Thursday 13 December 2012

What if?

A couple of weeks ago, I gave a talk as part of the University's "What If" lecture series, asking "What if alcohol were not as socially costly as everyone says?" It's now up on the University's site and is embedded below.

There is one question at the end that I handled poorly. An attendee wondered whether the change in the reported binge drinking rates could be due to definitional changes. Because the numbers I was using were using the older and consistent standard, I said that wasn't it. But the reported increases were coming from some more recent reports. And I just don't know whether they were using a consistent standard for binge drinking.


  1. Excellent primer on the field Eric.

    Things I took away:

    1. voodoo magic conclusions, such as BERL's, from poor methodology. Seems like people (maybe Sir Geoff included) read the papers and went wild fire in the media.

    2. J curves... (who knew?!?!)

    3. Think about the margins... I'm in the middle so why should the edge's affect me as dramatically as the extremities when it comes to policy?

    4. Margins may be hard to target with policy. Broad brush policy may do more harm than good in the middle than affecting the problematic margins. Weight margins if you can - at least try to target margins.

    I didn't go back and compare/review but I am guessing one of the other key points is that the diagnostic definition of harm at the start didn't quite meet the J curve expectancy of harm. That being correct, some of those studies should be disregarded given the cold light of day of the J curve analysis.

    So the real question is, how do we get the tee-totallers causing us harm get on the alcohol band wagon to bring down externalities? :-)

    On a personal note, when you got animated it really hits home

    (the flippant conclusions work... they really imprint... e.g. group 1 carcinogens! As a Pharma guy it cracked me up)

    Nice work.

  2. All sounds about right.

    Harm from an economic perspective will differ also from harm from a medical perspective. When they used a four standard drink per day on average threshold for medical harm, that isn't far from the J-curve cut point. But that's not the same as the threshold for economic harm UNLESS the only thing that people care about is their health. If you choose to have 5 instead of 4, knowing that you will incur a bit of health risk, but weighing the consumption benefits as being worth it, there is no economic harm.

    The thing that hit me in the video is that when I'm looking at people in the front row of a tiered theatre, it looks like I'm looking down at my notes. I need to pitch to the back of the room.