Skip to content

Are All Temperance Policies Doomed to Fail?

Jay Brooks always has his eyes on the modern day prohibitionists. And thank goodness. The real prohibitionists of the late-19th/early-20th centuries were powerfully organized and they are the textbook example of a transnational activist movement ultimately successful at passing some form of national alcohol prohibition legislation not just in the United States but around the world. Back then, like many of the capital-P Progressive groups of the era their central arguments were framed around ideas of religious (mostly Protestant) and personal morality.

Modern prohibitionists—consistent with contemporary trends in other fields—tend to frame their arguments in terms of statistics, “markets,” science, and public health. This often makes it more difficult to criticize them because arguments based on science and statistics arrive with an often false sense of certainty and finality. Only people sensitive to the ways that science can mask subjectivity are capable of pointing out the truth behind the “truth.” Jay Brooks is one of those.

On the other hand, there is a tendency to respond to the artificial black-and-whiteness of a scientific (or “scientific”) piece with an equally artificial white-and-blackness. In this post from over the weekend, Brooks points out that a recent study concluded that a ban on quantity purchases and “buy one get on free” marketing tactics did not reduce overall consumption in Scotland. Brooks comments:

The main findings were that they didn’t work — shock, surprise.


You’d think at this point that policymakers would realize that trying to stop people from buying as much alcohol at one time as they want would do nothing except inconvenience adult purchasers of products they’re legally entitled to buy and consume.


Didn’t thirteen years of Prohibition make that abundantly clear? Yet all this type of regulation accomplishes is to punish the law-abiding, responsible adults who want to enjoy a legal adult beverage.

Whenever anybody starts to say that the results of a study are “obvious,” I get suspicious. Most people, myself included, tend to put more faith in studies that prove what we already believe and to overly scrutinize (or most often, simply ignore and forget) studies that prove the opposite. When studies confirm our presumptions, we are more prone to generalize their findings, to ignore fallacious reasoning,  to overlook small sample sizes, and to under scrutinize base rates and effect sizes because if it was “obviously” true before and is confirmed in Scotland, there is no harm at all in assuming that it naturally must hold true everywhere and at all times.

There are two conflicting assumptions at work when talking about “average alcohol consumption.” One is that at any given moment, the desired amount of human consumption is a static determinable number (normally captured by some proxy, liquor sales, for example). The other is that humans are rational actors that change their behavior to fit social, environmental, and legal (among other) contexts. The first assumption allows us to generate maps like this one from The Economist that determine the average consumption of entire nations.


More or less, all thing being equal (or as we like to say in the biz “ceteris paribus”) human consumption just doesn’t change all that much. If I prefer to drink 4 nights a week and consume between 3 and 4 drinks each time I drink, then that’s my preference. And if Colorado passed a law tomorrow that made drinking more than 2 drinks at any given time more inconvenient (by passing a law that forbade pub owners from serving more than 2 drinks in a 24-hour span to the same person, for example) I would still have those preferences. So I would look for ways to achieve my preferred level of drunkenness (by attending two bars near each other so that I could have two drinks at one and then go across the street for my next two.)

But in the real world, things are not all ceterisy paribusy. Look back up at that average consumption map. The most striking thing on the map is that it is not all one color. “Average consumption” differs by region. Why do Americans drink more than Saudis? Why do Russians drink more than…everybody? Let’s drill down to just one country. This site lists the prevalence of people who self-identify as casual drinkers. You will see it ranges from a high of 67.8% in Wisconsin to a low of 28.7% in Utah (average: 53.6%, roughly what they’re drinking in Virginia, 54%). The map below shows the amount of “binge drinkers” by state. Here we can see that from state to state alcohol consumption rates vary within a country. Why do Illini drink more than Hoosiers? Why do Utah…er…ites? drink less than everybody?


And here is a chart of median consumption numbers in Britain going back to 1992.


Although the numbers don’t change much, they do change. And that’s the point. There is nothing predetermined about the level at which humans want to, can, or will drink. Some drink nothing at all and some people drink so much they die from it. Anything in between is subject to a variety of pressures. Some of these pressures are very personal, but for most of us, social factors will play a major role. Some of these social factors are legal in nature. Look at that map of binge drinking averages and compare it to this Wikipedia map of where “dry” and “moist” counties are located. It is interesting that a prevalence of “dry” counties seems to correlate with less binge drinking, but a prevalence of “moist” counties seems to correlate with more binge drinking.


I wouldn’t necessarily jump to the conclusion that living in a dry county makes people not binge drink, but that the two are unrelated is not “abundantly clear.” The picture is extremely muddy. What is apparent is that rates of alcohol consumption are variable and something must account for those differences. I would suggest that policy is at least one factor.

My point is not that the Scotland study is erroneous. Only that it’s finding were not ex ante obvious and that policymakers are under no pressure to simply admit that temperance measures of any sort are universally condemned to failure (even if outright prohibition certainly is). These maps make it clear (at least to me) that policies do (or at least can) influence the choices that people make. We may not be able to measure these differences over just a few short months—it takes a while to develop new habits. And the Scotland policy was probably designed poorly. In general it’s more effective to just tax “undesirable” activities than to attempt to submerge the tax as in the Scotland case. However, I don’t think it was at all obvious that it was doomed to fail.

Share and Enjoy:
  • Print
  • Digg
  • StumbleUpon
  • Facebook
  • Yahoo! Buzz
  • Twitter
  • Google Bookmarks
  • Add to favorites
  • Google Buzz
  • LinkedIn
  • Tumblr

Posted in The Drinking Class.