Here is some interesting research, summarized in this month's NBER Digest:
In the long run, low rates of fertility are associated with diminished economic growth, according to a new study by NBER Research Associate David Bloom and his co-authors David Canning, Günther Fink, and Jocelyn Finlay. In The Cost of Low Fertility in Europe (NBER Working Paper No. 14820), they observe that in the short term, low fertility rates raise per capita income by lowering families’ costs of child-rearing and boosting the share of working-age people. But as that working-age population moves into retirement, the number of workers who replace them will shrink. So, whatever short-term boon European nations may have gained from low youth dependency will be overwhelmed eventually by the economic burdens of old-age dependency.<!--break-->
The summary describes a set of calculations to maximize the working-age share of the population. That is an interesting way to think about the issue -- the per-capita cost to the government of pay-as-you-go social insurance and other age-related transfers (like education) should be minimized that way. Here's an example:
In France, for example, where life expectancy is 80, the fertility rate that would maximize the working-age share of France’s population would be 2.1 if young people started working at age 20 and retired at age 60. With retirement at age 55, the working-age share-maximizing fertility rate would have to rise to 3.1. With retirement delayed until 70, that rate would drop to two.
Read the whole thing.