Scott Alexander, curated
← Back to curation

Literally Inconceivable: Contraceptives And Abortion Rates

Quality
70
Strong
Claude Shift
48
Moderate
RWI
2
of 10

Follows up on

Fetal Attraction: Abortion and the Principle of Charity — Essay · May 2013

Summary

An even-handed empirical wade through whether contraception raises or lowers abortion rates. Engages the anti-contraception case (Bad Catholic's country-by-country 'simultaneous rise' correlations, the Peltzman/risk-compensation mechanism) and dismantles the sloppy version (the correlations are just '20th-century trended up'; Canada legalized abortion the same year as the pill; Portugal's inflection is the 2007 legalization), then the pro-contraception case (teen-birth drop post-1960 Pill, abortion-patient surveys, the St. Louis free-contraceptive study with its self-selection flaw). Lands on the Guttmacher synthesis: contraception + abortion often rise together because both flow from the same 'lots of sex, few kids' demographic transition (not necessarily causal); but in an already-transitioned society like the US, more (and more effective) contraception decreases abortion.

Why this score

Quality 70 · Strong. Strong, mid-upper. Fair, careful empirical work on a charged topic — the country-by-country debunking of bogus correlations is sharp, and the contraception+abortion-as-joint-symptoms-of-the-transition synthesis is clarifying. Held mid-Strong because it's bounded (one question, messy data) and the conclusion is appropriately hedged; an early (2013) piece that nonetheless models good even-handed data-wrangling.

Claude’s paradigm shift 48 · Moderate. Moderate (48). The demographic-transition synthesis largely draws on the Guttmacher explanation he quotes; the fresh angle is applying the Peltzman effect and the careful both-sides debunk. Useful reframe, not paradigm-shifting.

Real-world impact 2 · Minor. Minor (2). A careful analysis of a policy-relevant question whose reach is within-discourse; aligns with the mainstream public-health view, no policy directly moved.