In previous posts, I have referred to cross-country data about reported levels of happiness, and suggested that these surveys support the notion that increases in GDP per capita, beyond a certain point, don’t make people happier. I have further suggested that this is encouraging given that if we are to deal with the problem of climate change in an ethical manner, this will require a drop in income for rich-nation inhabitants.
But in recent paper, Helen Johns and Paul Ormerod critique happiness studies and dispute that policy conclusions can be drawn from the results of such studies. I find their critique and conclusions persuasive, for the most part. It is important to note that they are not critiquing or supporting the policy positions that have been supported by happiness data, but rather the meaningfulness and use of the data. In the end, perhaps this serves as another reminder of the limitations of utilitarianism as a guide to policy-making, and of the inevitability of making ethical and value choices when making public policy decisions.
Johns and Ormerod’s argument is that:
There are at least two alternative interpretations to the mainstream view that happiness has remained flat over decades because economic growth does not make us happier. First, we could conclude from this flat trend that attempting to improve the human lot through any policy – not just through pursuing economic growth – is entirely futile. Second, and alternatively, that happiness data over time shows little movement because it is an extremely insensitive measure of welfare.
We argue that the evidence points to the latter. This can be demonstrated both from empirical reasoning and by examining the mathematical properties of the measure itself.
Above all, we argue that average happiness time series are, by construction, incapable of conveying useful information on the level of overall social well-being, and their use should therefore be rejected by policy-makers and social scientists.(pp.139-140)
In relation to “empirical reasoning” Johns and Ormerod note that:
there is no correlation in time series data between reported happiness levels and a whole series of factors which might reasonably be thought to affect well-being: income, public spending, longevity, gender equality, income inequality – even the incidence of depression in a population. (p.141)
With regard to the properties of the surveys as they are constructed (e.g. 1 not happy, 2 fairly happy…), Johns and Ormerod suggest that:
people have to undergo large discrete change in their happiness in order for this to be registered by the indicator; and once they have reached the top category they officially can’t experience any further increase in their happiness. As a consequence, noticeable changes in average happiness can only come about through substantial numbers of people moving category. (p.141)
And in regard to using the surveys as long-term, time-series, data, they note:
by construction, the happiness data can exhibit no indefinite trend. As individuals answer a survey in which they are asked to state their own level of happiness on an n-point scale, the data is therefore bounded between one and n. Over any particular short period of time, an apparent trend either up or down might exist, but by definition it cannot persist. In contrast, at least as it is presently defined, real GNP can exhibit no upper bound. (p.142)
Note: Both GDP, and GNP (sometimes known as GNI) are measures of the market value of production of an economy.
Finally, Johns and Ormerod (pp.142-145) discuss an analysis by Johns which shows that:
the variations which we observe in measured happiness are completely consistent with the view that they are simply fluctuations based on sampling error. (p.142)
Well I’m convinced.
I still believe that increases in GDP per capita, beyond a certain point, don’t make people happier, but now I shall have to make other arguments and find other evidence to make my case.
This paper has had me musing on the first possible interpretation that Johns and Ormerod floated: that attempting to improve the human lot through any policy is futile. It’s a heretical thought for most green activists! Certainly, if you look at the enduring features of human life and existence, and at the enduring wisdom contained in the world’s great spiritual literature and traditions, it seems plausible. On the other hand, I am damn glad to have electricity and running hot and cold water! Perhaps then, part of the problem here is that well-being is not directly equivalent to “happiness” and that “happiness” and “well-being” are not unitary concepts. I also suspect that while we may (arguably) be a bit vague about how to improve happiness and well-being we have a lot of effective knowledge about how to make people and societies miserable. Perhaps many of the policies people argue for in terms of improving well-being might more accurately (if less glamorously) be argued for in terms of avoiding misery? Finally, I suspect a good portion of happiness and well-being – at societal level as well as for individuals – is tied up with such intangibles as hope and vision – as they said in the musical:
Happy talk, keep talkin’ happy talk,
Talk about things you’d like to do.
You got to have a dream,
If you don’t have a dream,
How you gonna have a dream come true?
From “South Pacific” (Hammerstein / Rodgers)