More people generally is better.
I do some fantasy world-building in the little free time I have. The setting is diesel-punk/atom-punk, so 1950s era. A little bit of disco-punk mixed in too. Essentially, modern post-war USA.
So, when I try to model a city, or a country in this era, you gotta go after the data. Because we actually have it for this setting.
Medieval data is just horrible in comparison. The high fantasy settings will get a little bit of data on city dynamics and some on 'nation' level data and the like. But most of the modeling is starting from pretty much guessing with wide errorbars.
But in the atom-punk setting I have, there is just mountains of data from all over the world. It makes modeling really easy and you can get really fine-grained about it (much to the detriment of story-telling, I get lost in those tables and charts, happily so).
One thing that really stands out in trying to model this data is that population is literally everything.
How grimy a city is, how tall the skyscrapers are, the walking speed of people in the city, the talking speed of bank tellers, the number of patents per capita, the monthly charge for sewer access, tonnage of warships, the length of subway track and it's heights, etc. Pretty much every metric will correlate best with the population of the area. It's not just cities, it's states and countries and smaller towns too. Really anything about the setting and world-building is best modeled by looking at the population first and then working from there.
To be clear here, not the population density. The only metric I could find in my notes that better correlated with population-density was the number of subway stops on a line. I'm sure there are other examples, too.
But, in the years that I've been doing this, the number one thing that seems to pop out to me is: Raw population is everything.
Now, I'm mostly interested in themes and settings and the like. But most of the data is coming from the motivation of trying to increase the quality of life of people. So I get to stare at that a lot and make graphs.
And I can say from my eyeballing of all this data over the last few years: More people = better.
Most of the metrics we use to quantify if things are 'good' or 'bad' will correlate with more people = better. It's not, by any means, a 1-1 thing. This is human population stuff, If you see a r2 value of, like, 0.6, that's pretty darn good. There is a lot of noise here and the stats that are taken have just huge issues with them.
I'm not going to be able to really prove this, I'm just mucking around with a setting and world-building. But from the years that I have been staring at my poorly organized notes and the various governmental and NGO data tables, the underlying message is that things get better with more people around.