People aren't "choosing annihilation". There is no weekly meeting. One Japanese person having 10 kids does not change the headline. Birth rates are a textbook-perfect example of a tragedy of the commons coordination problem scenario. Furthermore, if we believe that social or institutional trust is an input to birth rates, the implication is that Sweden is a low trust civilization and South Sudan is high trust. It just doesn't pass the smell test.

People have fewer kids when the medium-term alternatives are better. This is why desperately poor people have tons of kids and why fabulously rich societies like Japan have fewer. An extremely poor person does not lose opportunity by having children. A rich person does. We fix it by inverting the cost and benefit. If being single and childless has fewer medium term rewards than being a parent, people will become parents.

And the "medium term" is I think an illuminating point to emphasize. Humans, I have noticed, tend to operate on 2-3 year time horizons. If humans operated on a 5 minute time horizon we'd have more kids because unprotected sex is fun. If we operated on a 50 year time horizons we'd have more kids because being 75 years old with no surviving family is for many a terribly lonely thought. I think it's also why so much of the birth rate conversation focuses on childcare, diaper changes, and sleep loss, in spite of the fact that the years involving those challenges are an extremely small portion of the life-long parenting experience.