The iPhone Birth Rate Theory, and Why It Broke
In 2007, something happened to the birth rate in a dozen countries at once. Not a slow bend. A break. And when you line those breaks up against the year each country got smartphones, they match almost too well.
That chart went viral in 2026. Then a pair of economists tried to turn it into evidence. This is what they found, what fell apart under scrutiny, and the one piece nobody is arguing about.
The chart that started it
The Financial Times published an analysis comparing national fertility rates against smartphone adoption. Paul Graham amplified it. Within days it was everywhere, mostly because of one visual property: the lines don't drift apart gradually. They hinge.
Chart above: Financial Times, graphic by John Burn-Murdoch, as shared by Paul Graham on X.
Pick a country, find the year smartphones arrived, and the fertility line changes direction within a year or two of that date. Do it for a rich country and a poor one and you get the same shape at different times.
| Country | Smartphone arrival | Fertility inflection |
|---|---|---|
| US, UK, Australia | ~2007 | ~2007 |
| France, Poland | ~2009 | ~2009 |
| Mexico, Indonesia | ~2012 | ~2012 |
| Ghana, Nigeria, Senegal | 2013–2015 | 2013–2015 |
The staggering is the interesting part. A global recession hits everyone in the same year. This didn't. It moved country by country, in the order the phones arrived.
The 2008 objection, and why it isn't enough
The obvious counter is the financial crisis. People stop having children when money gets tight, and 2008 is right there.
Except the crisis doesn't explain Ghana in 2014 or Indonesia in 2012. It doesn't explain why countries that skipped the worst of 2008 still show the break. And within the UK and the US, the regions that got 4G first are the regions where births fell fastest, a pattern that has nothing to do with the housing market.
Then economists tried to prove it
In June 2026, Caitlin Myers and Ezekiel Hooper published a working paper with a title that does not hedge: Is the iPhone Birth Control?
Their trick was a real one. From 2007 to 2011, the iPhone ran exclusively on AT&T in the US. If you lived where AT&T had coverage, you could own one. If you didn't, you couldn't, regardless of how much you wanted one. That is close to a natural experiment, and natural experiments are how economists get at cause instead of correlation.
The results were not subtle:
- Births down 4.5–8.0% among 15–19 year olds
- Births down 3.2–6.6% among 20–24 year olds
- Smaller declines in older groups
- Roughly a third to a half of the entire US fertility decline attributed to the iPhone
The proposed mechanism is not what most people assume. It isn't that phones make you infertile, or even that they make you less interested. It's simpler and stranger: teenagers stopped being in the same room as each other.
What fell apart
Demographers went at the paper hard, and some of the hits landed.
The sharpest one is a placebo test. Run the same model on Verizon and Sprint, carriers that had no iPhone at all during those years, and you still get a significant effect. If a phone nobody could buy produces the same result as the phone in question, the model is probably measuring something else. Most likely the thing that separates AT&T counties from the rest: they were richer and more urban, and urban fertility behaves differently for reasons that predate the iPhone by decades.
There's also a timing problem. By 2011, the end of the study window, only around 12–13% of Americans aged 20–24 owned an iPhone. Instagram launched in late 2010. TikTok didn't exist. The "everyone is scrolling instead of socializing" story needs an era that hadn't started yet.
None of this makes the correlation disappear. It means the causal claim is unproven, not that it's wrong.
The part nobody disputes
Strip out the fertility argument entirely and one finding survives every version of this debate: young people spend dramatically less time physically together than they used to.
That trend is measured directly, in time-use surveys, without needing any theory about phones. It's one of the most consistent behavioural shifts of the last two decades.
And it doesn't require a grand explanation. An evening has a fixed number of hours. An app designed by a team of engineers to hold your attention is competing for those hours against a friend who has to text you first, agree on a place, and travel there. One of those has meaningfully less friction than the other.
That is the whole mechanism. Not addiction, not damage, not a moral failing. Just friction, applied in one direction for fifteen years.
What you can do with this
None of the above proves that your phone is costing you anything specific. Population-level statistics say nothing about any individual, and anyone who tells you otherwise is selling something.
But the friction argument works in reverse too, and that part you can test yourself in a week. Make the app slightly harder to open than the thing you'd rather be doing, and watch which one wins.
That's the entire idea behind Scrolly: a physical tag that blocks the apps you lose time to, and needs a deliberate tap to unblock them. Everything you rely on stays open. The pause is the point, the second between reaching for the phone and deciding whether you meant to.
Fifteen years of friction pointed one way. It's worth finding out what happens when you point a little of it back.
