Disqus won’t let me comment, so I post my comment on this post hereI am honestly impressed that you didn’t mention lead. So I will. The single parent household peak came later than the murder peak — as one would expect. Single parent households last (until another marriage & good luck with that Ms single mom). You graph a stock (OK 3 stocks) each correspond to flows — births to single moms and separations of couples. Teenage births peak when the girls (and 18 and 19 year old women) in queston had maximum lead exposure as babies and toddlers. The boys (and men) aren’t always even identified on birth certificates, but ages of mothers and fathers are usually similar. Hasty marriages (including of couples who are racing to be spouses before they are parents) are less likely to last.
i am surprized to be more piombophobic than you, but I think it was lead. Certainly the evidence on lead and teenage pregnancy looks similar to the evidence on lead and crime.
In any case, to understand causes, it is almost always better to look at flows than stocks