Econ prof @Middlebury. Applied micro, causal inference, health, public, labor, fertility, economics of abortion. Widow. Wife. Mom of 4.
Caitlin Myers
Loading...
Placebos: We fake the iPhone’s launch in earlier years, and swap in Sprint and Verizon’s footprints. All null. E.g., Verizon-but-not-AT&T counties show no effect.
Is it really the iPhone—or were AT&T counties trending differently for other reasons? We throw a lot at the result to kill it: entropy-balanced Poisson and synthetic diff-in-differences. We come up empty. It holds.
Y'all...am I the only one who doesn't totally "get" the implicit weights in continuous-treatment TWFE? Just in case there are others out there, @causalinf.bsky.social does a great job walking us through it in our latest episode of "The Odd Couple". causalinf.substack.com/p/episode-10...
We then map AT&T coverage to changes in births. Births to young people fell far more where AT&T—and the iPhone—reached. For older women, births rose less.
Zeke is my former student @middleburycollege.bsky.social --- and my stepson. This project grew out of family dinner-table debates about iGen. We found @jonathanhaidt.bsky.social and @jeantwenge.bsky.social compelling… but how do you know the link between smartphones and falling fertility is causal?
Is it the phones?
New NBER WP this morning with Zeke Hooper: we estimate the diffusion of the iPhone explains 33–52% of the decline in US births from 2007–2011. #econsky🧵
www.nber.org/papers/w35310
Many "what's up with kids these days?" dinner table conversations turned into a co-authored NBER WP with my stepson that was covered in the New York Times this morning. Didn't have that on my life bingo card.
www.nytimes.com/2026/06/08/u...
Our experiment ends in 2011, but smartphones kept spreading—and likely kept pushing births down. There's another new working paper that takes up those later years using a different research design. It also finds large effects of technology. papers.ssrn.com/sol3/papers....
Bottom line: absent the iPhone, the 2007–2011 fertility decline would’ve been less steep.
Our natural experiment: when the iPhone launched in 2007, Apple gave AT&T the exclusive right to sell it. That monopoly lasted nearly four years, until 2011.But AT&T didn’t have coverage everywhere. Using validated SBI broadband data, we map AT&T coverage county by county.
