A few weeks ago I read a popular article online in the Washington Post about the international ranking of alcohol consumption habits. Every Irish person is instantly interested in this subject, as although we in Ireland are not impressed by the stereotype of ourselves as drunken messes, we as a small country are proud of any opportunity to be the best at something, regardless of what that something is. A few years ago, it was revealed that the Irish drank more cups of tea per person per day than any other nationality, and this was reported as the leading story on the national evening news that day. Similarly, in my current home of Austria, the Austrians are very proud of topping the international beer drinking charts a few years ago. Sadly, both Austria and Ireland have been overtaken, and indeed put to shame, by the Czech Republic over the past decade, and I wasn’t surprised to see the Czechs top the rankings the Washington Post’s study from last month. Figure 1 below reports the top 30 country rankings I pulled from the World Health Organisation’s (WHO) Global Information System on Alcohol and Health (GISAH), which is the source of the Washington Posts data. The graph reports average ml of alcohol consumed per person per year.
Of course as a major sufferer of small-country-syndrome (which is exacerbated by also being an ex-pat), I was irrationally annoyed at Ireland only coming in third place, outdone by two countries that didn’t even exist when I was growing up. So naturally, as a social scientist, I started thinking of ways to analyse the data better (in a way that would hopefully further Irelands position in the ranking). After thinking about it for a while, I settled on creating an index of Functional Boozing. For while Estonia and the Czech Republic do engage in a lot of drinking, they pay for it dearly, as both countries are not really up to much in the world economy, or anything else. Ireland, meanwhile had an economic boom, lots of jobs and plenty of tourism and international prestige. This didn’t end well, but at least it was something. During the boom period of the Celtic Tiger in Ireland, we were consistently ranked in the top 5 beer drinking nations in the world, which is something not many countries can lay claim to. I therefore settled on the idea that it was not just about how much a country drinks; it is about the productivity, quality, and indeed happiness of the country itself that should also be a factor in this ranking.
How to do this exactly was problematic, and time consuming (particularly before I decided to narrow the sample down to just the top 30 countries ranked in the WHO’s alcohol consumption database). Naturally I first turned to GDP per capita, as this data is very easy to acquire, and it is generally used as a proxy for happiness and success by very lazy researchers. I tried to create a model based on GDP and alcohol consumption, but this always overstated the Western European countries, and therefore was just a measure of the richest drinking countries. I added in health measures such as international ranking of national health systems, life expectancy, even penis size (which is surprisingly negatively correlated with GDP), but the relative weights of all of these things was difficult to justify. Finally, like a good researcher, I decided to be even lazier and fall back on the work of others. In a given year, a diverse range of research institutes produce a dazzling array of qualitative indexes aimed at measuring the (something) of a country, and most of these reports end up being a vain attempt to show that what they have done is empirically justified. This solved both of my problems, as not only were all my variables already present in a given index, there is also a wealth of literature at hand to tell one of you to read if you disagree with my methodology. In the end, I picked three of these indexes to use in my analysis, which are described in table 1 below.
Table 1: Indexes used
|Happy Planet Index||New Economics Foundation||Positivity towards achieving life goals, life expectancy, and effect on the environment|
|Quality of Life Index||Economist Intelligence Unit||A number of quantified economic, political and social variables|
|World Happiness Report||UN Sustainable Development Solutions Network||Lots of experts talking about what “Happiness” might be. A score from 1-10 emerges.|
All exist and are entirably googleable. There is some overlap of course, as “life expectancy” does occur in two of the indexes, and the top 5 of the two ‘happiness’ indexes are almost identical (in the countries, not in the ranking), yet I would argue that between the three indexes, the quality of life in any country would be proxied to some minimum degree. In some cases, observations were missing for some countries in the World Happiness Report. When this occured, I computed a value for that country using an average of its neighbouring countries, as well as the overall average of the index. A further issue was that while the World Happiness Report and the Quality of Life Index results are delivered as a score from 1-10, the Happy Planet Index is more open, and there is no upper bound to the scores. The highest score in the index is just over 50, and since the index is less relevant to the overall point of the exercise (being environmentally friendly is not something I care about when it comes to functional alcoholism), I decided to divide the HPI by 10, which means that it has less weight in the final scores of the Functional Boozing Index (FBI). Given below is the formula for computing the FBI:
Where QoL is the Quality of Life Index, hpi is the Happy Planet Index, and happy is the World Happiness Report. Booze refers to the amount of alcohol consumed per year (in ml) from the World Health Organisation’s database (shown previously in Figure 1). The FBI is therefore the ratio of the sum of the three indexes (with hpi divided by 10) to the alcohol consumed per person per year. Listed below in table 2 are the data, along with GDP per capita for each country, and the results.
As you can see, the table has changed dramatically. The first thing to note is that the top two highest drinking countries from the original WHO report are now right at the bottom of the table. As I originally predicted, these countries may drink a lot, but they really only drink out of despair: there is not a whole lot else going on (I am sorry, Czechs, I actually like your country, I am trying to be objective here). Unfortunately, my other original intent was for Ireland to usurp those two pretenders, and sneak into the top spot. This didn’t occur, and Ireland now languishes pathetically in mid-table, possibly as a result of the happiness-vacuum that exists in the country since the economic depression burst its bubble. What must be admired however, is the place of the Netherlands in the table. In the WHO index, the Dutch were at the foot of the table, and now they claim pride of place as the most functionally alcoholic nation on earth. They can handle all that Heineken and sugary schnapps, as well as maintaining a decent level of contentedness.
I imply no causal link between consumption of alcohol and ability to perform economically, as well as in the pursuit of happiness: it is merely a juggling act. If a country can drink a lot (and all these countries do, they are the top 30 out of 200+ nations), as well as reporting high levels of happiness and various other quality of life measures, this is something that could be seen as positive. It isn’t all about money either, as low GDP per capita countries such as Poland and Slovakia have risen in the rankings when “my” happiness variables are taken into account. What I have hoped to achieve with this blog entry is foremostly, to investigate the interactions between alcohol consumption and happiness on a national (and perhaps even personal!) level. The construction of an index such as this is never perfect, but the next time you read about a major new qualitative index, please consider that it is constructed in an almost identical way to the analysis I have just demonstrated. I was not entirely serious about the construction of this FBI, but I put enough work into it (a few hours constructing the graphs and searching for the other indexes) to consider it a worthy index until someone constructs something similar and argues differently. If you have ideas about how to do that, please let me know!