Last week, I reported the surprising realization that official population projections from the United Nations adhere to a notion of future fertility that appears to be immediately at odds with present real trends. The recent rapid decline in population growth—even pre-COVID—suggests that a population peak prior to 2050 is not outlandish, provided that current drivers continue to apply. Recent declines in fertility rates, together with a flattening age distribution of young folks, combine to set the stage for population peak and decline.
In the previous post, I performed two embarrassingly crude projections of recent trends (simple curve fits) to demonstrate that a population peak as soon as 2042 or even 2033 should not be ruled out, and in fact seems to be where we’re heading if present trends continue.
I mentioned that I was working on tooling up a more sophisticated model to do some exploration of my own. The goal was to track the nuances of actual age distributions across the world, together with alternative ideations of fertility evolution (greater weight on what is actually happening lately), and allowance for non-monotonic evolution of medical care and life expectancy going forward. It was a daunting task, but I was consumed with curiosity and powered through the exercise over a few intense days.
In this post, I will give an overview of what goes into my demographic projection model, why I believe it works well enough to be useful, and what top-level questions we can explore using it.
Views: 10468