes positive over the longer term (maybe 60 years?):
In the short term, having more people means you have less stuff per person---there will be more people sharing each car factory or office building or farm. But most of the stuff that matters, at least weighted by economic value, is built by humans. Even land is mostly valuable because it’s developed or in cities. Over the longer term, a larger population will mostly make more stuff (including building new cities).
Many of the benefits of a larger population take the form of faster technological progress and more know-how, which disproportionately benefit future generations. The fact that capitalism is able to incentivizes technological progress at all, using only the returns over the next few decades before innovations diffuse, feels like basically a lucky coincidence.
The changeover point is complicated and contingent, I’m just guessing “a few decades after the new people become adults” based on the time required to build new stuff.
In the very short term I think the welfare effects are dominated by whether people enjoy having bigger families and younger societies.
It's not obvious whether we should care about average welfare, especially over multiple generations. I'm thinking about it because it would make the change more robustly positive, though for many ethical views the benefits (or costs) of “more people get to live” would likely dominate.
Thinking about this question quantitatively changed my views. I had cached a vague cosmopolitan intuition that more people would quickly make average welfare higher, without having ever really checked that intuition.
More people feels intuitively beneficial to me because my day-to-day experience is dominated by the benefits of a bigger world, e.g. having more good software, better TV, and more people who share any particular interest. But about 30-40% of GDP flows to capital, suggesting under plausible assumptions that increasing the population 10% without increasing the capital stock would make each of us about 3-4% poorer, and it seems unlikely that the short-term economies of scale from +10% of people would be +3-4% of GDP. That said, there are tons of ways that 3-4% cost estimate could be wrong.
The main reasons I can see long-term effects being worse than short-term effects are large harms from climate change. I think the additional per capita cost of mitigating climate change from a bigger population would be small relative to the other costs and benefits (e.g. I think it would cost less than 1% of GDP to reduce per-capita carbon emissions by 10%). That said, we may just decide not to do it and the additional long-term harms from warming over the next 100-200 years could easily be 3%+ of GDP from a 10% increase in population. This becomes particularly plausible if technological progress is a lot slower than I expect over the next 100-200 years.