Tuesday, 6 January 2015

More things to ban

Suppose I told you about something that's totally preventable and is associated with an average excess death rate of 6.7%. It's worse for youths: the excess death rate for those aged 20-29, because of this totally preventable problem, is 25.4%.

And there would be one simple fix to end those excess deaths entirely.

Ladies and Gentlemen, it's time to ban birthdays. The simple policy fix: stop recording date of birth on birth certificates, and ban kids from having birthday parties so they never grow up knowing their birthday.

Here's the original study by Pablo Peña.


This study estimates average excess death rates on and around birthdays, and explores differences between birthdays falling on weekends and birthdays falling on weekdays. Using records from the U.S. Social Security Administration for 25 million people who died during the period from 1998 to 2011, average excess death rates are estimated controlling for seasonality of births and deaths. The average excess death rate on birthdays is 6.7% (p < 0.0001). No evidence is found of dips in average excess death rates in a ±10 day neighborhood around birthdays that could offset the spikes on birthdays. Significant differences are found between age groups and between weekend and weekday birthdays. Younger people have greater average excess death rates on birthdays, reaching up to 25.4% (p < 0.0001) for ages 20–29. Younger people also show the largest differences between average excess death rates on weekend birthdays and weekday birthdays, reaching up to 64.5 percentage points (p = 0.0063) for ages 1–9. Over the 13-year period analyzed, the estimated excess deaths on birthdays are 4590.
You can also check the Washington Post's report.

Now hear me out. We might think it a bit silly to ban birthdays, but the excess death finding seems pretty robust. And we know that anything that saves even one life is worthwhile. I'm sure that the fine folks in the New Zealand Government will use the price elasticity of demand for kid birthday parties to argue that the forgone benefits would add up to not much more than maybe $400 a day over the whole country. And nobody will bother fisking the number or even giving it a basic plausibility check before the National Party goes and implements the ban to avoid being outflanked on the "showing you care" margin by the newly competent Labour Party.

Now for a more worrying econometric point: a pile of regression discontinuity design regressions use birthday cutoffs where policy legalises something at an age as the source of the discontinuity, attributing all changes then to the policy and nothing to birthday effects. So RDD methods find increased mortality when kids hit the legal alcohol purchase age. But that's confounded with this birthday effect, as Stillman and Boes demonstrated. Be careful using RDDs around birthdays....


  1. For the life of me I can't see any links to Scott's blog post. It's here: http://slatestarcodex.com/2013/06/22/social-psychology-is-a-flamethrower/

  2. Akh, how did I manage to do that? Will fix.

  3. It's also a nice illustration that a quoted 6.7% excess rate translates to 4590 excess deaths out of 25 million -- ie, only about 0.2%

  4. "I can’t find the link for this, but negatively phrased information can sometimes reinforce the positive version of that information"

    Amongst others Nathan Mikaere-Wallis (http://www.brainwave.org.nz/about-us/our-people/presenters/nathan-mikaere-wallis/) explains this on one of his DVD's in the context of child development. Apparently it has to do with how the brain processes language by associating visual images with words. Words like "don't" don't really register apparently.

    Telling a child "don't stand on that chair" gets processed as "stand chair". Pretty much explains why so many forms of regulation are so ineffective.

  5. I really lol'd at this one. For reals.