Who’s happiest: Denmark or Vanuatu?
When two different countries can top two different happiness surveys in the same month, you know there’s something dodgy about happy stats.
Imagine there were competing systems for awarding points to teams in a football league. According to the first system, one set of teams was in the top 10, while judging by the other system, a completely different set of teams was at the top. Observers would probably conclude that at least one and possibly both systems were flawed.
Yet no one seems to be drawing such a conclusion from the recent publication of two sets of global happiness league tables (reproduced at the bottom of this article). In the World Map of Happiness, compiled by Adrian White at the University of Leicester, the top 10 are mainly European countries with Denmark first, followed by Switzerland and then Austria (1). In the Happy Planet Index, compiled by the New Economics Foundation (NEF) in association with Friends of the Earth, the top country is Vanuatu (an island nation in the Pacific), but the list is dominated by smaller Latin American countries (2).
Wealthy countries do much better in the World Map of Happiness than in the Happy Planet Index. America is ranked 23 in the World Map of Happiness but 150 in the Happy Planet Index, while Britain is 41 in the first and 108 in the second. Healthcare and education also help countries rank highly in the World Map of Happiness. On the other hand, Colombia, a country which has suffered a 40-year civil war, somehow makes it to number two on the NEF ranking.
Despite the vast differences between the two tables, apparently measuring the same thing, the contradiction does not seem to have attracted much comment (3). The publication of both sets of league tables in July has generally been reported uncritically (4). Journalists seem to take it as given that the league tables provide meaningful information for their readers.
Before unpicking this puzzle it is important to point out there is nothing inherently wrong with measuring subjective perceptions of happiness. If individuals in one country are on average happier than in another country, or if happiness varies over time in the same country, it is worth investigating. But it is important to recognise that such perceptions should be the starting point rather than the end point of a study. All sorts of factors, including social and cultural ones, can influence what people say about their levels of happiness. The job of the researcher into such data is to try to work out why people say what they do.
The first and most astonishing fact about the two happiness league tables is that neither are measures of happiness. Although they are promoted as being rankings of happiness, a glance at the small print shows that subjective wellbeing is only a component of what they measure. Both are based on various combinations of self-reported happiness combined with a variety of other indicators. In fact, the University of Leicester Study uses data from the NEF, which in turn compiled the Happy Planet Index.
Full information on how the World Map of Happiness is calculated is not provided on the Leicester University website, although it says: ‘The projection, which is to be published in a psychology journal this September, will be presented at a conference later in the year.’ (5) In addition to self-reported happiness, it seems to include access to schooling, life expectancy and GDP per head as part of the calculation.
In contrast, the NEF gives a rough idea of how its index is calculated (6):
The HPI incorporates three separate indicators: ecological footprint, life-satisfaction and life expectancy. The statistical calculations that underlie the HPI are quite complex. However, conceptually, it is straightforward and intuitive:
|HPI = ||Life satisfaction x Life expectancy|
Out of these indicators life expectancy and life satisfaction are fairly straightforward. Life expectancy is clearly easy to quantify and it is an important objective measure of wellbeing. Life satisfaction is a subjective measure of happiness and, as already argued, is worth knowing. However, whether it is meaningful to combine the two in a single measure is open to question. One is an objective measure of wellbeing while the other is subjective. In addition, there are many other objective measures of wellbeing that could be used in addition to life expectancy.
The most problematic measure, however, is that of the ‘ecological footprint’. The NEF says: ‘The ecological footprint measures how much land area is required to sustain a given population at present levels of consumption, technological development and resource efficiency, and is expressed in global-average hectares.’(7) Even assuming for a moment that this concept is useful, it is hard to see what it has to do with happiness. It is quite conceivable that people living in countries with a large ecological footprint could be happy even if it was the case that their lifestyles were unsustainable.
As it happens, the notion of an ecological footprint is a dubious one. It is usually presented as a fixed measure but, as the definition suggests, greater technological development can completely change the measure. The amount of area needed to sustain a given population at one level of technology can be completely different from that at a higher level of technological development. Better technology, as well as greater resource efficiency, means a smaller area of land can sustain a given population. Yet the discussion of the ecological footprint typically includes meaningless statements such as ‘humans are using more resources than are available on the planet’ (8).
In the context of happiness rankings, to include the ecological footprint inevitably means that countries with high levels of consumption will be downgraded. So this assumption alone means there is an in-built bias against the developed countries in the NEF’s Happy Planet Index. Therefore, middle-income nations – where individuals have a reasonable life expectancy but consumption is still relatively low – do best. In contrast, the University of Leicester Study seems to award positive points for countries with a high GDP per head, so developed countries do well in that ranking.
This shows that world happiness rankings reveal more about the prejudices of their compilers than they do about the inhabitants of different nations. This is most clear in the NEF Happy Planet Index, which, through the use of the ecological footprint, has that bias against developed nations. The conclusions that it draws – including the need for humans to restrict their consumption – are built into the assumptions of the index. It is a circular argument.
What ‘happiness’ means in this context is the happiness of the index compilers rather than that of the world’s inhabitants. As environmentalists, the NEF’s preference is for a world in which humans restrict their consumption. The implication is that it would prefer Americans or Britons to reduce their consumption levels to that of people in Colombia or Costa Rica. Therefore, it has constructed an index which reflects the qualities it sees as desirable.
Statistics can be a valuable aid to understanding society. If they are properly used they can help point to areas that demand further investigation by researchers. But those who use statistics should never forget they need to be handled with skill and care. They would do well to remember the old maxim of computer programmers: garbage in, garbage out.
Daniel Ben-Ami is speaking at the Battle for Affluence session at the Battle of Ideas festival in London in October 2006.
|World Map of Happiness – Top 10|
(University of Leicester)
Source: University of Leicester
|Happy Planet Index – Top 10|
(New Economics Foundation)
|10||Saint Vincent and the Grenadines||61.4|
Source: Happy Planet Index
(1) University of Leicester produces the first ever World Map of Happiness, press release, 28 July 2006
(2) The full report can be downloaded here
(3) The New Scientist magazine briefly referred to the contradiction between the two rankings: Roxanne Khamsi, Wealthy nations hold the keys to happiness, New Scientist, 28 July 2006
(4) An honourable exception to the often uncritical reporting of happiness research is Will Wilkinson, a policy analyst at the Cato Institute. See http://happinesspolicy.com/
(5) University of Leicester produces the first ever World Map of Happiness, press release, 28 July 2006
(6) The Happy Planet Index – how it is calculated
(7) The Happy Planet Index – ecological footprint
(8) For a critique of the idea of an ecological footprint see ‘Treading lightly’, The Economist, 19 September 2002. This article is in turn based on a longer paper from Denmark’s Institut for Miljøvurdering (Environmental Assessment Institute): ‘Assessing the Ecological Footprint: A look at the WWF’s World Planet Report 2002’, available in pdf format here.